Artificial Intelligence in Sensors

Dear Colleagues,

This topic comprises several interdisciplinary research areas that cover the main aspects of sensor sciences.

There has been an increase in both the capabilities and challenges related within numerous application fields, e.g., Robotics, Industry 4.0, Automotive, Smart Cities, Medicine, Diagnosis, Food, Telecommunication, Environmental and Civil Applications, Health, and Security.

The associated applications constantly require novel sensors to improve their capabilities and challenges. Thus, Sensor Sciences represents a paradigm characterized by the integration of modern nanotechnologies and nanomaterials into manufacturing and industrial practice to develop tools for several application fields. The primary underlying goal of Sensor Sciences is to facilitate the closer interconnection and control of complex systems, machines, devices, and people to increase the support provided to humans in several application fields.

Sensor Sciences comprises a set of significant research fields, including:

  • Chemical Sensors;
  • Biosensors;
  • Physical Sensors;
  • Optical Sensors;
  • Microfluidics;
  • Sensor Networks;
  • Electronics and Mechanicals;
  • Mechatronics;
  • Internet of Things platforms and their applications;
  • Materials and Nanomaterials;
  • Data Security;
  • Artificial Intelligence;
  • Robotics;
  • UAV; UGV;
  • Remote Sensing;
  • Measurement Science and Technology;
  • Cognitive Computing Platforms and Applications, including technologies related to Artificial Intelligence, Machine Learning, as well as Big Data Processing and Analytics;
  • Advanced Interactive Technologies, including Augmented/Virtual Reality;
  • Advanced Data Visualization Techniques;
  • Instrumentation Science and Technology;
  • Nanotechnology;
  • Organic Electronics, Biophotonics, and Smart Materials;
  • Optoelectronics, Photonics, and Optical fibers;
  • MEMS, Microwaves, and Acoustic waves;
  • Physics and Biophysics;
  • Interdisciplinary Sciences.

This topic aims to collect the results of research in these fields and others. Therefore, submitting papers within those areas connected to sensors is strongly encouraged.

Prof. Dr. YangQuan Chen
Prof. Dr. Nunzio Cennamo
Prof. Dr. Subhas Mukhopadhyay
Prof. Dr. Simone Morais
Topic Editors

Deadline for abstract submissions: 30 October 2022.
Deadline for manuscript submissions: 31 December 2022.

Topic Board

Prof. Dr. YangQuan Chen
grade E-Mail Website
Topic Editor-in-Chief
Department of Mechanical Engineering (ME), University of California, Merced, CA 95343, USA
Interests: drone-based remote sensing; 3D photogrammetry; optimal sensing in cyber-physical systems; distributed collaborative sensing in distributed parameter systems; mobile sensor networks; virtual metrology
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Subhas Mukhopadhyay
E-Mail Website
Topic Associate Editor-in-Chief
School of Engineering, Macquarie University, Sydney, NSW 2109, Australia
Interests: sensors and sensing technology; instrumentation; wireless sensor networks; Internet of Things; mechatronics and robotics
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Nunzio Cennamo
E-Mail Website
Topic Associate Editor-in-Chief
Prof. Dr. M. Jamal Deen
E-Mail Website
Topic Editor
Department of Electrical and Computer Engineering, McMaster University, Hamilton, ON, Canada
Interests: biomedical electronics; biomedical technologies; CMOS image sensors; device modeling; smart sensors
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Junseop Lee
E-Mail Website
Topic Editor
Department of Materials Science and Engineering, Gachon University, 1342 Seongnam-Daero, Sujeong-Gu, Seongnam-Si, Gyeonggi-Do 13120, Korea
Interests: FET biosensors; image sensors; fluorescent sensors; wireless smart sensors; RFID wireless sensing networks; functional polymers; conducting polymer nanomaterials; inorganic nanomaterials; composite nanomaterials; electronic signal transport in nanomaterials
Special Issues, Collections and Topics in MDPI journals

Keywords

  • chemical sensors
  • biosensors
  • remote sensing
  • physical sensors
  • optical sensors
  • microfluidics
  • sensor networks
  • electronics and mechanicals
  • mechatronics
  • internet of things platforms and their applications
  • materials and nanomaterials
  • data security
  • measurement science and technology
  • cognitive computing platforms and applications, including technologies related to artificial intelligence, machine learning, as well as big data processing and analytics
  • advanced interactive technologies, including augmented/virtual reality
  • advanced data visualization techniques
  • instrumentation science and technology
  • nanotechnology
  • organic electronics, biophotonics, and smart materials
  • optoelectronics, photonics, and optical fibers
  • MEMS, microwaves, and acoustic waves
  • physics and biophysics
  • interdisciplinary sciences
  • Artificial Intelligence
  • sensor networks
  • WSN
  • robotics
  • machine vision
  • deep learning
  • machine learning
  • computer vision
  • image processing
  • smart sensing
  • smart sensor
  • intelligent sensor
  • unmanned aerial vehicle
  • UAV
  • unmanned ground vehicle
  • UGV

Relevant Journals List

Journal Name Impact Factor CiteScore Launched Year First Decision (median) APC
Sensors
sensors
3.576 5.8 2001 15.06 Days 2200 CHF Submit
Remote Sensing
remotesensing
4.848 6.6 2009 16.06 Days 2400 CHF Submit
Applied Sciences
applsci
2.679 3.0 2011 13.8 Days 2000 CHF Submit
Electronics
electronics
2.397 2.7 2012 15.03 Days 1800 CHF Submit
Drones
drones
- 5.4 2017 14.45 Days 1400 CHF Submit

Published Papers (35 papers)

