Recognition Scheme Research Articles

A LIGHTWEIGHT MULTI-PERSON POSE ESTIMATION SCHEME BASED ON JETSON NANO

As the basic technology of human action recognition, pose estimation is attracting more and more researchers' attention, while edge application scenarios pose a higher challenge. This paper proposes a lightweight multi-person pose estimation scheme to meet the needs of real-time human action recognition on the edge end. This scheme uses AlphaPose to extract human skeleton nodes, and adds ResNet and Dense Upsampling Revolution to improve its accuracy. Meanwhile, we use YOLO to enhance AlphaPose’s support for multi-person pose estimation, and optimize the proposed model with TensorRT. In addition, this paper sets Jetson Nano as the Edge AI deployment device of the proposed model and successfully realizes the model migration to the edge end. The experimental results show that the speed of the optimized object detection model can reach 20 FPS, and the optimized multi-person pose estimation model can reach 10 FPS. With the image resolution of 320×240, the model’s accuracy is 73.2%, which can meet the real-time requirements. In short, our scheme can provide a basis for lightweight multi-person action recognition scheme on the edge end.

FedLVR: a federated learning-based fine-grained vehicle recognition scheme in intelligent traffic system

FedLVR: a federated learning-based fine-grained vehicle recognition scheme in intelligent traffic system

Computer-Aided Diagnosis of COVID-19 from Chest X-ray Images Using Hybrid-Features and Random Forest Classifier

In recent years, a lot of attention has been paid to using radiology imaging to automatically find COVID-19. (1) Background: There are now a number of computer-aided diagnostic schemes that help radiologists and doctors perform diagnostic COVID-19 tests quickly, accurately, and consistently. (2) Methods: Using chest X-ray images, this study proposed a cutting-edge scheme for the automatic recognition of COVID-19 and pneumonia. First, a pre-processing method based on a Gaussian filter and logarithmic operator is applied to input chest X-ray (CXR) images to improve the poor-quality images by enhancing the contrast, reducing the noise, and smoothing the image. Second, robust features are extracted from each enhanced chest X-ray image using a Convolutional Neural Network (CNNs) transformer and an optimal collection of grey-level co-occurrence matrices (GLCM) that contain features such as contrast, correlation, entropy, and energy. Finally, based on extracted features from input images, a random forest machine learning classifier is used to classify images into three classes, such as COVID-19, pneumonia, or normal. The predicted output from the model is combined with Gradient-weighted Class Activation Mapping (Grad-CAM) visualisation for diagnosis. (3) Results: Our work is evaluated using public datasets with three different train-test splits (70-30%, 80-20%, and 90-10%) and achieved an average accuracy, F1 score, recall, and precision of 97%, 96%, 96%, and 96%, respectively. A comparative study shows that our proposed method outperforms existing and similar work. The proposed approach can be utilised to screen COVID-19-infected patients effectively. (4) Conclusions: A comparative study with the existing methods is also performed. For performance evaluation, metrics such as accuracy, sensitivity, and F1-measure are calculated. The performance of the proposed method is better than that of the existing methodologies, and it can thus be used for the effective diagnosis of the disease.

An Open-Set Modulation Recognition Scheme With Deep Representation Learning

This letter proposes a deep representation learning based automatic modulation recognition (AMR) algorithm in the open-set recognition (OSR) regime. The challenging recognition risk of unknown modulation classes is first analyzed for most state-of-the-art approaches, and interesting insights into this problem is then provided. Based on this, an open-set AMR scheme is proposed with a combination of feature representation and classification, where a triplet loss function from metric learning is employed for the representor to form distinct clusters for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> known modulation classes. Then, the degree of membership is calculated via extreme value theory (EVT) by modeling the distance between known training data to its corresponding clustering center, followed by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> binary classifiers. Comprehensive experiments on public dataset confirm that the proposed scheme outperforms the other state-of-the-arts in terms of both balanced accuracy and openness.

