Search

This article proposes a simple sparse imaging scheme using a linear sparse aperiodic array and a new fast Fourier transform matched filtering (FFTMF) algorithm for a THz multistatic multiple-input and multiple-output (MIMO) imaging system. The simple linear sparse aperiodic array and multipass interferometric synthetic aperture focusing technique are used to achieve fast sampling, low system cost, and high imaging performance. Unlike a traditional generalized synthetic aperture focusing technique (GSAFT) for multistatic MIMO imaging, which is time-consuming and exhibits increased reconstruction time with increased data volume, the proposed FFTMF image reconstruction algorithm is capable of providing comparable image quality but significantly reducing the reconstruction time. For example, we show that for an image of 300 × 320 mm with a pixel size of 0.75 × 0.83 mm, the reconstruction time is reduced from about 1.50 min to 0.25 s in the 220-GHz five-pass synthetic imaging experiments. The proposed imaging algorithm uses internal zero padding, a multipass interferometric synthetic aperture focusing technique, and a wideband imaging technique to improve the imaging performance under a low-cost, sparse sampling scheme. It shows a strong antinoise ability and a high tolerance to target focusing distance. In addition, integrated with an algorithm of principal component pursuit by alternating directions method (PCPADM), sparse imaging is available to further save system cost and sampling data without a loss of image quality while the novel use of an error matrix provides an additional detection capability for imaging systems.

© 2021 The Authors. Many biomedical applications require accurate non-rigid image registration that can cope with complex deformations. However, popular diffeomorphic Demons registration algorithms suffer from difficulties for complex and serious distortions since they only use image greyscale and gradient information. To address these difficulties, a new diffeomorphic Demons registration algorithm is proposed using hierarchical neighbourhood spectral features namely HNSF Log-Demons in this paper. In view of three important properties of hierarchical neighbourhood spectral features based on line graph such as rotation invariance, invariance of linear changes of brightness, and robustness to noise, the hierarchical neighbourhood spectral features of a reference image and a moving image is first extracted and these novel spectral features are incorporated into the energy function of the diffeomorphic registration framework to improve the capability of capturing complex distortions. Secondly, the Nystr ö o ̈ m approximation based on random singular value decomposition is employed to effectively enhance the computational efficiency of HNSF Log-Demons. Finally, the hybrid multi-resolution strategy based on wavelet decomposition in the registration process is utilised to further improve the registration accuracy and efficiency. Experimental results show that the proposed HNSF Log-Demons not only effectively ensures the generation of smooth and reversible deformation field, but also achieves better performance than state-of-the-art algorithms. National Natural Science Foundation of China. Grant Numbers: 61762058, 61861024, 61871259; Natural Science Foundation of Gansu Province of China. Grant Number: 20JR5RA404; Natural Science Basic Research Program of Shaanxi. Grant Number: 2021JC-47.

Permanent pacemaker (PPM) implantation occurs in up to 5 % of patients after cardiac surgery but there is little consensus on how long to wait between surgery and PPM insertion. Predicting the likelihood of a patient being pacing dependent 30 days after implant can aid with this timing decision and avoid unnecessary observation time waiting for intrinsic conduction to recover. In this paper, we introduce a new approach for the prediction of PPM dependency at 30 days after implant in patients who have undergone recent cardiac surgery. The aim is to create an automatic detection model able to support clinicians in the decision-making process. We first applied Synthetic Minority Oversampling Technique (SMOTE) and Bayesian Networks (BN) to the dataset, to balance the inherently imbalanced data and create additional synthetic data respectively. The six resultant datasets were then used to train four different classifiers to predict pacing dependence at 30 days, all using the same testing set. The Bagged Trees classifier achieved the best results, reaching an area under the receiver operating curve (AUC) of 90 % in the train phase, and 83 % in the test phase. The overall classification performance was clearly enhanced when using SMOTE and synthetic data created with BN to create a combined and balanced dataset. This technique could be of great use in answering clinical questions where the original dataset is imbalanced.

Solder paste printing is the first part of the surface mount process flow; its postprinting defect inspection is particularly important. In this paper, we focus on studying the printing defects inspection algorithm for solder paste on PCB (Printed Circuit Board) images. +e work proposes a number of methods to enhance the defects inspection performance of solder paste printing: a regional multidirectional data fusion image interpolation method, which can achieve fast and high precision image interpolation; a method for detecting solder paste areas with better accuracy, efficiency, and robustness; an improved connected domain labeling method to reduce time complexity; and defects detection and types classification method, which extracts features and centroid of every solder paste region and completes the inspection by comparing with a standard image. +e experiments show that the defects inspection algorithm can detect the most common types of defects with low time consumption, high inspection accuracy, and classification accuracy. Key Project of Shaanxi Province Innovation Program, China (2017ZDCXL-GY-11-02-02)

10.13039/501100001809-National Natural Science Foundation of China (Grant Number: 62076033 and U1931202).