Order results
Result details
Select all
Export citation of selected articles as:
Article
A Hybrid Vision Processing Unit with a Pipelined Workflow for Convolutional Neural Network Accelerating and Image Signal Processing
by and
Electronics 2021, 10(23), 2989; https://doi.org/10.3390/electronics10232989 - 01 Dec 2021
Abstract
Vision processing chips have been widely used in image processing and recognition tasks. They are conventionally designed based on the image signal processing (ISP) units directly connected with the sensors. In recent years, convolutional neural networks (CNNs) have become the dominant tools for [...] Read more.
Vision processing chips have been widely used in image processing and recognition tasks. They are conventionally designed based on the image signal processing (ISP) units directly connected with the sensors. In recent years, convolutional neural networks (CNNs) have become the dominant tools for many state-of-the-art vision processing tasks. However, CNNs cannot be processed by a conventional vision processing unit (VPU) with a high speed. On the other side, the CNN processing units cannot process the RAW images from the sensors directly and an ISP unit is required. This makes a vision system inefficient with a lot of data transmission and redundant hardware resources. Additionally, many CNN processing units suffer from a low flexibility for various CNN operations. To solve this problem, this paper proposed an efficient vision processing unit based on a hybrid processing elements array for both CNN accelerating and ISP. Resources are highly shared in this VPU, and a pipelined workflow is introduced to accelerate the vision tasks. We implement the proposed VPU on the Field-Programmable Gate Array (FPGA) platform and various vision tasks are tested on it. The results show that this VPU achieves a high efficiency for both CNN processing and ISP and shows a significant reduction in energy consumption for vision tasks consisting of CNNs and ISP. For various CNN tasks, it maintains an average multiply accumulator utilization of over 94% and achieves a performance of 163.2 GOPS with a frequency of 200 MHz. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
The Whale Optimization Algorithm Approach for Deep Neural Networks
Sensors 2021, 21(23), 8003; https://doi.org/10.3390/s21238003 - 30 Nov 2021
Abstract
One of the biggest challenge in the field of deep learning is the parameter selection and optimization process. In recent years different algorithms have been proposed including bio-inspired solutions to solve this problem, however, there are many challenges including local minima, saddle points, [...] Read more.
One of the biggest challenge in the field of deep learning is the parameter selection and optimization process. In recent years different algorithms have been proposed including bio-inspired solutions to solve this problem, however, there are many challenges including local minima, saddle points, and vanishing gradients. In this paper, we introduce the Whale Optimisation Algorithm (WOA) based on the swarm foraging behavior of humpback whales to optimise neural network hyperparameters. We wish to stress that to the best of our knowledge this is the first attempt that uses Whale Optimisation Algorithm for the optimisation task of hyperparameters. After a detailed description of the WOA algorithm we formulate and explain the application in deep learning, present the implementation, and compare the proposed algorithm with other well-known algorithms including widely used Grid and Random Search methods. Additionally, we have implemented a third dimension feature analysis to the original WOA algorithm to utilize 3D search space (3D-WOA). Simulations show that the proposed algorithm can be successfully used for hyperparameters optimization, achieving accuracy of 89.85% and 80.60% for Fashion MNIST and Reuters datasets, respectively. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Recognition of Noisy Radar Emitter Signals Using a One-Dimensional Deep Residual Shrinkage Network
Sensors 2021, 21(23), 7973; https://doi.org/10.3390/s21237973 - 29 Nov 2021
Abstract
Signal features can be obscured in noisy environments, resulting in low accuracy of radar emitter signal recognition based on traditional methods. To improve the ability of learning features from noisy signals, a new radar emitter signal recognition method based on one-dimensional (1D) deep [...] Read more.
Signal features can be obscured in noisy environments, resulting in low accuracy of radar emitter signal recognition based on traditional methods. To improve the ability of learning features from noisy signals, a new radar emitter signal recognition method based on one-dimensional (1D) deep residual shrinkage network (DRSN) is proposed, which offers the following advantages: (i) Unimportant features are eliminated using the soft thresholding function, and the thresholds are automatically set based on the attention mechanism; (ii) without any professional knowledge of signal processing or dimension conversion of data, the 1D DRSN can automatically learn the features characterizing the signal directly from the 1D data and achieve a high recognition rate for noisy signals. The effectiveness of the 1D DRSN was experimentally verified under different types of noise. In addition, comparison with other deep learning methods revealed the superior performance of the DRSN. Last, the mechanism of eliminating redundant features using the soft thresholding function was analyzed. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Estimation of Knee Joint Extension Force Using Mechanomyography Based on IGWO-SVR Algorithm
Electronics 2021, 10(23), 2972; https://doi.org/10.3390/electronics10232972 - 29 Nov 2021
Abstract
Muscle force is an important physiological parameter of the human body. Accurate estimation of the muscle force can improve the stability and flexibility of lower limb joint auxiliary equipment. Nevertheless, the existing force estimation methods can neither satisfy the accuracy requirement nor ensure [...] Read more.
Muscle force is an important physiological parameter of the human body. Accurate estimation of the muscle force can improve the stability and flexibility of lower limb joint auxiliary equipment. Nevertheless, the existing force estimation methods can neither satisfy the accuracy requirement nor ensure the validity of estimation results. It is a very challenging task that needs to be solved. Among many optimization algorithms, gray wolf optimization (GWO) is widely used to find the optimal parameters of the regression model because of its superior optimization ability. Due to the traditional GWO being prone to fall into local optimum, a new nonlinear convergence factor and a new position update strategy are employed to balance local and global search capability. In this paper, an improved gray wolf optimization (IGWO) algorithm to optimize the support vector regression (SVR) is developed to estimate knee joint extension force accurately and timely. Firstly, mechanomyography (MMG) of the lower limb is measured by acceleration sensors during leg isometric muscle contractions extension training. Secondly, root mean square (RMS), mean absolute value (MAV), zero crossing (ZC), mean power frequency (MPF), and sample entropy (SE) of the MMG are extracted to construct feature sets as candidate data sets for regression analysis. Lastly, the features are fed into IGWO-SVR for further training. Experiments demonstrate that the IGWO-SVR provides the best performance indexes in the estimation of knee joint extension force in terms of RMSE, MAPE, and R compared with the other state-of-art models. These results are expected to become the most effective as guidance for rehabilitation training, muscle disease diagnosis, and health evaluation. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
One Spatio-Temporal Sharpening Attention Mechanism for Light-Weight YOLO Models Based on Sharpening Spatial Attention
Sensors 2021, 21(23), 7949; https://doi.org/10.3390/s21237949 - 28 Nov 2021
Abstract
Attention mechanisms have demonstrated great potential in improving the performance of deep convolutional neural networks (CNNs). However, many existing methods dedicate to developing channel or spatial attention modules for CNNs with lots of parameters, and complex attention modules inevitably affect the performance of [...] Read more.
Attention mechanisms have demonstrated great potential in improving the performance of deep convolutional neural networks (CNNs). However, many existing methods dedicate to developing channel or spatial attention modules for CNNs with lots of parameters, and complex attention modules inevitably affect the performance of CNNs. During our experiments of embedding Convolutional Block Attention Module (CBAM) in light-weight model YOLOv5s, CBAM does influence the speed and increase model complexity while reduce the average precision, but Squeeze-and-Excitation (SE) has a positive impact in the model as part of CBAM. To replace the spatial attention module in CBAM and offer a suitable scheme of channel and spatial attention modules, this paper proposes one Spatio-temporal Sharpening Attention Mechanism (SSAM), which sequentially infers intermediate maps along channel attention module and Sharpening Spatial Attention (SSA) module. By introducing sharpening filter in spatial attention module, we propose SSA module with low complexity. We try to find a scheme to combine our SSA module with SE module or Efficient Channel Attention (ECA) module and show best improvement in models such as YOLOv5s and YOLOv3-tiny. Therefore, we perform various replacement experiments and offer one best scheme that is to embed channel attention modules in backbone and neck of the model and integrate SSAM into YOLO head. We verify the positive effect of our SSAM on two general object detection datasets VOC2012 and MS COCO2017. One for obtaining a suitable scheme and the other for proving the versatility of our method in complex scenes. Experimental results on the two datasets show obvious promotion in terms of average precision and detection performance, which demonstrates the usefulness of our SSAM in light-weight YOLO models. Furthermore, visualization results also show the advantage of enhancing positioning ability with our SSAM. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Spherically Stratified Point Projection: Feature Image Generation for Object Classification Using 3D LiDAR Data
Sensors 2021, 21(23), 7860; https://doi.org/10.3390/s21237860 - 25 Nov 2021
Abstract
Three-dimensional point clouds have been utilized and studied for the classification of objects at the environmental level. While most existing studies, such as those in the field of computer vision, have detected object type from the perspective of sensors, this study developed a [...] Read more.
Three-dimensional point clouds have been utilized and studied for the classification of objects at the environmental level. While most existing studies, such as those in the field of computer vision, have detected object type from the perspective of sensors, this study developed a specialized strategy for object classification using LiDAR data points on the surface of the object. We propose a method for generating a spherically stratified point projection (sP2) feature image that can be applied to existing image-classification networks by performing pointwise classification based on a 3D point cloud using only LiDAR sensors data. The sP2’s main engine performs image generation through spherical stratification, evidence collection, and channel integration. Spherical stratification categorizes neighboring points into three layers according to distance ranges. Evidence collection calculates the occupancy probability based on Bayes’ rule to project 3D points onto a two-dimensional surface corresponding to each stratified layer. Channel integration generates sP2 RGB images with three evidence values representing short, medium, and long distances. Finally, the sP2 images are used as a trainable source for classifying the points into predefined semantic labels. Experimental results indicated the effectiveness of the proposed sP2 in classifying feature images generated using the LeNet architecture. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
A Domain-Independent Generative Adversarial Network for Activity Recognition Using WiFi CSI Data
Sensors 2021, 21(23), 7852; https://doi.org/10.3390/s21237852 - 25 Nov 2021
Abstract
Over the past years, device-free sensing has received considerable attention due to its unobtrusiveness. In this regard, context recognition using WiFi Channel State Information (CSI) data has gained popularity, and various techniques have been proposed that combine unobtrusive sensing and deep learning to [...] Read more.
Over the past years, device-free sensing has received considerable attention due to its unobtrusiveness. In this regard, context recognition using WiFi Channel State Information (CSI) data has gained popularity, and various techniques have been proposed that combine unobtrusive sensing and deep learning to accurately detect various contexts ranging from human activities to gestures. However, research has shown that the performance of these techniques significantly degrades due to change in various factors including sensing environment, data collection configuration, diversity of target subjects, and target learning task (e.g., activities, gestures, emotions, vital signs). This problem, generally known as the domain change problem, is typically addressed by collecting more data and learning the data distribution that covers multiple factors impacting the performance. However, activity recognition data collection is a very labor-intensive and time consuming task, and there are too many known and unknown factors impacting WiFi CSI signals. In this paper, we propose a domain-independent generative adversarial network for WiFi CSI based activity recognition in combination with a simplified data pre-processing module. Our evaluation results show superiority of our proposed approach compared to the state of the art in terms of increased robustness against domain change, higher accuracy of activity recognition, and reduced model complexity. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Self-Supervised Denoising Image Filter Based on Recursive Deep Neural Network Structure
Sensors 2021, 21(23), 7827; https://doi.org/10.3390/s21237827 - 24 Nov 2021
Abstract
The purpose of this paper is to propose a novel noise removal method based on deep neural networks that can remove various types of noise without paired noisy and clean data. Because this type of filter generally has relatively poor performance, the proposed [...] Read more.
The purpose of this paper is to propose a novel noise removal method based on deep neural networks that can remove various types of noise without paired noisy and clean data. Because this type of filter generally has relatively poor performance, the proposed noise-to-blur-estimated clean (N2BeC) model introduces a stage-dependent loss function and a recursive learning stage for improved denoised image quality. The proposed loss function regularizes the existing loss function so that the proposed model can better learn image details. Moreover, the recursive learning stage provides the proposed model with an additional opportunity to learn image details. The overall deep neural network consists of three learning stages and three corresponding loss functions. We determine the essential hyperparameters via several simulations. Consequently, the proposed model showed more than 1 dB superior performance compared with the existing noise-to-blur model. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Sensor-Based Human Activity Recognition Using Adaptive Class Hierarchy
Sensors 2021, 21(22), 7743; https://doi.org/10.3390/s21227743 - 21 Nov 2021
Abstract
In sensor-based human activity recognition, many methods based on convolutional neural networks (CNNs) have been proposed. In the typical CNN-based activity recognition model, each class is treated independently of others. However, actual activity classes often have hierarchical relationships. It is important to consider [...] Read more.
In sensor-based human activity recognition, many methods based on convolutional neural networks (CNNs) have been proposed. In the typical CNN-based activity recognition model, each class is treated independently of others. However, actual activity classes often have hierarchical relationships. It is important to consider an activity recognition model that uses the hierarchical relationship among classes to improve recognition performance. In image recognition, branch CNNs (B-CNNs) have been proposed for classification using class hierarchies. B-CNNs can easily perform classification using hand-crafted class hierarchies, but it is difficult to manually design an appropriate class hierarchy when the number of classes is large or there is little prior knowledge. Therefore, in our study, we propose a class hierarchy-adaptive B-CNN, which adds a method to the B-CNN for automatically constructing class hierarchies. Our method constructs the class hierarchy from training data automatically to effectively train the B-CNN without prior knowledge. We evaluated our method on several benchmark datasets for activity recognition. As a result, our method outperformed standard CNN models without considering the hierarchical relationship among classes. In addition, we confirmed that our method has performance comparable to a B-CNN model with a class hierarchy based on human prior knowledge. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Review
Dragonfly Algorithm and Its Hybrids: A Survey on Performance, Objectives and Applications
Sensors 2021, 21(22), 7542; https://doi.org/10.3390/s21227542 - 13 Nov 2021
Abstract
Swarm intelligence is a discipline which makes use of a number of agents for solving optimization problems by producing low cost, fast and robust solutions. The dragonfly algorithm (DA), a recently proposed swarm intelligence algorithm, is inspired by the dynamic and static swarming [...] Read more.
Swarm intelligence is a discipline which makes use of a number of agents for solving optimization problems by producing low cost, fast and robust solutions. The dragonfly algorithm (DA), a recently proposed swarm intelligence algorithm, is inspired by the dynamic and static swarming behaviors of dragonflies, and it has been found to have a higher performance in comparison to other swarm intelligence and evolutionary algorithms in numerous applications. There are only a few surveys about the dragonfly algorithm, and we have found that they are limited in certain aspects. Hence, in this paper, we present a more comprehensive survey about DA, its applications in various domains, and its performance as compared to other swarm intelligence algorithms. We also analyze the hybrids of DA, the methods they employ to enhance the original DA, their performance as compared to the original DA, and their limitations. Moreover, we categorize the hybrids of DA according to the type of problem that they have been applied to, their objectives, and the methods that they utilize. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Memory-Based Pruning of Deep Neural Networks for IoT Devices Applied to Flood Detection
Sensors 2021, 21(22), 7506; https://doi.org/10.3390/s21227506 - 12 Nov 2021
Abstract
Automatic flood detection may be an important component for triggering damage control systems and minimizing the risk of social or economic impacts caused by flooding. Riverside images from regular cameras are a widely available resource that can be used for tackling this problem. [...] Read more.
Automatic flood detection may be an important component for triggering damage control systems and minimizing the risk of social or economic impacts caused by flooding. Riverside images from regular cameras are a widely available resource that can be used for tackling this problem. Nevertheless, state-of-the-art neural networks, the most suitable approach for this type of computer vision task, are usually resource-consuming, which poses a challenge for deploying these models within low-capability Internet of Things (IoT) devices with unstable internet connections. In this work, we propose a deep neural network (DNN) architecture pruning algorithm capable of finding a pruned version of a given DNN within a user-specified memory footprint. Our results demonstrate that our proposed algorithm can find a pruned DNN model with the specified memory footprint with little to no degradation of its segmentation performance. Finally, we show that our algorithm can be used in a memory-constraint wireless sensor network (WSN) employed to detect flooding events of urban rivers, and the resulting pruned models have competitive results compared with the original models. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Infer Thermal Information from Visual Information: A Cross Imaging Modality Edge Learning (CIMEL) Framework
Sensors 2021, 21(22), 7471; https://doi.org/10.3390/s21227471 - 10 Nov 2021
Abstract
The measurement accuracy and reliability of thermography is largely limited by a relatively low spatial-resolution of infrared (IR) cameras in comparison to digital cameras. Using a high-end IR camera to achieve high spatial-resolution can be costly or sometimes infeasible due to the high [...] Read more.
The measurement accuracy and reliability of thermography is largely limited by a relatively low spatial-resolution of infrared (IR) cameras in comparison to digital cameras. Using a high-end IR camera to achieve high spatial-resolution can be costly or sometimes infeasible due to the high sample rate required. Therefore, there is a strong demand to improve the quality of IR images, particularly on edges, without upgrading the hardware in the context of surveillance and industrial inspection systems. This paper proposes a novel Conditional Generative Adversarial Networks (CGAN)-based framework to enhance IR edges by learning high-frequency features from corresponding visual images. A dual-discriminator, focusing on edge and content/background, is introduced to guide the cross imaging modality learning procedure of the U-Net generator in high and low frequencies respectively. Results demonstrate that the proposed framework can effectively enhance barely visible edges in IR images without introducing artefacts, meanwhile the content information is well preserved. Different from most similar studies, this method only requires IR images for testing, which will increase the applicability of some scenarios where only one imaging modality is available, such as active thermography. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Model-Independent Lens Distortion Correction Based on Sub-Pixel Phase Encoding
Sensors 2021, 21(22), 7465; https://doi.org/10.3390/s21227465 - 10 Nov 2021
Abstract
Lens distortion can introduce deviations in visual measurement and positioning. The distortion can be minimized by optimizing the lens and selecting high-quality optical glass, but it cannot be completely eliminated. Most existing correction methods are based on accurate distortion models and stable image [...] Read more.
Lens distortion can introduce deviations in visual measurement and positioning. The distortion can be minimized by optimizing the lens and selecting high-quality optical glass, but it cannot be completely eliminated. Most existing correction methods are based on accurate distortion models and stable image characteristics. However, the distortion is usually a mixture of the radial distortion and the tangential distortion of the lens group, which makes it difficult for the mathematical model to accurately fit the non-uniform distortion. This paper proposes a new model-independent lens complex distortion correction method. Taking the horizontal and vertical stripe pattern as the calibration target, the sub-pixel value distribution visualizes the image distortion, and the correction parameters are directly obtained from the pixel distribution. A quantitative evaluation method suitable for model-independent methods is proposed. The method only calculates the error based on the characteristic points of the corrected picture itself. Experiments show that this method can accurately correct distortion with only 8 pictures, with an error of 0.39 pixels, which provides a simple method for complex lens distortion correction. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Towards Autonomous Drone Racing without GPU Using an OAK-D Smart Camera
Sensors 2021, 21(22), 7436; https://doi.org/10.3390/s21227436 - 09 Nov 2021
Abstract
Recent advances have shown for the first time that it is possible to beat a human with an autonomous drone in a drone race. However, this solution relies heavily on external sensors, specifically on the use of a motion capture system. Thus, a [...] Read more.
Recent advances have shown for the first time that it is possible to beat a human with an autonomous drone in a drone race. However, this solution relies heavily on external sensors, specifically on the use of a motion capture system. Thus, a truly autonomous solution demands performing computationally intensive tasks such as gate detection, drone localisation, and state estimation. To this end, other solutions rely on specialised hardware such as graphics processing units (GPUs) whose onboard hardware versions are not as powerful as those available for desktop and server computers. An alternative is to combine specialised hardware with smart sensors capable of processing specific tasks on the chip, alleviating the need for the onboard processor to perform these computations. Motivated by this, we present the initial results of adapting a novel smart camera, known as the OpenCV AI Kit or OAK-D, as part of a solution for the ADR running entirely on board. This smart camera performs neural inference on the chip that does not use a GPU. It can also perform depth estimation with a stereo rig and run neural network models using images from a 4K colour camera as the input. Additionally, seeking to limit the payload to 200 g, we present a new 3D-printed design of the camera’s back case, reducing the original weight 40%, thus enabling the drone to carry it in tandem with a host onboard computer, the Intel Stick compute, where we run a controller based on gate detection. The latter is performed with a neural model running on an OAK-D at an operation frequency of 40 Hz, enabling the drone to fly at a speed of 2 m/s. We deem these initial results promising toward the development of a truly autonomous solution that will run intensive computational tasks fully on board. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
A Hybrid Deep Learning Model for Recognizing Actions of Distracted Drivers
Sensors 2021, 21(21), 7424; https://doi.org/10.3390/s21217424 - 08 Nov 2021
Abstract
With the rapid spreading of in-vehicle information systems such as smartphones, navigation systems, and radios, the number of traffic accidents caused by driver distractions shows an increasing trend. Timely identification and warning are deemed to be crucial for distracted driving and the establishment [...] Read more.
With the rapid spreading of in-vehicle information systems such as smartphones, navigation systems, and radios, the number of traffic accidents caused by driver distractions shows an increasing trend. Timely identification and warning are deemed to be crucial for distracted driving and the establishment of driver assistance systems is of great value. However, almost all research on the recognition of the driver’s distracted actions using computer vision methods neglected the importance of temporal information for action recognition. This paper proposes a hybrid deep learning model for recognizing the actions of distracted drivers. Specifically, we used OpenPose to obtain skeleton information of the human body and then constructed the vector angle and modulus ratio of the human body structure as features to describe the driver’s actions, thereby realizing the fusion of deep network features and artificial features, which improve the information density of spatial features. The K-means clustering algorithm was used to preselect the original frames, and the method of inter-frame comparison was used to obtain the final keyframe sequence by comparing the Euclidean distance between manually constructed vectors representing frames and the vector representing the cluster center. Finally, we constructed a two-layer long short-term memory neural network to obtain more effective spatiotemporal features, and one softmax layer to identify the distracted driver’s action. The experimental results based on the collected dataset prove the effectiveness of this framework, and it can provide a theoretical basis for the establishment of vehicle distraction warning systems. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
TSInsight: A Local-Global Attribution Framework for Interpretability in Time Series Data
Sensors 2021, 21(21), 7373; https://doi.org/10.3390/s21217373 - 05 Nov 2021
Abstract
With the rise in the employment of deep learning methods in safety-critical scenarios, interpretability is more essential than ever before. Although many different directions regarding interpretability have been explored for visual modalities, time series data has been neglected, with only a handful of [...] Read more.
With the rise in the employment of deep learning methods in safety-critical scenarios, interpretability is more essential than ever before. Although many different directions regarding interpretability have been explored for visual modalities, time series data has been neglected, with only a handful of methods tested due to their poor intelligibility. We approach the problem of interpretability in a novel way by proposing TSInsight, where we attach an auto-encoder to the classifier with a sparsity-inducing norm on its output and fine-tune it based on the gradients from the classifier and a reconstruction penalty. TSInsight learns to preserve features that are important for prediction by the classifier and suppresses those that are irrelevant, i.e., serves as a feature attribution method to boost the interpretability. In contrast to most other attribution frameworks, TSInsight is capable of generating both instance-based and model-based explanations. We evaluated TSInsight along with nine other commonly used attribution methods on eight different time series datasets to validate its efficacy. The evaluation results show that TSInsight naturally achieves output space contraction; therefore, it is an effective tool for the interpretability of deep time series models. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Time Series Segmentation Based on Stationarity Analysis to Improve New Samples Prediction
Sensors 2021, 21(21), 7333; https://doi.org/10.3390/s21217333 - 04 Nov 2021
Abstract
A wide range of applications based on sequential data, named time series, have become increasingly popular in recent years, mainly those based on the Internet of Things (IoT). Several different machine learning algorithms exploit the patterns extracted from sequential data to support multiple [...] Read more.
A wide range of applications based on sequential data, named time series, have become increasingly popular in recent years, mainly those based on the Internet of Things (IoT). Several different machine learning algorithms exploit the patterns extracted from sequential data to support multiple tasks. However, this data can suffer from unreliable readings that can lead to low accuracy models due to the low-quality training sets available. Detecting the change point between high representative segments is an important ally to find and thread biased subsequences. By constructing a framework based on the Augmented Dickey-Fuller (ADF) test for data stationarity, two proposals to automatically segment subsequences in a time series were developed. The former proposal, called Change Detector segmentation, relies on change detection methods of data stream mining. The latter, called ADF-based segmentation, is constructed on a new change detector derived from the ADF test only. Experiments over real-file IoT databases and benchmarks showed the improvement provided by our proposals for prediction tasks with traditional Autoregressive integrated moving average (ARIMA) and Deep Learning (Long short-term memory and Temporal Convolutional Networks) methods. Results obtained by the Long short-term memory predictive model reduced the relative prediction error from 1 to 0.67, compared to time series without segmentation. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Lossless Compression of Sensor Signals Using an Untrained Multi-Channel Recurrent Neural Predictor
Appl. Sci. 2021, 11(21), 10240; https://doi.org/10.3390/app112110240 - 01 Nov 2021
Abstract
The use of sensor applications has been steadily increasing, leading to an urgent need for efficient data compression techniques to facilitate the storage, transmission, and processing of digital signals generated by sensors. Unlike other sequential data such as text sequences, sensor signals have [...] Read more.
The use of sensor applications has been steadily increasing, leading to an urgent need for efficient data compression techniques to facilitate the storage, transmission, and processing of digital signals generated by sensors. Unlike other sequential data such as text sequences, sensor signals have more complex statistical characteristics. Specifically, in every signal point, each bit, which corresponds to a specific precision scale, follows its own conditional distribution depending on its history and even other bits. Therefore, applying existing general-purpose data compressors usually leads to a relatively low compression ratio, since these compressors do not fully exploit such internal features. What is worse, partitioning a bit stream into groups with a preset size will sometimes break the integrity of each signal point. In this paper, we present a lossless data compressor dedicated to compressing sensor signals which is built upon a novel recurrent neural architecture named multi-channel recurrent unit (MCRU). Each channel in the proposed MCRU models a specific precision range of each signal point without breaking data integrity. During compressing and decompressing, the mirrored network will be trained on observed data; thus, no pre-training is needed. The superiority of our approach over other compressors is demonstrated experimentally on various types of sensor signals. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Deep Learning-Based Optimal Smart Shoes Sensor Selection for Energy Expenditure and Heart Rate Estimation
Sensors 2021, 21(21), 7058; https://doi.org/10.3390/s21217058 - 25 Oct 2021
Abstract
Wearable technologies are known to improve our quality of life. Among the various wearable devices, shoes are non-intrusive, lightweight, and can be used for outdoor activities. In this study, we estimated the energy consumption and heart rate in an environment (i.e., running on [...] Read more.
Wearable technologies are known to improve our quality of life. Among the various wearable devices, shoes are non-intrusive, lightweight, and can be used for outdoor activities. In this study, we estimated the energy consumption and heart rate in an environment (i.e., running on a treadmill) using smart shoes equipped with triaxial acceleration, triaxial gyroscope, and four-point pressure sensors. The proposed model uses the latest deep learning architecture which does not require any separate preprocessing. Moreover, it is possible to select the optimal sensor using a channel-wise attention mechanism to weigh the sensors depending on their contributions to the estimation of energy expenditure (EE) and heart rate (HR). The performance of the proposed model was evaluated using the root mean squared error (RMSE), mean absolute error (MAE), and coefficient of determination (R2). Moreover, the RMSE was 1.05 ± 0.15, MAE 0.83 ± 0.12 and R2 0.922 ± 0.005 in EE estimation. On the other hand, and RMSE was 7.87 ± 1.12, MAE 6.21 ± 0.86, and R2 0.897 ± 0.017 in HR estimation. In both estimations, the most effective sensor was the z axis of the accelerometer and gyroscope sensors. Through these results, it is demonstrated that the proposed model could contribute to the improvement of the performance of both EE and HR estimations by effectively selecting the optimal sensors during the active movements of participants. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Biosignal-Based Driving Skill Classification Using Machine Learning: A Case Study of Maritime Navigation
Appl. Sci. 2021, 11(20), 9765; https://doi.org/10.3390/app11209765 - 19 Oct 2021
Abstract
This work presents a novel approach to detecting stress differences between experts and novices in Situation Awareness (SA) tasks during maritime navigation using one type of wearable sensor, Empatica E4 Wristband. We propose that for a given workload state, the values of biosignal [...] Read more.
This work presents a novel approach to detecting stress differences between experts and novices in Situation Awareness (SA) tasks during maritime navigation using one type of wearable sensor, Empatica E4 Wristband. We propose that for a given workload state, the values of biosignal data collected from wearable sensor vary in experts and novices. We describe methods to conduct a designed SA task experiment, and collected the biosignal data on subjects sailing on a 240° view simulator. The biosignal data were analysed by using a machine learning algorithm, a Convolutional Neural Network. The proposed algorithm showed that the biosingal data associated with the experts can be categorized as different from that of the novices, which is in line with the results of NASA Task Load Index (NASA-TLX) rating scores. This study can contribute to the development of a self-training system in maritime navigation in further studies. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Super-Resolution Network with Information Distillation and Multi-Scale Attention for Medical CT Image
Sensors 2021, 21(20), 6870; https://doi.org/10.3390/s21206870 - 16 Oct 2021
Abstract
The CT image is an important reference for clinical diagnosis. However, due to the external influence and equipment limitation in the imaging, the CT image often has problems such as blurring, a lack of detail and unclear edges, which affect the subsequent diagnosis. [...] Read more.
The CT image is an important reference for clinical diagnosis. However, due to the external influence and equipment limitation in the imaging, the CT image often has problems such as blurring, a lack of detail and unclear edges, which affect the subsequent diagnosis. In order to obtain high-quality medical CT images, we propose an information distillation and multi-scale attention network (IDMAN) for medical CT image super-resolution reconstruction. In a deep residual network, instead of only adding the convolution layer repeatedly, we introduce information distillation to make full use of the feature information. In addition, in order to better capture information and focus on more important features, we use a multi-scale attention block with multiple branches, which can automatically generate weights to adjust the network. Through these improvements, our model effectively solves the problems of insufficient feature utilization and single attention source, improves the learning ability and expression ability, and thus can reconstruct the higher quality medical CT image. We conduct a series of experiments; the results show that our method outperforms the previous algorithms and has a better performance of medical CT image reconstruction in the objective evaluation and visual effect. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Filter Pruning via Measuring Feature Map Information
Sensors 2021, 21(19), 6601; https://doi.org/10.3390/s21196601 - 02 Oct 2021
Abstract
Neural network pruning, an important method to reduce the computational complexity of deep models, can be well applied to devices with limited resources. However, most current methods focus on some kind of information about the filter itself to prune the network, rarely exploring [...] Read more.
Neural network pruning, an important method to reduce the computational complexity of deep models, can be well applied to devices with limited resources. However, most current methods focus on some kind of information about the filter itself to prune the network, rarely exploring the relationship between the feature maps and the filters. In this paper, two novel pruning methods are proposed. First, a new pruning method is proposed, which reflects the importance of filters by exploring the information in the feature maps. Based on the premise that the more information there is, more important the feature map is, the information entropy of feature maps is used to measure information, which is used to evaluate the importance of each filter in the current layer. Further, normalization is used to realize cross layer comparison. As a result, based on the method mentioned above, the network structure is efficiently pruned while its performance is well reserved. Second, we proposed a parallel pruning method using the combination of our pruning method above and slimming pruning method which has better results in terms of computational cost. Our methods perform better in terms of accuracy, parameters, and FLOPs compared to most advanced methods. On ImageNet, it is achieved 72.02% top1 accuracy for ResNet50 with merely 11.41M parameters and 1.12B FLOPs.For DenseNet40, it is obtained 94.04% accuracy with only 0.38M parameters and 110.72M FLOPs on CIFAR10, and our parallel pruning method makes the parameters and FLOPs are just 0.37M and 100.12M, respectively, with little loss of accuracy. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
FGFF Descriptor and Modified Hu Moment-Based Hand Gesture Recognition
Sensors 2021, 21(19), 6525; https://doi.org/10.3390/s21196525 - 29 Sep 2021
Abstract
Gesture recognition has been studied for decades and still remains an open problem. One important reason is that the features representing those gestures are not sufficient, which may lead to poor performance and weak robustness. Therefore, this work aims at a comprehensive and [...] Read more.
Gesture recognition has been studied for decades and still remains an open problem. One important reason is that the features representing those gestures are not sufficient, which may lead to poor performance and weak robustness. Therefore, this work aims at a comprehensive and discriminative feature for hand gesture recognition. Here, a distinctive Fingertip Gradient orientation with Finger Fourier (FGFF) descriptor and modified Hu moments are suggested on the platform of a Kinect sensor. Firstly, two algorithms are designed to extract the fingertip-emphasized features, including palm center, fingertips, and their gradient orientations, followed by the finger-emphasized Fourier descriptor to construct the FGFF descriptors. Then, the modified Hu moment invariants with much lower exponents are discussed to encode contour-emphasized structure in the hand region. Finally, a weighted AdaBoost classifier is built based on finger-earth mover’s distance and SVM models to realize the hand gesture recognition. Extensive experiments on a ten-gesture dataset were carried out and compared the proposed algorithm with three benchmark methods to validate its performance. Encouraging results were obtained considering recognition accuracy and efficiency. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Estimation of Various Walking Intensities Based on Wearable Plantar Pressure Sensors Using Artificial Neural Networks
Sensors 2021, 21(19), 6513; https://doi.org/10.3390/s21196513 - 29 Sep 2021
Abstract
Walking has been demonstrated to improve health in people with diabetes and peripheral arterial disease. However, continuous walking can produce repeated stress on the plantar foot and cause a high risk of foot ulcers. In addition, a higher walking intensity (i.e., including different [...] Read more.
Walking has been demonstrated to improve health in people with diabetes and peripheral arterial disease. However, continuous walking can produce repeated stress on the plantar foot and cause a high risk of foot ulcers. In addition, a higher walking intensity (i.e., including different speeds and durations) will increase the risk. Therefore, quantifying the walking intensity is essential for rehabilitation interventions to indicate suitable walking exercise. This study proposed a machine learning model to classify the walking speed and duration using plantar region pressure images. A wearable plantar pressure measurement system was used to measure plantar pressures during walking. An Artificial Neural Network (ANN) was adopted to develop a model for walking intensity classification using different plantar region pressure images, including the first toe (T1), the first metatarsal head (M1), the second metatarsal head (M2), and the heel (HL). The classification consisted of three walking speeds (i.e., slow at 0.8 m/s, moderate at 1.6 m/s, and fast at 2.4 m/s) and two walking durations (i.e., 10 min and 20 min). Of the 12 participants, 10 participants (720 images) were randomly selected to train the classification model, and 2 participants (144 images) were utilized to evaluate the model performance. Experimental evaluation indicated that the ANN model effectively classified different walking speeds and durations based on the plantar region pressure images. Each plantar region pressure image (i.e., T1, M1, M2, and HL) generates different accuracies of the classification model. Higher performance was achieved when classifying walking speeds (0.8 m/s, 1.6 m/s, and 2.4 m/s) and 10 min walking duration in the T1 region, evidenced by an F1-score of 0.94. The dataset T1 could be an essential variable in machine learning to classify the walking intensity at different speeds and durations. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
A Class-Imbalanced Deep Learning Fall Detection Algorithm Using Wearable Sensors
Sensors 2021, 21(19), 6511; https://doi.org/10.3390/s21196511 - 29 Sep 2021
Abstract
Falling represents one of the most serious health risks for elderly people; it may cause irreversible injuries if the individual cannot obtain timely treatment after the fall happens. Therefore, timely and accurate fall detection algorithm research is extremely important. Recently, a number of [...] Read more.
Falling represents one of the most serious health risks for elderly people; it may cause irreversible injuries if the individual cannot obtain timely treatment after the fall happens. Therefore, timely and accurate fall detection algorithm research is extremely important. Recently, a number of researchers have focused on fall detection and made many achievements, and most of the relevant algorithm studies are based on ideal class-balanced datasets. However, in real-life applications, the possibilities of Activities of Daily Life (ADL) and fall events are different, so the data collected by wearable sensors suffers from class imbalance. The previously developed algorithms perform poorly on class-imbalanced data. In order to solve this problem, this paper proposes an algorithm that can effectively distinguish falls from a large amount of ADL signals. Compared with the state-of-the-art fall detection algorithms, the proposed method can achieve the highest score in multiple evaluation methods, with a sensitivity of 99.33%, a specificity of 91.86%, an F-Score of 98.44% and an AUC of 98.35%. The results prove that the proposed algorithm is effective on class-imbalanced data and more suitable for real-life application compared to previous works. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Estimation of Mechanical Power Output Employing Deep Learning on Inertial Measurement Data in Roller Ski Skating
Sensors 2021, 21(19), 6500; https://doi.org/10.3390/s21196500 - 29 Sep 2021
Abstract
The ability to optimize power generation in sports is imperative, both for understanding and balancing training load correctly, and for optimizing competition performance. In this paper, we aim to estimate mechanical power output by employing a time-sequential information-based deep Long Short-Term Memory (LSTM) [...] Read more.
The ability to optimize power generation in sports is imperative, both for understanding and balancing training load correctly, and for optimizing competition performance. In this paper, we aim to estimate mechanical power output by employing a time-sequential information-based deep Long Short-Term Memory (LSTM) neural network from multiple inertial measurement units (IMUs). Thirteen athletes conducted roller ski skating trials on a treadmill with varying incline and speed. The acceleration and gyroscope data collected with the IMUs were run through statistical feature processing, before being used by the deep learning model to estimate power output. The model was thereafter used for prediction of power from test data using two approaches. First, a user-dependent case was explored, reaching a power estimation within 3.5% error. Second, a user-independent case was developed, reaching an error of 11.6% for the power estimation. Finally, the LSTM model was compared to two other machine learning models and was found to be superior. In conclusion, the user-dependent model allows for precise estimation of roller skiing power output after training the model on data from each athlete. The user-independent model provides less accurate estimation; however, the accuracy may be sufficient for providing valuable information for recreational skiers. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Recovery of Ionospheric Signals Using Fully Convolutional DenseNet and Its Challenges
Sensors 2021, 21(19), 6482; https://doi.org/10.3390/s21196482 - 28 Sep 2021
Abstract
The technique of active ionospheric sounding by ionosondes requires sophisticated methods for the recovery of experimental data on ionograms. In this work, we applied an advanced algorithm of deep learning for the identification and classification of signals from different ionospheric layers. We collected [...] Read more.
The technique of active ionospheric sounding by ionosondes requires sophisticated methods for the recovery of experimental data on ionograms. In this work, we applied an advanced algorithm of deep learning for the identification and classification of signals from different ionospheric layers. We collected a dataset of 6131 manually labeled ionograms acquired from low-latitude ionosondes in Taiwan. In the ionograms, we distinguished 11 different classes of the signals according to their ionospheric layers. We developed an artificial neural network, FC-DenseNet24, based on the FC-DenseNet convolutional neural network. We also developed a double-filtering algorithm to reduce incorrectly classified signals. That made it possible to successfully recover the sporadic E layer and the F2 layer from highly noise-contaminated ionograms whose mean signal-to-noise ratio was low, SNR = 1.43. The Intersection over Union (IoU) of the recovery of these two signal classes was greater than 0.6, which was higher than the previous models reported. We also identified three factors that can lower the recovery accuracy: (1) smaller statistics of samples; (2) mixing and overlapping of different signals; (3) the compact shape of signals. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Data-Driven Investigation of Gait Patterns in Individuals Affected by Normal Pressure Hydrocephalus
Sensors 2021, 21(19), 6451; https://doi.org/10.3390/s21196451 - 27 Sep 2021
Abstract
Normal pressure hydrocephalus (NPH) is a chronic and progressive disease that affects predominantly elderly subjects. The most prevalent symptoms are gait disorders, generally determined by visual observation or measurements taken in complex laboratory environments. However, controlled testing environments can have a significant influence [...] Read more.
Normal pressure hydrocephalus (NPH) is a chronic and progressive disease that affects predominantly elderly subjects. The most prevalent symptoms are gait disorders, generally determined by visual observation or measurements taken in complex laboratory environments. However, controlled testing environments can have a significant influence on the way subjects walk and hinder the identification of natural walking characteristics. The study aimed to investigate the differences in walking patterns between a controlled environment (10 m walking test) and real-world environment (72 h recording) based on measurements taken via a wearable gait assessment device. We tested whether real-world environment measurements can be beneficial for the identification of gait disorders by performing a comparison of patients’ gait parameters with an aged-matched control group in both environments. Subsequently, we implemented four machine learning classifiers to inspect the individual strides’ profiles. Our results on twenty young subjects, twenty elderly subjects and twelve NPH patients indicate that patients exhibited a considerable difference between the two environments, in particular gait speed (p-value p=0.0073), stride length (p-value p=0.0073), foot clearance (p-value p=0.0117) and swing/stance ratio (p-value p=0.0098). Importantly, measurements taken in real-world environments yield a better discrimination of NPH patients compared to the controlled setting. Finally, the use of stride classifiers provides promise in the identification of strides affected by motion disorders. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Graphical abstract

Article
An Embeddable Algorithm for Automatic Garbage Detection Based on Complex Marine Environment
Sensors 2021, 21(19), 6391; https://doi.org/10.3390/s21196391 - 24 Sep 2021
Abstract
With the continuous development of artificial intelligence, embedding object detection algorithms into autonomous underwater detectors for marine garbage cleanup has become an emerging application area. Considering the complexity of the marine environment and the low resolution of the images taken by underwater detectors, [...] Read more.
With the continuous development of artificial intelligence, embedding object detection algorithms into autonomous underwater detectors for marine garbage cleanup has become an emerging application area. Considering the complexity of the marine environment and the low resolution of the images taken by underwater detectors, this paper proposes an improved algorithm based on Mask R-CNN, with the aim of achieving high accuracy marine garbage detection and instance segmentation. First, the idea of dilated convolution is introduced in the Feature Pyramid Network to enhance feature extraction ability for small objects. Secondly, the spatial-channel attention mechanism is used to make features learn adaptively. It can effectively focus attention on detection objects. Third, the re-scoring branch is added to improve the accuracy of instance segmentation by scoring the predicted masks based on the method of Generalized Intersection over Union. Finally, we train the proposed algorithm in this paper on the Transcan dataset, evaluating its effectiveness by various metrics and comparing it with existing algorithms. The experimental results show that compared to the baseline provided by the Transcan dataset, the algorithm in this paper improves the mAP indexes on the two tasks of garbage detection and instance segmentation by 9.6 and 5.0, respectively, which significantly improves the algorithm performance. Thus, it can be better applied in the marine environment and achieve high precision object detection and instance segmentation. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
The Seismo-Performer: A Novel Machine Learning Approach for General and Efficient Seismic Phase Recognition from Local Earthquakes in Real Time
Sensors 2021, 21(18), 6290; https://doi.org/10.3390/s21186290 - 19 Sep 2021
Abstract
When recording seismic ground motion in multiple sites using independent recording stations one needs to recognize the presence of the same parts of seismic waves arriving at these stations. This problem is known in seismology as seismic phase picking. It is challenging to [...] Read more.
When recording seismic ground motion in multiple sites using independent recording stations one needs to recognize the presence of the same parts of seismic waves arriving at these stations. This problem is known in seismology as seismic phase picking. It is challenging to automate the accurate picking of seismic phases to the level of human capabilities. By solving this problem, it would be possible to automate routine processing in real time on any local network. A new machine learning approach was developed to classify seismic phases from local earthquakes. The resulting model is based on spectrograms and utilizes the transformer architecture with a self-attention mechanism and without any convolution blocks. The model is general for various local networks and has only 57 k learning parameters. To assess the generalization property, two new datasets were developed, containing local earthquake data collected from two different regions using a wide variety of seismic instruments. The data were not involved in the training process for any model to estimate the generalization property. The new model exhibits the best classification and computation performance results on its pre-trained weights compared with baseline models from related work. The model code is available online and is ready for day-to-day real-time processing on conventional seismic equipment without graphics processing units. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Human Activity Classification Using Multilayer Perceptron
Sensors 2021, 21(18), 6207; https://doi.org/10.3390/s21186207 - 16 Sep 2021
Abstract
The number of smart homes is rapidly increasing. Smart homes typically feature functions such as voice-activated functions, automation, monitoring, and tracking events. Besides comfort and convenience, the integration of smart home functionality with data processing methods can provide valuable information about the well-being [...] Read more.
The number of smart homes is rapidly increasing. Smart homes typically feature functions such as voice-activated functions, automation, monitoring, and tracking events. Besides comfort and convenience, the integration of smart home functionality with data processing methods can provide valuable information about the well-being of the smart home residence. This study is aimed at taking the data analysis within smart homes beyond occupancy monitoring and fall detection. This work uses a multilayer perceptron neural network to recognize multiple human activities from wrist- and ankle-worn devices. The developed models show very high recognition accuracy across all activity classes. The cross-validation results indicate accuracy levels above 98% across all models, and scoring evaluation methods only resulted in an average accuracy reduction of 10%. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Photo-Realistic Image Dehazing and Verifying Networks via Complementary Adversarial Learning
Sensors 2021, 21(18), 6182; https://doi.org/10.3390/s21186182 - 15 Sep 2021
Abstract
Physical model-based dehazing methods cannot, in general, avoid environmental variables and undesired artifacts such as non-collected illuminance, halo and saturation since it is difficult to accurately estimate the amount of the illuminance, light transmission and airlight. Furthermore, the haze model estimation process requires [...] Read more.
Physical model-based dehazing methods cannot, in general, avoid environmental variables and undesired artifacts such as non-collected illuminance, halo and saturation since it is difficult to accurately estimate the amount of the illuminance, light transmission and airlight. Furthermore, the haze model estimation process requires very high computational complexity. To solve this problem by directly estimating the radiance of the haze images, we present a novel dehazing and verifying network (DVNet). In the dehazing procedure, we enhanced the clean images by using a correction network (CNet), which uses the ground truth to learn the haze network. Haze images are then restored through a haze network (HNet). Furthermore, a verifying method verifies the error of both CNet and HNet using a self-supervised learning method. Finally, the proposed complementary adversarial learning method can produce results more naturally. Note that the proposed discriminator and generators (HNet & CNet) can be learned via an unpaired dataset. Overall, the proposed DVNet can generate a better dehazed result than state-of-the-art approaches under various hazy conditions. Experimental results show that the DVNet outperforms state-of-the-art dehazing methods in most cases. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Automatic Inside Point Localization with Deep Reinforcement Learning for Interactive Object Segmentation
Sensors 2021, 21(18), 6100; https://doi.org/10.3390/s21186100 - 11 Sep 2021
Abstract
In the task of interactive image segmentation, the Inside-Outside Guidance (IOG) algorithm has demonstrated superior segmentation performance leveraging Inside-Outside Guidance information. Nevertheless, we observe that the inconsistent input between training and testing when selecting the inside point will result in significant performance degradation. [...] Read more.
In the task of interactive image segmentation, the Inside-Outside Guidance (IOG) algorithm has demonstrated superior segmentation performance leveraging Inside-Outside Guidance information. Nevertheless, we observe that the inconsistent input between training and testing when selecting the inside point will result in significant performance degradation. In this paper, a deep reinforcement learning framework, named Inside Point Localization Network (IPL-Net), is proposed to infer the suitable position for the inside point to help the IOG algorithm. Concretely, when a user first clicks two outside points at the symmetrical corner locations of the target object, our proposed system automatically generates the sequence of movement to localize the inside point. We then perform the IOG interactive segmentation method for precisely segmenting the target object of interest. The inside point localization problem is difficult to define as a supervised learning framework because it is expensive to collect image and their corresponding inside points. Therefore, we formulate this problem as Markov Decision Process (MDP) and then optimize it with Dueling Double Deep Q-Network (D3QN). We train our network on the PASCAL dataset and demonstrate that the network achieves excellent performance. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
A Sensor Fused Rear Cross Traffic Detection System Using Transfer Learning
Sensors 2021, 21(18), 6055; https://doi.org/10.3390/s21186055 - 09 Sep 2021
Abstract
Recent emerging automotive sensors and innovative technologies in Advanced Driver Assistance Systems (ADAS) increase the safety of driving a vehicle on the road. ADAS enhance road safety by providing early warning signals for drivers and controlling a vehicle accordingly to mitigate a collision. [...] Read more.
Recent emerging automotive sensors and innovative technologies in Advanced Driver Assistance Systems (ADAS) increase the safety of driving a vehicle on the road. ADAS enhance road safety by providing early warning signals for drivers and controlling a vehicle accordingly to mitigate a collision. A Rear Cross Traffic (RCT) detection system is an important application of ADAS. Rear-end crashes are a frequently occurring type of collision, and approximately 29.7% of all crashes are rear-ended collisions. The RCT detection system detects obstacles at the rear while the car is backing up. In this paper, a robust sensor fused RCT detection system is proposed. By combining the information from two radars and a wide-angle camera, the locations of the target objects are identified using the proposed sensor fused algorithm. Then, the transferred Convolution Neural Network (CNN) model is used to classify the object type. The experiments show that the proposed sensor fused RCT detection system reduced the processing time 15.34 times faster than the camera-only system. The proposed system has achieved 96.42% accuracy. The experimental results demonstrate that the proposed sensor fused system has robust object detection accuracy and fast processing time, which is vital for deploying the ADAS system. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Article
Tactile Object Recognition for Humanoid Robots Using New Designed Piezoresistive Tactile Sensor and DCNN
Sensors 2021, 21(18), 6024; https://doi.org/10.3390/s21186024 - 08 Sep 2021
Abstract
A tactile sensor array is a crucial component for applying physical sensors to a humanoid robot. This work focused on developing a palm-size tactile sensor array (56.0 mm × 56.0 mm) to apply object recognition for the humanoid robot hand. This sensor was [...] Read more.
A tactile sensor array is a crucial component for applying physical sensors to a humanoid robot. This work focused on developing a palm-size tactile sensor array (56.0 mm × 56.0 mm) to apply object recognition for the humanoid robot hand. This sensor was based on a PCB technology operating with the piezoresistive principle. A conductive polymer composites sheet was used as a sensing element and the matrix array of this sensor was 16 × 16 pixels. The sensitivity of this sensor was evaluated and the sensor was installed on the robot hand. The tactile images, with resolution enhancement using bicubic interpolation obtained from 20 classes, were used to train and test 19 different DCNNs. InceptionResNetV2 provided superior performance with 91.82% accuracy. However, using the multimodal learning method that included InceptionResNetV2 and XceptionNet, the highest recognition rate of 92.73% was achieved. Moreover, this recognition rate improved when the object exploration was applied to demonstrate. Full article
(This article belongs to the Topic Artificial Intelligence in Sensors)
Show Figures

Figure 1

Back to TopTop