HAND RECOGNITION SYSTEM BASED ON ELECTROMYOGRAPHY FOR REAL-TIME HAND PROSTHETIC CONTROL

Upper limb amputation is a significant limitation for achieving routine activities. Myoelectric signals detected by electrodes well-known as Electromyography (EMG) have been targeted to control upper limb prostheses of such lost limbs. Unfortunately, the acquisition, processing and use of such myoelectric signals are sophisticated. Furthermore, it necessarily requires complex computation to fulfil accuracy, robustness, and time-consumption execution for the real-time prosthesis application. Thus, machine learning schemes for pattern recognition are a potential approach to improve the traditional control for hand prostheses due to the movement of users and muscle contraction. This paper presents real-time hand posture recognition based on three hand postures using surface EMG (sEMG) signals. sEMG signals are acquired by the electrode channel and simultaneously collected while making a hand posture. Performance evaluation relies on classification accuracy and time consumption. The performance of six real-time recognition models is evaluated which combine two projection techniques and three classifiers. Results indicate that EMG-based pattern recognition (EMG-PR) control outperforms the traditional control for hand prostheses in real-time application. The highest classification accuracy is approximately 96%, whereas the lowest time consumption is 4 ms. In addition, the accuracy is dropped when the number of electrodes decreases nearly to 3%. These outcomes can apply to real-time hand prostheses to alleviate the limited prostheses available.

UWB Localization Based on Improved Robust Adaptive Cubature Kalman Filter.

Aiming at the problems of Non-Line-of-Sight (NLOS) observation errors and inaccurate kinematic model in ultra-wideband (UWB) systems, this paper proposed an improved robust adaptive cubature Kalman filter (IRACKF). Robust and adaptive filtering can weaken the influence of observed outliers and kinematic model errors on filtering, respectively. However, their application conditions are different, and improper use may reduce positioning accuracy. Therefore, this paper designed a sliding window recognition scheme based on polynomial fitting, which can process the observation data in real-time to identify error types. Simulation and experimental results indicate that compared to the robust CKF, adaptive CKF, and robust adaptive CKF, the IRACKF algorithm reduces the position error by 38.0%, 45.1%, and 25.3%, respectively. The proposed IRACKF algorithm significantly improves the positioning accuracy and stability of the UWB system.

A survey of human-computer interaction (HCI) & natural habits-based behavioural biometric modalities for user recognition schemes

The proliferation of Internet of Things (IoT) systems is having a profound impact across all aspects of life. Recognising and identifying particular users is central to delivering the personalised experience that citizens want to experience, and that organisations wish to deliver. This article presents a survey of human-computer interaction-based (HCI-based) and natural habits-based behavioural biometrics that can be acquired unobtrusively through smart devices or IoT sensors for user recognition purposes. Robust and usable user recognition is also a security requirement for emerging IoT ecosystems to protect them from adversaries. Typically, it can be specified as a fundamental building block for most types of human-to-things accountability principles and access-control methods. However, end-users are facing numerous security and usability challenges in using currently available knowledge- and token-based recognition (i.e., authentication and identification) schemes. To address the limitations of conventional recognition schemes, biometrics, naturally come as a first choice to supporting sophisticated user recognition solutions. We perform a comprehensive review of touch-stroke, swipe, touch signature, hand-movements, voice, gait and footstep behavioural biometrics modalities. This survey analyzes the recent state-of-the-art research of these behavioural biometrics with a goal to identify their attributes and features for generating unique identification signatures. Finally, we present security, privacy, and usability evaluations that can strengthen the designing of robust and usable user recognition schemes for IoT applications.

A discrete side-lobe clutter recognition method based on sliding filter response loss for space-based radar

Different from ground-based or airborne early warning radar, space-based radar (SBR) possesses large coverage capability. As a result, several discrete strong scatter points from the antenna side-lobe shares the same feature with the real targets in range-Doppler domain, which leads to false alarms when conducting constant false alarm rate (CFAR) detection process, and the detection performance with regard to SBR deteriorates seriously. In this paper, a discrete side-lobe clutter recognition method based on sliding filter response loss is proposed for space-based radar. Firstly, considering both the echo inhomogeneity and the limited degrees of freedom (DOFs) after dimension-reduced space-time adaptive processing (STAP), the sliding window design strategy is employed to segment range cells for the observation scene. Then, the images related to different range segments are registered after clutter suppression, in this way, the candidate target parameters, including the position information and the amplitude information are counted. On this basis, the reliable recognition scheme between the real target and the discrete side-lobe clutter can be realized by comparing these filter response losses. Compared with recent works, experimental results based on real measured data show that the proposed method significantly improves the fault-tolerant discrimination ability, which possesses high robustness in algorithm performance as well as good prospect in engineering application.

An N400 identification method based on the combination of Soft-DTW and transformer.

As a time-domain EEG feature reflecting the semantic processing of the human brain, the N400 event-related potentials still lack a mature classification and recognition scheme. To address the problems of low signal-to-noise ratio and difficult feature extraction of N400 data, we propose a Soft-DTW-based single-subject short-distance event-related potential averaging method by using the advantages of differentiable and efficient Soft-DTW loss function, and perform partial Soft-DTW averaging based on DTW distance within a single-subject range, and propose a Transformer-based ERP recognition classification model, which captures contextual information by introducing location coding and a self-attentive mechanism, combined with a Softmax classifier to classify N400 data. The experimental results show that the highest recognition accuracy of 0.8992 is achieved on the ERP-CORE N400 public dataset, verifying the effectiveness of the model and the averaging method.

A Multi-Dimensional Covert Transaction Recognition Scheme for Blockchain

Covert communication was widely studied in recent years in terms of keeping the communication of entities on the Internet secret from the point of view of information security. Due to the anonymity of accounts and the publicness of the ledger, blockchain is a natural and ideal channel for helping users establish covert communication channels. Senders can embed secret messages into certain fields in transactions, and receivers can extract those messages from the transactions without attracting the attention of other users. However, to the best of our knowledge, most existing works have aimed at designing blockchain-based covert communication schemes. Few studies concentrated on the recognition of transactions used for covert communication. In this paper, we first analyze convolutional neural network (CNN)-based and attention-based covert transaction recognition schemes, and we explore the deep relationship between the appropriate extraction of features and the embedded fields of covert transactions. We further propose a multi-dimensional covert transaction recognition (M-CTR) scheme. It can simultaneously support both one-dimensional and two-dimensional feature extraction to recognize covert transactions. The experimental results show that the precision and recall of the M-CTR in recognizing covert transactions outperformed those of existing covert communication schemes.

Skeleton-Based Fall Detection with Multiple Inertial Sensors Using Spatial-Temporal Graph Convolutional Networks.

The application of wearable devices for fall detection has been the focus of much research over the past few years. One of the most common problems in established fall detection systems is the large number of false positives in the recognition schemes. In this paper, to make full use of the dependence between human joints and improve the accuracy and reliability of fall detection, a fall-recognition method based on the skeleton and spatial-temporal graph convolutional networks (ST-GCN) was proposed, using the human motion data of body joints acquired by inertial measurement units (IMUs). Firstly, the motion data of five inertial sensors were extracted from the UP-Fall dataset and a human skeleton model for fall detection was established through the natural connection relationship of body joints; after that, the ST-GCN-based fall-detection model was established to extract the motion features of human falls and the activities of daily living (ADLs) at the spatial and temporal scales for fall detection; then, the influence of two hyperparameters and window size on the algorithm performance was discussed; finally, the recognition results of ST-GCN were also compared with those of MLP, CNN, RNN, LSTM, TCN, TST, and MiniRocket. The experimental results showed that the ST-GCN fall-detection model outperformed the other seven algorithms in terms of accuracy, precision, recall, and F1-score. This study provides a new method for IMU-based fall detection, which has the reference significance for improving the accuracy and robustness of fall detection.

DRMT: A Decentralized IoT Device Recognition and Management Technology in Smart Cities

The Internet of Things (IoT) faces many security and privacy risks due to the large number of devices accessing the network while providing convenience to the current society. At present, the management of IoT devices is centralized. All terminal devices are managed in a centralized manner, and the data accessing IoT devices can only be connected to the network with the permission of the centralized nodes. In this scenario, the increase in the amount of equipment can cause central node overload or single node failure. Moreover, flexible expansion requirements are difficult to achieve in such a centralized architecture. This paper proposes a safe and efficient IoT device recognition and management scheme based on blockchain and edge computing, DRMT. This scheme realizes feature extraction, clustering, and learning of IoT devices’ traffic based on the characteristics of distributed deployment of edge servers. In this process, the recognition model can be continuously updated. The scheme also has an advantage that it is possible to enhance the management efficiency and scalability of the system by establishing a blockchain network between edge servers to share identification information and automate deployment security. This paper also verifies the scheme’s effectiveness by building an experimental network and a blockchain system. The results showed that DRMT would help realize the effective management of large-scale IoT devices in smart cities.

Fabrication and characterization of highly sensitive flexible strain sensor based on biodegradable gelatin nanocomposites and double strain layered structures with crack for gesture recognition

Flexible sensors have attracted extensive attention in the field of human-computer interaction. However, it is still a challenging task to realize accuracy gesture recognition with flexible sensor, which requires sensor not only have high sensitivity, but also have appropriate strain detection range. Here, a high gauge factor flexible sensor (gauge factor ∼ 1296 under 12–20 % strain) based on crack structure is reported. The sensor is made of a biodegradable and stretchable gelatin composite combined with fabric bases, with good repeatability (6000 cycles) and a fast response (60 ms). Because of the double-layer structure, it has a suitable detection range (20 % strain). The sensor is manufactured by a screen-printing process, and it has been used to make data gloves and has realized 9 gestures recognition with machine learning algorithm (99.6 % accuracy). In general, this study offers a wearable gestures recognition scheme through the proposed sensor.

Facial Image Encryption for Secure Face Recognition System

A biometric authentication system is more convenient and secure than graphical or textual passwords when accessing information systems. Unfortunately, biometric authentication systems have the disadvantage of being susceptible to spoofing attacks. Authentication schemes based on biometrics, including face recognition, are susceptible to spoofing. This paper proposes an image encryption scheme to counter spoofing attacks by integrating it into the pipeline of Linear Discriminant Analysis (LDA) based face recognition. The encryption scheme uses XOR pixels substitution and cellular automata for scrambling. A single key is used to encrypt the training and testing datasets in LDA face recognition system. For added security, the encryption step requires input images of faces to be encrypted with the correct key before the system can recognize the images. An LDA face recognition scheme based on random forest classifiers has achieved 96.25% accuracy on ORL dataset in classifying encrypted test face images. In a test where original test face images were not encrypted with keys used for encrypted feature databases, the system achieved 8.75% accuracy only showing it is capable of resisting spoofing attacks.

Deterministic size discovery and topology recognition in radio networks with short labels

We consider size discovery and topology recognition in radio networks without collision detection, modeled by simple undirected connected graphs. The aim of size discovery is that all nodes output the number of nodes in the graph, and in the task of topology recognition, each node has to learn the topology of the graph and its position in it. Nodes are assigned labels which are (not necessarily different) binary strings. Each node uses its own label as input when executing the algorithm. The length of a labeling scheme is the largest length of a label. Our goal is to construct short labeling schemes for size discovery and topology recognition in arbitrary radio networks, and to design efficient deterministic distributed algorithms for each of these tasks, using these short schemes. In both cases, the length of our labeling schemes is asymptotically optimal.

Carbon footprints of travel to World Heritage Sites: communicating climate to potential tourists through a consumption-based life-cycle assessment

ABSTRACT In this study, an input-output-based hybrid life cycle assessment method is used to calculate the carbon footprint of a tourist’s travel to sixteen United Nations Educational, Scientific and Cultural Organization (UNESCO) World Heritage Sites (WHS) including direct and indirect emissions. The carbon footprint of three trip styles (budget, mid-range expenses, and high-end) is calculated for each WHS, reflecting options available to heritage tourists. This study is part of a more extensive study that created a climate communication recognition scheme (CCRS) for UNESCO WHS to communicate information about climate change to potential tourists and heritage site managers, elaborating and extending upon communication tools such as environmental product information schemes, certifications, and ecolabels. The results are intended for educational purposes and contribute to a holistic approach to carbon management in heritage tourism. The results are available online as an ArcGIS StoryMap titled Climate Footprints of Heritage Tourism.

Optimized Deep CNN with Deviation Relevance-based LBP for Skin Cancer Detection: Hybrid Metaheuristic Enabled Feature Selection

Segmentation of skin lesions is a significant and demanding task in dermoscopy images. This paper proposes a new skin cancer recognition scheme, with: “Pre-processing, Segmentation, Feature extraction, Optimal Feature Selection and Classification”. Here, pre-processing is done with certain processes. The pre-processed images are segmented via the “Otsu Thresholding model”. The third phase is feature extraction, where Deviation Relevance-based “Local Binary Pattern (DRLBP), Gray-Level Co-Occurrence Matrix (GLCM) features and Gray Level Run-Length Matrix (GLRM) features” are extracted. From these extracted features, the optimal features are chosen via Particle Updated WOA (PU-WOA) model. Subsequently, classification occurs via Optimized DCNN and NN to classify the skin lesion. To make the classification more precise, the DCNN is optimized by the introduced algorithm. The result has shown a higher accuracy of 0.998737, when compared with other extant models like IPSO, IWOA, PSO+CNN, WOA+CNN and CNN schemes.

Optimal Ensemble Scheme for Human Activity Recognition and Floor Detection Based on AutoML and Weighted Soft Voting Using Smartphone Sensors

In this article, human activity recognition (HAR) attempts to recognize the activities of an object in a multistory building from data retrieved via smartphone-based sensors (SBSs). Most publications based on machine learning (ML) report the development of a suitable architecture to improve classification accuracy by increasing the parameters of the architecture pertaining to HAR. Due to robust and automated ML, it is quite possible to develop a versatile approach to improve the accuracy of HAR. This research proposes an optimal ensemble HAR and floor detection (OEC-HAFD) scheme based on automated learning (AutoML) and weighted soft voting (WSF) using SBS to improve the recognition rate. The proposed HAR-FD scheme is developed based on twofold mechanisms. First, an AutoML paradigm is employed to find optimal supervised models based on performance. Second, top-ranked (optimal) models are combined using the WSF mechanism to classify HAR on various floors of a multistory building. The proposed scheme is developed based on real-time SBS: accelerometer and barometer data. The accelerometer data are used to detect activity by observing the magnitude of the sensor measurements. Similarly, the barometer sensor detects floor height by using pressure and altitude data. Furthermore, we analyze the performance of the proposed optimal ensemble classifier (OEC) and compare them with the state-of-the-art classifiers. Based on the result analysis, it is clearly seen that the proposed OEC outperforms the performance of the individual classifiers. Moreover, our proposed HAR-FD can be leveraged as a robust solution to accurately recognize human activities in multistory buildings compared to the existing standalone models.

Internet of Things Centric-Based Multiactivity Recognition in Smart Home Environment

In recent times, numerous human activity recognition (HAR) schemes have been proposed with embedding sensors, wearable devices, smart phones, and vision and ambient sensors. Though the systems have shown better performance they are mostly standalone and still lack the ability to share, host, and perform real-time analysis and visualization of activity data. The Internet of Things (IoT) paradigm has a solution to render the limitations and this will pave the way for HAR in the smart home environment. Thus in this article, an IoT-centric multiactivity recognition system is proposed and deployed on the cloud platform for activity data tracking in the smart home environment. The proposed system collects the real-time data collected using IMU sensors and transmitted to the IoT-Edge Server via Wi-Fi where the data has been fused and classified using light-weight deep learning models. This system has a provision of a Web-based dashboard which is helpful for the home dwellers to monitor the activities in the remote. The performance evaluation justified that the developed system can measure IoT-based activity recognition with greater efficiency in terms of accuracy and F1-score in a shorter response time as of deployment in the cloud platform to detect the activity.

Adaptive federated learning scheme for recognition of malicious attacks in an IoT network

Adaptive federated learning scheme for recognition of malicious attacks in an IoT network