Forest fires can destroy forest and inflict great damage to the ecosystem. Fortunately, forest fire detection with video has achieved remarkable results in enabling timely and accurate fire warnings. However, the traditional forest fire detection method relies heavily on artificially designed features; CNN-based methods require a large number of parameters. In addition, forest fire detection is easily disturbed by fog. To solve these issues, a lightweight YOLOX-L and defogging algorithm-based forest fire detection method, GXLD, is proposed. GXLD uses the dark channel prior to defog the image to obtain a fog-free image. After the lightweight improvement of YOLOX-L by GhostNet, depth separable convolution, and SENet, we obtain the YOLOX-L-Light and use it to detect the forest fire in the fog-free image. To evaluate the performance of YOLOX-L-Light and GXLD, mean average precision (mAP) was used to evaluate the detection accuracy, and network parameters were used to evaluate the lightweight effect. Experiments on our forest fire dataset show that the number of the parameters of YOLOX-L-Light decreased by 92.6%, and the mAP increased by 1.96%. The mAP of GXLD is 87.47%, which is 2.46% higher than that of YOLOX-L; and the average fps of GXLD is 26.33 when the input image size is 1280 × 720. Even in a foggy environment, the GXLD can detect a forest fire in real time with a high accuracy, target confidence, and target integrity. This research proposes a lightweight forest fire detection method (GXLD) with fog removal. Therefore, GXLD can detect a forest fire with a high accuracy in real time. The proposed GXLD has the advantages of defogging, a high target confidence, and a high target integrity, which makes it more suitable for the development of a modern forest fire video detection system. Full article

High-resolution image transmission is required in safety helmet detection problems in the construction industry, which makes it difficult for existing image detection methods to achieve high-speed detection. To overcome this problem, a novel super-resolution (SR) reconstruction module is designed to improve the resolution of images before the detection module. In the super-resolution reconstruction module, the multichannel attention mechanism module is used to improve the breadth of feature capture. Furthermore, a novel CSP (Cross Stage Partial) module of YOLO (You Only Look Once) v5 is presented to reduce information loss and gradient confusion. Experiments are performed to validate the proposed algorithm. The PSNR (peak signal-to-noise ratio) of the proposed module is 29.420, and the SSIM (structural similarity) reaches 0.855. These results show that the proposed model works well for safety helmet detection in construction industries. Full article

Since the advent of visual sensors, smart cities have generated massive surveillance video data, which can be intelligently inspected to detect anomalies. Computer vision-based automated anomaly detection techniques replace human intervention to secure video surveillance applications in place from traditional video surveillance systems that rely on human involvement for anomaly detection, which is tedious and inaccurate. Due to the diverse nature of anomalous events and their complexity, it is however, very challenging to detect them automatically in a real-world scenario. By using Artificial Intelligence of Things (AIoT), this research work presents an efficient and robust framework for detecting anomalies in surveillance large video data. A hybrid model integrating 2D-CNN and ESN are proposed in this research study for smart surveillance, which is an important application of AIoT. The CNN is used as feature extractor from input videos which are then inputted to autoencoder for feature refinement followed by ESN for sequence learning and anomalous events detection. The proposed model is lightweight and implemented over edge devices to ensure their capability and applicability over AIoT environments in a smart city. The proposed model significantly enhanced performance using challenging surveillance datasets compared to other methods. Full article

With the proliferation of IoT applications, more and more smart, connected devices will be required to communicate with one another, operating in situations that involve diverse levels of range and cost requirements, user interactions, mobility, and energy constraints. Wireless technologies that can satisfy the aforementioned requirements will be vital to realise emerging market opportunities in the IoT sector. Bluetooth Mesh is a new wireless protocol that extends the core Bluetooth low energy (BLE) stack and promises to support reliable and scalable IoT systems where thousands of devices such as sensors, smartphones, wearables, robots, and everyday appliances operate together. In this article, we present a comprehensive discussion on current research directions and existing use cases for Bluetooth Mesh, with recommendations for best practices so that researchers and practitioners can better understand how they can use Bluetooth Mesh in IoT scenarios. Full article

Recent advances in unmanned aerial vehicles (UAV), mini and mobile sensors, and GeoAI (a blend of geospatial and artificial intelligence (AI) research) are the main highlights among agricultural innovations to improve crop productivity and thus secure vulnerable food systems. This study investigated the versatility of UAV-borne multisensory data fusion within a framework of multi-task deep learning for high-throughput phenotyping in maize. UAVs equipped with a set of miniaturized sensors including hyperspectral, thermal, and LiDAR were collected in an experimental corn field in Urbana, IL, USA during the growing season. A full suite of eight phenotypes was in situ measured at the end of the season for ground truth data, specifically, dry stalk biomass, cob biomass, dry grain yield, harvest index, grain nitrogen utilization efficiency (Grain NutE), grain nitrogen content, total plant nitrogen content, and grain density. After being funneled through a series of radiometric calibrations and geo-corrections, the aerial data were analytically processed in three primary approaches. First, an extended version normalized difference spectral index (NDSI) served as a simple arithmetic combination of different data modalities to explore the correlation degree with maize phenotypes. The extended NDSI analysis revealed the NIR spectra (750–1000 nm) alone in a strong relation with all of eight maize traits. Second, a fusion of vegetation indices, structural indices, and thermal index selectively handcrafted from each data modality was fed to classical machine learning regressors, Support Vector Machine (SVM) and Random Forest (RF). The prediction performance varied from phenotype to phenotype, ranging from R² = 0.34 for grain density up to R² = 0.85 for both grain nitrogen content and total plant nitrogen content. Further, a fusion of hyperspectral and LiDAR data completely exceeded limitations of single data modality, especially addressing the vegetation saturation effect occurring in optical remote sensing. Third, a multi-task deep convolutional neural network (CNN) was customized to take a raw imagery data fusion of hyperspectral, thermal, and LiDAR for multi-predictions of maize traits at a time. The multi-task deep learning performed predictions comparably, if not better in some traits, with the mono-task deep learning and machine learning regressors. Data augmentation used for the deep learning models boosted the prediction accuracy, which helps to alleviate the intrinsic limitation of a small sample size and unbalanced sample classes in remote sensing research. Theoretical and practical implications to plant breeders and crop growers were also made explicit during discussions in the studies. Full article

Machine learning is an expanding field with an ever-increasing role in everyday life, with its utility in the industrial, agricultural, and medical sectors being undeniable. Recently, this utility has come in the form of machine learning implementation on embedded system devices. While there have been steady advances in the performance, memory, and power consumption of embedded devices, most machine learning algorithms still have a very high power consumption and computational demand, making the implementation of embedded machine learning somewhat difficult. However, different devices can be implemented for different applications based on their overall processing power and performance. This paper presents an overview of several different implementations of machine learning on embedded systems divided by their specific device, application, specific machine learning algorithm, and sensors. We will mainly focus on NVIDIA Jetson and Raspberry Pi devices with a few different less utilized embedded computers, as well as which of these devices were more commonly used for specific applications in different fields. We will also briefly analyze the specific ML models most commonly implemented on the devices and the specific sensors that were used to gather input from the field. All of the papers included in this review were selected using Google Scholar and published papers in the IEEExplore database. The selection criterion for these papers was the usage of embedded computing systems in either a theoretical study or practical implementation of machine learning models. The papers needed to have provided either one or, preferably, all of the following results in their studies—the overall accuracy of the models on the system, the overall power consumption of the embedded machine learning system, and the inference time of their models on the embedded system. Embedded machine learning is experiencing an explosion in both scale and scope, both due to advances in system performance and machine learning models, as well as greater affordability and accessibility of both. Improvements are noted in quality, power usage, and effectiveness. Full article

Due to the recent advances in the domain of smart agriculture as a result of integrating traditional agriculture and the latest information technologies including the Internet of Things (IoT), cloud computing, and artificial intelligence (AI), there is an urgent need to address the information security-related issues and challenges in this field. In this article, we propose the integration of lightweight cryptography techniques into the IoT ecosystem for smart agriculture to meet the requirements of resource-constrained IoT devices. Moreover, we investigate the adoption of a lightweight encryption protocol, namely, the Expeditious Cipher (X-cipher), to create a secure channel between the sensing layer and the broker in the Message Queue Telemetry Transport (MQTT) protocol as well as a secure channel between the broker and its subscribers. Our case study focuses on smart irrigation systems, and the MQTT protocol is deployed as the application messaging protocol in these systems. Smart irrigation strives to decrease the misuse of natural resources by enhancing the efficiency of agricultural irrigation. This secure channel is utilized to eliminate the main security threat in precision agriculture by protecting sensors’ published data from eavesdropping and theft, as well as from unauthorized changes to sensitive data that can negatively impact crops’ development. In addition, the secure channel protects the irrigation decisions made by the data analytics (DA) entity regarding the irrigation time and the quantity of water that is returned to actuators from any alteration. Performance evaluation of our chosen lightweight encryption protocol revealed an improvement in terms of power consumption, execution time, and required memory usage when compared with the Advanced Encryption Standard (AES). Moreover, the selected lightweight encryption protocol outperforms the PRESENT lightweight encryption protocol in terms of throughput and memory usage. Full article

Due to climate change, soil moisture may increase, and outflows could become more frequent, which will have a considerable impact on crop growth. Crops are affected by soil moisture; thus, soil moisture prediction is necessary for irrigating at an appropriate time according to weather changes. Therefore, the aim of this study is to develop a future soil moisture (SM) prediction model to determine whether to conduct irrigation according to changes in soil moisture due to weather conditions. Sensors were used to measure soil moisture and soil temperature at a depth of 10 cm, 20 cm, and 30 cm from the topsoil. The combination of optimal variables was investigated using soil moisture and soil temperature at depths between 10 cm and 30 cm and weather data as input variables. The recurrent neural network long short-term memory (RNN-LSTM) models for predicting SM was developed using time series data. The loss and the coefficient of determination (R²) values were used as indicators for evaluating the model performance and two verification datasets were used to test various conditions. The best model performance for 10 cm depth was an R² of 0.999, a loss of 0.022, and a validation loss of 0.105, and the best results for 20 cm and 30 cm depths were an R² of 0.999, a loss of 0.016, and a validation loss of 0.098 and an R² of 0.956, a loss of 0.057, and a validation loss of 2.883, respectively. The RNN-LSTM model was used to confirm the SM predictability in soybean arable land and could be applied to supply the appropriate moisture needed for crop growth. The results of this study show that a soil moisture prediction model based on time-series weather data can help determine the appropriate amount of irrigation required for crop cultivation. Full article

In this study, a snapshot-based hyperspectral imaging (HSI) algorithm that converts RGB images to HSI images is designed using the Raspberry Pi environment. A Windows-based Python application is also developed to control the Raspberry Pi camera and processor. The mean gray values (MGVs) of two distinct regions of interest (ROIs) are selected from three samples of 100 NTD Taiwanese currency notes and compared with three samples of counterfeit 100 NTD notes. Results suggest that the currency notes can be easily differentiated on the basis of MGV values within shorter wavelengths, between 400 nm and 500 nm. However, the MGV values are similar in longer wavelengths. Moreover, if an ROI has a security feature, then the classification method is considerably more efficient. The key features of the module include portability, lower cost, a lack of moving parts, and no processing of images required. Full article

Real-time monitoring of large marine structures’ health, including drilling platforms, submarine pipelines, dams, and ship hulls, is greatly needed. Among the various kinds of monitoring methods, optical fiber sensors (OFS) have gained a lot of concerns and showed several distinct advantages, such as small size, high flexibility and durability, anti-electromagnetic interference, and high transmission rate. In this paper, three types of OFS used for marine structural health monitoring (SHM), including point sensing, quasi-distributed sensing, and distributed sensing, are reviewed. Emphases are given to the applicability of each type of the sensors by analyzing the operating principles and characteristics of the OFSs. The merits and demerits of different sensing schemes are discussed, as well as the challenges and future developments in OFSs for the marine SHM field. Full article

Ensuring safe food supplies has recently become a serious problem all over the world. Controlling the quality, spoilage, and standing time for products with a short shelf life is a quite difficult problem. However, electronic noses can make all these controls possible. In this study, which aims to develop a different approach to the solution of this problem, electronic nose data obtained from 12 different beef cuts were classified. In the dataset, there are four classes (1: excellent, 2: good, 3: acceptable, and 4: spoiled) indicating beef quality. The classifications were performed separately for each cut and all cut shapes. The ANOVA method was used to determine the active features in the dataset with data for 12 features. The same classification processes were carried out by using the three active features selected by the ANOVA method. Three different machine learning methods, Artificial Neural Network, K Nearest Neighbor, and Logistic Regression, which are frequently used in the literature, were used in classifications. In the experimental studies, a classification accuracy of 100% was obtained as a result of the classification performed with ANN using the data obtained by combining all the tables in the dataset. Full article

Humans are a vision-dominated species; what we perceive depends on where we look. Therefore, eye movements (EMs) are essential to our interactions with the environment, and experimental findings show EMs are affected in neurodegenerative disorders (ND). This could be a reason for some cognitive and movement disorders in ND. Therefore, we aim to establish whether changes in EM-evoked responses can tell us about the progression of ND, such as Alzheimer’s (AD) and Parkinson’s diseases (PD), in different stages. In the present review, we have analyzed the results of psychological, neurological, and EM (saccades, antisaccades, pursuit) tests to predict disease progression with machine learning (ML) methods. Thanks to ML algorithms, from the high-dimensional parameter space, we were able to find significant EM changes related to ND symptoms that gave us insights into ND mechanisms. The predictive algorithms described use various approaches, including granular computing, Naive Bayes, Decision Trees/Tables, logistic regression, C-/Linear SVC, KNC, and Random Forest. We demonstrated that EM is a robust biomarker for assessing symptom progression in PD and AD. There are navigation problems in 3D space in both diseases. Consequently, we investigated EM experiments in the virtual space and how they may help find neurodegeneration-related brain changes, e.g., related to place or/and orientation problems. In conclusion, EM parameters with clinical symptoms are powerful precision instruments that, in addition to their potential for predictions of ND progression with the help of ML, could be used to indicate the different preclinical stages of both diseases. Full article

We designed a tunable diode laser absorption spectroscopy (TDLAS) sensor for the online monitoring of CO₂ and H₂O concentrations. It comprised a small self-design multi-pass cell, home-made laser drive circuits, and a data acquisition circuit. The optical and electrical parts and the gas circuit were integrated into a portable carrying case (height = 134 mm, length = 388 mm, and width = 290 mm). A TDLAS drive module (size: 90 mm × 45 mm) was designed to realize the function of laser current and temperature control with a temperature control accuracy of ±1.4 mK and a current control accuracy of ±0.5 μA, and signal acquisition and demodulation. The weight and power consumption of the TDLAS system were only 5 kg and 10 W, respectively. Distributed feedback lasers (2004 nm and 1392 nm) were employed to target CO₂ and H₂O absorption lines, respectively. According to Allan analysis, the detection limits of CO₂ and H₂O were 0.13 ppm and 3.7 ppm at an average time of 18 s and 35 s, respectively. The system response time was approximately 10 s. Sensor performance was verified by measuring atmospheric CO₂ and H₂O concentrations for 240 h. Experimental results were compared with those obtained using a commercial instrument LI-7500, which uses non-dispersive infrared technology. Measurements of the developed gas analyzer were in good agreement with those of the commercial instrument, and its accuracy was comparable. Therefore, the TDLAS sensor has strong application prospects in atmospheric CO₂ and H₂O concentration detection and ecological soil flux monitoring. Full article

A detachable miniaturized three-element spirals radiator button antenna integrated with a compact leaky-wave wearable antenna forming a dual-band three-port antenna is proposed. The leaky-wave antenna is fabricated on a denim (ε_r = 1.6, tan δ = 0.006) textile substrate with dimensions of 0.37 λ₀ × 0.25 λ₀ × 0.01 λ₀ mm³ and a detachable rigid button of 20 mm diameter (on a PTFE substrate ε_r = 2.01, tan δ = 0.001). It augments users’ comfort, making it one of the smallest to date in the literature. The designed antenna, with 3.25 to 3.65 GHz and 5.4 to 5.85 GHz operational bands, covers the wireless local area network (WLAN) frequency (5.1–5.5 GHz), the fifth-generation (5G) communication band. Low mutual coupling between the ports and the button antenna elements ensures high diversity performance. The performance of the specific absorption rate (SAR) and the envelope correlation coefficient (ECC) are also examined. The simulation and measurement findings agree well. Low SAR, <−0.05 of LCC, more than 9.5 dBi diversity gain, dual polarization, and strong isolation between every two ports all point to the proposed antenna being an ideal option for use as a MIMO antenna for communications. Full article

Reliable fault diagnosis and condition monitoring are essential for permanent magnet synchronous motor (PMSM) drive systems with high-reliability requirements. PMSMs can be subject to various types of damage during operation. Magnetic damage is a unique fault of PMSM and concerns the permanent magnet (PM) of the rotor. PM damage may be mechanical in nature or be related to the phenomenon of demagnetization. This article presents a machine learning (ML) based demagnetization fault diagnosis method for PMSM drives. The time-frequency domain analysis based on short-time Fourier transform (STFT) is applied in the process of PM fault feature extraction from the stator phase current signal. Moreover, two ML-based models are verified and compared in the process of the automatic fault detection of demagnetization fault. These models are k-nearest neighbors (KNN) and multiLayer perceptron (MLP). The influence of the input vector elements, key parameters and structures of the models used on their effectiveness is extensively analyzed. The results of the experimental verification confirm the very high effectiveness of the proposed method. Full article