Key Project of Shaanxi Province Industrial Innovation Program, China (Grant No. 2017ZDCXL-GY-11-02-02).

Recently, Multi-Object Tracking (MOT) has attracted rising attention, and accordingly, remarkable progresses have been achieved. However, the existing methods tend to use various basic models (e.g, detector and embedding model), and different training or inference tricks, etc. As a result, the construction of a good baseline for a fair comparison is essential. In this paper, a classic tracker, i.e., DeepSORT, is first revisited, and then is significantly improved from multiple perspectives such as object detection, feature embedding, and trajectory association. The proposed tracker, named StrongSORT, contributes a strong and fair baseline for the MOT community. Moreover, two lightweight and plug-and-play algorithms are proposed to address two inherent "missing" problems of MOT: missing association and missing detection. Specifically, unlike most methods, which associate short tracklets into complete trajectories at high computation complexity, we propose an appearance-free link model (AFLink) to perform global association without appearance information, and achieve a good balance between speed and accuracy. Furthermore, we propose a Gaussian-smoothed interpolation (GSI) based on Gaussian process regression to relieve the missing detection. AFLink and GSI can be easily plugged into various trackers with a negligible extra computational cost (1.7 ms and 7.1 ms per image, respectively, on MOT17). Finally, by fusing StrongSORT with AFLink and GSI, the final tracker (StrongSORT++) achieves state-of-the-art results on multiple public benchmarks, i.e., MOT17, MOT20, DanceTrack and KITTI. Codes are available at https://github.com/dyhBUPT/StrongSORT and https://github.com/open-mmlab/mmtracking., Accepted by IEEE Transactions on Multimedia 2023

Copyright: © 2021 Pereira, Meng and Hone. It is well recognised that social signals play an important role in communication effectiveness. Observation of videos to understand non-verbal behaviour is time-consuming and limits the potential to incorporate detailed and accurate feedback of this behaviour in practical applications such as communication skills training or performance evaluation. The aim of the current research is twofold: (1) to investigate whether off-the-shelf emotion recognition technology can detect social signals in media interviews and (2) to identify which combinations of social signals are most promising for evaluating trainees’ performance in a media interview. To investigate this, non-verbal signals were automatically recognised from practice on-camera media interviews conducted within a media training setting with a sample size of 34. Automated non-verbal signal detection consists of multimodal features including facial expression, hand gestures, vocal behaviour and ‘honest’ signals. The on-camera interviews were categorised into effective and poor communication exemplars based on communication skills ratings provided by trainers and neutral observers which served as a ground truth. A correlation-based feature selection method was used to select signals associated with performance. To assess the accuracy of the selected features, a number of machine learning classification techniques were used. Naive Bayes analysis produced the best results with an F-measure of 0.76 and prediction accuracy of 78%. Results revealed that a combination of body movements, hand movements and facial expression are relevant for establishing communication effectiveness in the context of media interviews. The results of the current study have implications for the automatic evaluation of media interviews with a number of potential application areas including enhancing communication training including current media skills training. UK Defence Science and Technology Laboratory; Engineering and Physical Sciences Research Council (DTP grant).

An autonomous vehicle navigates by perceiving the environment through the sensors and acting on the received data by making sense of the surroundings. In this paper, innovative holoscopic 3D scene classification based on the single aperture holoscopic 3D camera is proposed to recognise continuous 3D scene of environment. The deep learning network AlexNet is used to evaluate the proposed approach, and the outcome exhibits promising results compared to 2D images.

Simultaneous Localization and Mapping (SLAM) is a critical system for Unmanned Aerial Vehicles (UAV), Autonomous Navigation (AN), Augmented Reality (AR) and Virtual Reality (VR). However, due to volume and computation limitations, the current SLAM systems on mobile devices are not robust enough for practical applications. In this paper, a new SLAM system is proposed, which can be applied to AR or VR. With the RGB-D input data, this SLAM system can produce 3D scene reconstruction and camera pose estimation in almost real time. The parameters of the system are optimized for use on mobile devices. In addition, an adaptive keyframe selection method is proposed to balance trade-off between accuracy and speed of the system. Experimental results demonstrated its superior performance in comparison with the state-of-the-art methods in the same public benchmark dataset.

Emotion Artificial Intelligence (AI) is a novel technology for advanced human machine interaction in real-world applications. In interactive films, variant contents can be adjusted and displayed in the film based on audiences’ interaction such as voice, hand gesture or body movement, etc. In this paper, a specific emotion detection system is designed and implemented that can detect emotion continuously through the audience’s facial expression and give the feedback to the film immediately. Then the film can change its contents accordingly. In this system, A pre-trained convolutional neural network is used for feature extraction from video frames and then the emotion is predicted by a support vector regression model. The environmental noise is reduced in the pre-processing stage and the final prediction is smoothed in the post-processing. A database is recorded for this particular scene and the proposed system is trained on it. The experimental results demonstrate the effectiveness of the system and the built interactive film “RIOT” has been exhibited on several occasions with good performance.

As one of the most classic human inventions, music can be seen as another language, used to express the author’s thoughts and emotion. Music Emotion Recognition (MER) is an interesting research topic in artificial intelligence field for recognising emotions from music. The recognition methods and tools for music signals are growing fast recently. However the attentions are paid mainly on music signal feature extraction and machine learning methods. The theoretical knowledge of music on emotion is usually ignored. This paper propose a new method through important music units – chord to recognize emotion from music. We firstly build a new chord database with chord segments based on selected emotions. Then FFT and statistics (STAT) features are extracted from the music signals in the database. Then one-shot learning are used for chord identification and emotion recognition using Euclidean distance and correlation. The proposed method can recognise 6 emotions from music signals through chord identification with 76.6% accuracy of chord identification. This idea can be developed further using more advanced feature extraction and one-shot learning methods.

The popular deep convolutional networks used for image super-resolution (SR) reconstruction often increase the network depth and employ attention mechanism to improve image reconstruction effect. However, these networks suffer from two problems. The first is the deeper network easily causes higher computational cost and more GPU memory usage. The second is traditional attention mechanism often misses the spatial information of images leading the loss of image detail information. To address these issues, we propose a lightweight non-local network (LNLN) for image super resolution in this paper. The proposed network makes two contributions. First, we use non-local module instead of normal attention module to obtain larger receptive field and extract more comprehensive feature information, which is helpful for improving image SR reconstruction results. Secondly, we use the depthwise separable convolution (DSC) instead of the vanilla convolution to reconstruct the residual block, which greatly reduces the number of parameters and computational cost. The proposed LNLN and comparative networks are evaluated on five commonly public datasets, and experiments demonstrate that the proposed LNLN is superior to state-of-the-art networks in terms of reconstruction performance, the number of parameters and storage space.

The COVID-19 epidemic in 2020 posed a sudden and serious challenge to education where online teachings were adopted, and problems that people had not noticed before were exposed. Based on the experience of international online course and the study of online teaching, the paper focuses on the associated problems and puts forward some solutions. First, based on the performance of current software and hardware used in online teaching, the characteristics of commonly used teaching software have been analysed and compared. Then, efficient combinations of teaching software for complementary advantages are recommended to ensure a high-quality online teaching. Second, problems inside of class existing in online lecturing and learning are studied, and corresponding teaching methods have been explored including the aspects of class interaction design and implementation, alternative method for delivering essential video information, and online class management. Finally, some objective class evaluation criteria are studied in terms of coverage of class content, learning effect of specialized knowledge, and online class interaction. In the end, some feasible solutions are provided, and it can be served as a reference to improve the online teaching. [ABSTRACT FROM AUTHOR] Copyright of Scientific Programming is the property of Hindawi Limited and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Traditional fuzzy clustering algorithms suffer from two problems in image segmentations. One is that these algorithms are sensitive to outliers due to the non-sparsity of fuzzy memberships. The other is that these algorithms often cause image over-segmentation due to the loss of image local spatial information. To address these issues, we propose a robust self-sparse fuzzy clustering algorithm (RSSFCA) for image segmentation. The proposed RSSFCA makes two contributions. The first concerns a regularization under Gaussian metric that is integrated into the objective function of fuzzy clustering algorithms to obtain fuzzy membership with sparsity, which reduces a proportion of noisy features and improves clustering results. The second concerns a connected-component filtering based on area density balance strategy (CCF-ADB) that is proposed to address the problem of image over-segmentation. Compared to the integration of local spatial information into the objective functions, the presented CCF-ADB is simpler and faster for the removal of small areas. Experimental results show that the proposed RSSFCA addresses two problems in current fuzzy clustering algorithms, i.e., the outlier sensitivity and the over-segmentation, and it provides better image segmentation results than state-of-the-art algorithms. 10.13039/501100001809-National Natural Science Foundation of China (Grant Number: 61871259, 61811530325 (International Exchange Cooperation (IEC)\NSFC\170396, Royal Society, U.K.), 61871260, 61672333 and 61861024)

© 2020 by the authors. Wireless Capsule Endoscopy is a state-of-the-art technology for medical diagnoses of gastrointestinal diseases. The amount of data produced by an endoscopic capsule camera is huge. These vast amounts of data are not practical to be saved internally due to power consumption and the available size. So, this data must be transmitted wirelessly outside the human body for further processing. The data should be compressed and transmitted efficiently in the domain of power consumption. In this paper, a new approach in the design and implementation of a low complexity, multiplier-less compression algorithm is proposed. Statistical analysis of capsule endoscopy images improved the performance of traditional lossless techniques, like Huffman coding and DPCM coding. