Research

Open AccessArticle

Three-Dimensional Reconstruction from Single Image Base on Combination of CNN and Multi-Spectral Photometric Stereo

by

,

,

,

and

Sensors 2018, 18(3), 764; https://doi.org/10.3390/s18030764 - 02 Mar 2018

Cited by 12 | Viewed by 5080

Abstract

Multi-spectral photometric stereo can recover pixel-wise surface normal from a single RGB image. The difficulty lies in that the intensity in each channel is the tangle of illumination, albedo and camera response; thus, an initial estimate of the normal is required in optimization-based [...] Read more.

Multi-spectral photometric stereo can recover pixel-wise surface normal from a single RGB image. The difficulty lies in that the intensity in each channel is the tangle of illumination, albedo and camera response; thus, an initial estimate of the normal is required in optimization-based solutions. In this paper, we propose to make a rough depth estimation using the deep convolutional neural network (CNN) instead of using depth sensors or binocular stereo devices. Since high-resolution ground-truth data is expensive to obtain, we designed a network and trained it with rendered images of synthetic 3D objects. We use the model to predict initial normal of real-world objects and iteratively optimize the fine-scale geometry in the multi-spectral photometric stereo framework. The experimental results illustrate the improvement of the proposed method compared with existing methods. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

Archetype-Based Modeling of Persona for Comprehensive Personality Computing from Personal Big Data

by

Ao Guo

and

Jianhua Ma

Sensors 2018, 18(3), 684; https://doi.org/10.3390/s18030684 - 25 Feb 2018

Cited by 12 | Viewed by 9329

Abstract

A model describing the wide variety of human behaviours called personality, is becoming increasingly popular among researchers due to the widespread availability of personal big data generated from the use of prevalent digital devices, e.g., smartphones and wearables. Such an approach can be [...] Read more.

A model describing the wide variety of human behaviours called personality, is becoming increasingly popular among researchers due to the widespread availability of personal big data generated from the use of prevalent digital devices, e.g., smartphones and wearables. Such an approach can be used to model an individual and even digitally clone a person, e.g., a Cyber-I (cyber individual). This work is aimed at establishing a unique and comprehensive description for an individual to mesh with various personalized services and applications. An extensive research literature on or related to psychological modelling exists, i.e., into automatic personality computing. However, the integrity and accuracy of the results from current automatic personality computing is insufficient for the elaborate modeling in Cyber-I due to an insufficient number of data sources. To reach a comprehensive psychological description of a person, it is critical to bring in heterogeneous data sources that could provide plenty of personal data, i.e., the physiological data, and the Internet data. In addition, instead of calculating personality traits from personal data directly, an approach to a personality model derived from the theories of Carl Gustav Jung is used to measure a human subject’s persona. Therefore, this research is focused on designing an archetype-based modeling of persona covering an individual’s facets in different situations to approach a comprehensive personality model. Using personal big data to measure a specific persona in a certain scenario, our research is designed to ensure the accuracy and integrity of the generated personality model. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

Social Image Captioning: Exploring Visual Attention and User Attention

by

,

,

,

,

and

Sensors 2018, 18(2), 646; https://doi.org/10.3390/s18020646 - 22 Feb 2018

Cited by 11 | Viewed by 4085

Abstract

Image captioning with a natural language has been an emerging trend. However, the social image, associated with a set of user-contributed tags, has been rarely investigated for a similar task. The user-contributed tags, which could reflect the user attention, have been neglected in [...] Read more.

Image captioning with a natural language has been an emerging trend. However, the social image, associated with a set of user-contributed tags, has been rarely investigated for a similar task. The user-contributed tags, which could reflect the user attention, have been neglected in conventional image captioning. Most existing image captioning models cannot be applied directly to social image captioning. In this work, a dual attention model is proposed for social image captioning by combining the visual attention and user attention simultaneously.Visual attention is used to compress a large mount of salient visual information, while user attention is applied to adjust the description of the social images with user-contributed tags. Experiments conducted on the Microsoft (MS) COCO dataset demonstrate the superiority of the proposed method of dual attention. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Graphical abstract

Open AccessArticle

A Single RF Emitter-Based Indoor Navigation Method for Autonomous Service Robots

by

,

,

,

and

Sensors 2018, 18(2), 585; https://doi.org/10.3390/s18020585 - 14 Feb 2018

Cited by 5 | Viewed by 4803

Abstract

Location-aware services are one of the key elements of modern intelligent applications. Numerous real-world applications such as factory automation, indoor delivery, and even search and rescue scenarios require autonomous robots to have the ability to navigate in an unknown environment and reach mobile [...] Read more.

Location-aware services are one of the key elements of modern intelligent applications. Numerous real-world applications such as factory automation, indoor delivery, and even search and rescue scenarios require autonomous robots to have the ability to navigate in an unknown environment and reach mobile targets with minimal or no prior infrastructure deployment. This research investigates and proposes a novel approach of dynamic target localisation using a single RF emitter, which will be used as the basis of allowing autonomous robots to navigate towards and reach a target. Through the use of multiple directional antennae, Received Signal Strength (RSS) is compared to determine the most probable direction of the targeted emitter, which is combined with the distance estimates to improve the localisation performance. The accuracy of the position estimate is further improved using a particle filter to mitigate the fluctuating nature of real-time RSS data. Based on the direction information, a motion control algorithm is proposed, using Simultaneous Localisation and Mapping (SLAM) and A* path planning to enable navigation through unknown complex environments. A number of navigation scenarios were developed in the context of factory automation applications to demonstrate and evaluate the functionality and performance of the proposed system. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

Wearable Driver Distraction Identification On-The-Road via Continuous Decomposition of Galvanic Skin Responses

by

Omid Dehzangi

,

Vikas Rajendra

and

Mojtaba Taherisadr

Sensors 2018, 18(2), 503; https://doi.org/10.3390/s18020503 - 07 Feb 2018

Cited by 29 | Viewed by 5939

Abstract

One of the main reasons for fatal accidents on the road is distracted driving. The continuous attention of an individual driver is a necessity for the task of driving. While driving, certain levels of distraction can cause drivers to lose their attention, which [...] Read more.

One of the main reasons for fatal accidents on the road is distracted driving. The continuous attention of an individual driver is a necessity for the task of driving. While driving, certain levels of distraction can cause drivers to lose their attention, which might lead to an accident. Thus, the number of accidents can be reduced by early detection of distraction. Many studies have been conducted to automatically detect driver distraction. Although camera-based techniques have been successfully employed to characterize driver distraction, the risk of privacy violation is high. On the other hand, physiological signals have shown to be a privacy preserving and reliable indicator of driver state, while the acquisition technology might be intrusive to drivers in practical implementation. In this study, we investigate a continuous measure of phasic Galvanic Skin Responses (GSR) using a wristband wearable to identify distraction of drivers during a driving experiment on-the-road. We first decompose the raw GSR signal into its phasic and tonic components using Continuous Decomposition Analysis (CDA), and then the continuous phasic component containing relevant characteristics of the skin conductance signals is investigated for further analysis. We generated a high resolution spectro-temporal transformation of the GSR signals for non-distracted and distracted (calling and texting) scenarios to visualize the associated behavior of the decomposed phasic GSR signal in correlation with distracted scenarios. According to the spectrogram observations, we extract relevant spectral and temporal features to capture the patterns associated with the distracted scenarios at the physiological level. We then performed feature selection using support vector machine recursive feature elimination (SVM-RFE) in order to: (1) generate a rank of the distinguishing features among the subject population, and (2) create a reduced feature subset toward more efficient distraction identification on the edge at the generalization phase. We employed support vector machine (SVM) to generate the 10-fold cross validation (10-CV) identification performance measures. Our experimental results demonstrated cross-validation accuracy of 94.81% using all the features and the accuracy of 93.01% using reduced feature space. The SVM-RFE selected set of features generated a marginal decrease in accuracy while reducing the redundancy in the input feature space toward shorter response time necessary for early notification of distracted state of the driver. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

A Survey of Data Semantization in Internet of Things

by

,

,

and

Sensors 2018, 18(1), 313; https://doi.org/10.3390/s18010313 - 22 Jan 2018

Cited by 66 | Viewed by 7977

Abstract

With the development of Internet of Things (IoT), more and more sensors, actuators and mobile devices have been deployed into our daily lives. The result is that tremendous data are produced and it is urgent to dig out hidden information behind these volumous [...] Read more.

