COVID-Safe Spatial Occupancy Monitoring Using OFDM-Based Features and Passive WiFi Samples

2021 ◽  
Vol 12 (4) ◽  
pp. 1-24
Author(s):  
Junye Li ◽  
Aryan Sharma ◽  
Deepak Mishra ◽  
Gustavo Batista ◽  
Aruna Seneviratne
Keyword(s):  
Learning Algorithm ◽  
Low Cost ◽  
Support Vector ◽  
Svm Classifier ◽  
Accuracy Rate ◽  
Channel State ◽  
State Information ◽  
Social Distancing ◽  
Occupancy Monitoring ◽  
Safe Limit

During the COVID-19 pandemic, authorities have been asking for social distancing to prevent transmission of the virus. However, enforcing such distancing has been challenging in tight spaces such as elevators and unmonitored commercial settings such as offices. This article addresses this gap by proposing a low-cost and non-intrusive method for monitoring social distancing within a given space, using Channel State Information (CSI) from passive WiFi sensing. By exploiting the frequency selective behavior of CSI with a Support Vector Machine (SVM) classifier, we achieve an improvement in accuracy over existing crowd counting works. Our system counts the number of occupants with a 93% accuracy rate in an elevator setting and predicts whether the COVID-Safe limit is breached with a 97% accuracy rate. We also demonstrate the occupant counting capability of the system in a commercial office setting, achieving 97% accuracy. Our proposed occupancy monitoring outperforms existing methods by at least 7%. Overall, the proposed framework is inexpensive, requiring only one device that passively collects data and a lightweight supervised learning algorithm for prediction. Our lightweight model and accuracy improvements are necessary contributions for WiFi-based counting to be suitable for COVID-specific applications.

Using geometric constraints to improve performance of image classifiers for automatic segmentation of traffic signs

GEOMATICA ◽  
2021 ◽  
pp. 1-23
Author(s):  
Roholah Yazdan ◽  
Masood Varshosaz ◽  
Saied Pirasteh ◽  
Fabio Remondino
Keyword(s):  
Deep Learning ◽  
Learning Algorithm ◽  
Automatic Segmentation ◽  
Geometric Constraints ◽  
Support Vector ◽  
Svm Classifier ◽  
Data Set ◽  
Traffic Signs ◽  
Deep Learning Algorithm ◽  
Comparison Results

Automatic detection and recognition of traffic signs from images is an important topic in many applications. At first, we segmented the images using a classification algorithm to delineate the areas where the signs are more likely to be found. In this regard, shadows, objects having similar colours, and extreme illumination changes can significantly affect the segmentation results. We propose a new shape-based algorithm to improve the accuracy of the segmentation. The algorithm works by incorporating the sign geometry to filter out the wrong pixels from the classification results. We performed several tests to compare the performance of our algorithm against those obtained by popular techniques such as Support Vector Machine (SVM), K-Means, and K-Nearest Neighbours. In these tests, to overcome the unwanted illumination effects, the images are transformed into colour spaces Hue, Saturation, and Intensity, YUV, normalized red green blue, and Gaussian. Among the traditional techniques used in this study, the best results were obtained with SVM applied to the images transformed into the Gaussian colour space. The comparison results also suggested that by adding the geometric constraints proposed in this study, the quality of sign image segmentation is improved by 10%–25%. We also comparted the SVM classifier enhanced by incorporating the geometry of signs with a U-Shaped deep learning algorithm. Results suggested the performance of both techniques is very close. Perhaps the deep learning results could be improved if a more comprehensive data set is provided.

A Low-Cost Monocular Vision-Based Obstacle Avoidance Using SVM and Optical Flow

2018 ◽  
Vol 06 (04) ◽  
pp. 267-275
Author(s):  
Ajay Shankar ◽  
Mayank Vatsa ◽  
P. B. Sujit
Keyword(s):  
Optical Flow ◽  
Autonomous Navigation ◽  
Low Cost ◽  
Obstacle Detection ◽  
Monocular Vision ◽  
Support Vector ◽  
Svm Classifier ◽  
Range Finding ◽  
Image Patches ◽  
Computational Resources

Development of low-cost robots with the capability to detect and avoid obstacles along their path is essential for autonomous navigation. These robots have limited computational resources and payload capacity. Further, existing direct range-finding methods have the trade-off of complexity against range. In this paper, we propose a vision-based system for obstacle detection which is lightweight and useful for low-cost robots. Currently, monocular vision approaches used in the literature suffer from various environmental constraints such as texture and color. To mitigate these limitations, a novel algorithm is proposed, termed as Pyramid Histogram of Oriented Optical Flow ([Formula: see text]-HOOF), which distinctly captures motion vectors from local image patches and provides a robust descriptor capable of discriminating obstacles from nonobstacles. A support vector machine (SVM) classifier that uses [Formula: see text]-HOOF for real-time obstacle classification is utilized. To avoid obstacles, a behavior-based collision avoidance mechanism is designed that updates the probability of encountering an obstacle while navigating. The proposed approach depends only on the relative motion of the robot with respect to its surroundings, and therefore is suitable for both indoor and outdoor applications and has been validated through simulated and hardware experiments.

scholarly journals Wi-SL: Contactless Fine-Grained Gesture Recognition Uses Channel State Information

Sensors ◽  
2020 ◽  
Vol 20 (14) ◽  
pp. 4025
Author(s):  
Zhanjun Hao ◽  
Yu Duan ◽  
Xiaochao Dang ◽  
Yang Liu ◽  
Daiyang Zhang
Keyword(s):  
Sign Language ◽  
Gesture Recognition ◽  
Channel State Information ◽  
Computational Cost ◽  
Support Vector ◽  
Language Recognition ◽  
Sign Language Recognition ◽  
Channel State ◽  
State Information ◽  
Fine Grained

In recent years, with the development of wireless sensing technology and the widespread popularity of WiFi devices, human perception based on WiFi has become possible, and gesture recognition has become an active topic in the field of human-computer interaction. As a kind of gesture, sign language is widely used in life. The establishment of an effective sign language recognition system can help people with aphasia and hearing impairment to better interact with the computer and facilitate their daily life. For this reason, this paper proposes a contactless fine-grained gesture recognition method using Channel State Information (CSI), namely Wi-SL. This method uses a commercial WiFi device to establish the correlation mapping between the amplitude and phase difference information of the subcarrier level in the wireless signal and the sign language action, without requiring the user to wear any device. We combine an efficient denoising method to filter environmental interference with an effective selection of optimal subcarriers to reduce the computational cost of the system. We also use K-means combined with a Bagging algorithm to optimize the Support Vector Machine (SVM) classification (KSB) model to enhance the classification of sign language action data. We implemented the algorithms and evaluated them for three different scenarios. The experimental results show that the average accuracy of Wi-SL gesture recognition can reach 95.8%, which realizes device-free, non-invasive, high-precision sign language gesture recognition.

scholarly journals A Novel Algorithm for Feature Level Fusion Using SVM Classifier for Multibiometrics-Based Person Identification

2013 ◽  
Vol 2013 ◽  
pp. 1-11 ◽  
Author(s):  
Ujwalla Gawande ◽  
Mukesh Zaveri ◽  
Avichal Kapur
Keyword(s):  
Learning Algorithm ◽  
Recognition Performance ◽  
Recognition Rate ◽  
Rejection Rate ◽  
Support Vector ◽  
Svm Classifier ◽  
Fusion Algorithm ◽  
Biometric Systems ◽  
Feature Level Fusion ◽  
Level Fusion

Recent times witnessed many advancements in the field of biometric and ultimodal biometric fields. This is typically observed in the area, of security, privacy, and forensics. Even for the best of unimodal biometric systems, it is often not possible to achieve a higher recognition rate. Multimodal biometric systems overcome various limitations of unimodal biometric systems, such as nonuniversality, lower false acceptance, and higher genuine acceptance rates. More reliable recognition performance is achievable as multiple pieces of evidence of the same identity are available. The work presented in this paper is focused on multimodal biometric system using fingerprint and iris. Distinct textual features of the iris and fingerprint are extracted using the Haar wavelet-based technique. A novel feature level fusion algorithm is developed to combine these unimodal features using the Mahalanobis distance technique. A support-vector-machine-based learning algorithm is used to train the system using the feature extracted. The performance of the proposed algorithms is validated and compared with other algorithms using the CASIA iris database and real fingerprint database. From the simulation results, it is evident that our algorithm has higher recognition rate and very less false rejection rate compared to existing approaches.

scholarly journals Smart Face Detection and Recognition in Low Resolution Images Using Alexnet CNN Compare Accuracy with SVM

2021 ◽  
Vol 36 (1) ◽  
pp. 721-726
Author(s):  
S. Mahesh ◽  
Dr.G. Ramkumar
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Recognition Rate ◽  
Vital Role ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Svm Classifier ◽  
Low Resolution ◽  
Accuracy Ratio ◽  
Low Resolution Images

Aim: Machine learning algorithm plays a vital role in various biometric applications due to its admirable result in detection, recognition and classification. The main objective of this work is to perform comparative analysis on two different machine learning algorithms to recognize the person from low resolution images with high accuracy. Materials & Methods: AlexNet Convolutional Neural Network (ACNN) and Support Vector Machine (SVM) classifiers are implemented to recognize the face in a low resolution image dataset with 20 samples each. Results: Simulation result shows that ACNN achieves a significant recognition rate with 98% accuracy over SVM (89%). Attained significant accuracy ratio (p=0.002) in SPSS statistical analysis as well. Conclusion: For the considered low resolution images ACNN classifier provides better accuracy than SVM Classifier.

scholarly journals Wi-Alarm: Low-Cost Passive Intrusion Detection Using WiFi

Sensors ◽  
2019 ◽  
Vol 19 (10) ◽  
pp. 2335 ◽  
Author(s):  
Tao Wang ◽  
Dandan Yang ◽  
Shunqing Zhang ◽  
Yating Wu ◽  
Shugong Xu
Keyword(s):  
Intrusion Detection ◽  
Detection System ◽  
Low Cost ◽  
Influence Factors ◽  
Human Motion ◽  
Support Vector ◽  
Channel State ◽  
Svm Algorithm ◽  
Mean And Variance ◽  
The Mean

In this paper, we present a WiFi-based intrusion detection system called Wi-Alarm. Motivated by our observations and analysis that raw channel state information (CSI) of WiFi is sensitive enough to monitor human motion, Wi-Alarm omits data preprocessing. The mean and variance of the amplitudes of raw CSI data are used for feature extraction. Then, a support vector machine (SVM) algorithm is applied to determine detection results. We prototype Wi-Alarm on commercial WiFi devices and evaluate it in a typical indoor scenario. Results show that Wi-Alarm reduces much computational expense without losing accuracy and robustness. Moreover, different influence factors are also discussed in this paper.

scholarly journals Endpoints-Clipping CSI Amplitude for SVM-Based Indoor Localization

Sensors ◽  
2019 ◽  
Vol 19 (17) ◽  
pp. 3689 ◽  
Author(s):  
Zhanjun Hao ◽  
Yan Yan ◽  
Xiaochao Dang ◽  
Chenguang Shao
Keyword(s):  
Indoor Localization ◽  
Spatial Clustering ◽  
Support Vector ◽  
Localization Algorithm ◽  
Channel State ◽  
Positioning Accuracy ◽  
State Information ◽  
Inaccurate Data ◽  
Svm Algorithm ◽  
Fingerprint Database

With the wide application of Channel State Information (CSI) in the field of sensing, the accuracy of positioning accuracy of indoor fingerprint positioning is increasingly necessary. The flexibility of the CSI signals may lead to an increase in fingerprint noise and inaccurate data classification. This paper presents an indoor localization algorithm based on Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Endpoints-Clipping (EC) CSI amplitude, and Support Vector Machine (EC-SVM). In the offline phase, the CSI amplitude information collected through the three channels is combined and clipped using the EC, and then a fingerprint database is obtained. In the online phase, the SVM is used to train the data in the fingerprint database, and the corresponding relationship is found with the CSI data collected in real time to perform matching and positioning. The experimental results show that the positioning accuracy of the EC-SVM algorithm is superior to the state-of-art indoor CSI-based localization technique.

A Sensor Fusion Based Pan-Tilt Platform for Activity Tracking and Fall Detection

2017 ◽  
Author(s):  
Zhangjie Chen ◽  
Hanwei Liu ◽  
Yuqiao Wang ◽  
Ya Wang
Keyword(s):  
Sensor Fusion ◽  
Degrees Of Freedom ◽  
Learning Algorithm ◽  
Low Cost ◽  
Fall Detection ◽  
Support Vector ◽  
Fusion Algorithm ◽  
Activity Tracking ◽  
Term Care ◽  
Sensor Module

This paper presents a pan-tilt sensor fusion platform for activity tracking and fall-detection which can work as a reliable surveillance system with long-term care function. A low cost thermal array sensor and a distance sensor are integrated together as the sensor module. The sensor module is installed on a pan-tilt orienting mechanism with two rotation degrees of freedom to increase the field of view while reducing the number of sensors used on-board. The performance of the sensor test platform is analyzed. The location of the indoor object as well as its size can be estimated based on a novel sensor fusion algorithm. The support vector machine (SVM) based machine learning algorithm is applied for fall detection. The preliminary experiment result shows a 95% accuracy to identify falling action from similar normal indoor activity such as sitting and picking up stuff.

Sign in / Sign up

Export Citation Format

Share Document