With the development of Internet of Things (IoT), more and more sensors, actuators and mobile devices have been deployed into our daily lives. The result is that tremendous data are produced and it is urgent to dig out hidden information behind these volumous data. However, IoT data generated by multi-modal sensors or devices show great differences in formats, domains and types, which poses challenges for machines to process and understand. Therefore, adding semantics to Internet of Things becomes an overwhelming tendency. This paper provides a systematic review of data semantization in IoT, including its backgrounds, processing flows, prevalent techniques, applications, existing challenges and open issues. It surveys development status of adding semantics to IoT data, mainly referring to sensor data and points out current issues and challenges that are worth further study. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

A More Efficient Transportable and Scalable System for Real-Time Activities and Exercises Recognition

by

,

,

,

,

and

Sensors 2018, 18(1), 268; https://doi.org/10.3390/s18010268 - 18 Jan 2018

Cited by 17 | Viewed by 4324

Abstract

Many people in the world are affected by muscle wasting, especially the population hits by myotonic dystrophy type 1 (DM1). Those people are usually given a program of multiple physical exercises to do. While DM1 and many other people have difficulties attending commercial [...] Read more.

Many people in the world are affected by muscle wasting, especially the population hits by myotonic dystrophy type 1 (DM1). Those people are usually given a program of multiple physical exercises to do. While DM1 and many other people have difficulties attending commercial centers to realize their program, a solution is to develop such a program completable at home. To this end, we developed a portable system that patients could bring home. This prototype is an improved version of the previous one using Wi-Fi, as this new prototype runs on BLE technology. This new prototype conceptualized induces great results. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

Application of Computational Intelligence to Improve Education in Smart Cities

by

Everton Gomede

,

Fernando Henrique Gaffo

,

Gabriel Ulian Briganó

,

Rodolfo Miranda De Barros

and Leonardo De Souza Mendes

Sensors 2018, 18(1), 267; https://doi.org/10.3390/s18010267 - 18 Jan 2018

Cited by 42 | Viewed by 5430

Abstract

According to UNESCO, education is a fundamental human right and every nation’s citizens should be granted universal access with equal quality to it. Because this goal is yet to be achieved in most countries, in particular in the developing and underdeveloped countries, it [...] Read more.

According to UNESCO, education is a fundamental human right and every nation’s citizens should be granted universal access with equal quality to it. Because this goal is yet to be achieved in most countries, in particular in the developing and underdeveloped countries, it is extremely important to find more effective ways to improve education. This paper presents a model based on the application of computational intelligence (data mining and data science) that leads to the development of the student’s knowledge profile and that can help educators in their decision making for best orienting their students. This model also tries to establish key performance indicators to monitor objectives’ achievement within individual strategic planning assembled for each student. The model uses random forest for classification and prediction, graph description for data structure visualization and recommendation systems to present relevant information to stakeholders. The results presented were built based on the real dataset obtained from a Brazilian private k-9 (elementary school). The obtained results include correlations among key data, a model to predict student performance and recommendations that were generated for the stakeholders. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Open AccessArticle

Using Markov Chains and Multi-Objective Optimization for Energy-Efficient Context Recognition

by

Vito Janko

and

Mitja Luštrek

Sensors 2018, 18(1), 80; https://doi.org/10.3390/s18010080 - 29 Dec 2017

Cited by 5 | Viewed by 3373

Abstract

The recognition of the user’s context with wearable sensing systems is a common problem in ubiquitous computing. However, the typically small battery of such systems often makes continuous recognition impractical. The strain on the battery can be reduced if the sensor setting is [...] Read more.

The recognition of the user’s context with wearable sensing systems is a common problem in ubiquitous computing. However, the typically small battery of such systems often makes continuous recognition impractical. The strain on the battery can be reduced if the sensor setting is adapted to each context. We propose a method that efficiently finds near-optimal sensor settings for each context. It uses Markov chains to simulate the behavior of the system in different configurations and the multi-objective genetic algorithm to find a set of good non-dominated configurations. The method was evaluated on three real-life datasets and found good trade-offs between the system’s energy expenditure and the system’s accuracy. One of the solutions, for example, consumed five-times less energy than the default one, while sacrificing only two percentage points of accuracy. Full article

(This article belongs to the Special Issue Sensing, Data Analysis and Platforms for Ubiquitous Intelligence)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Special Issue "Sensing, Data Analysis and Platforms for Ubiquitous Intelligence"

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (9 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI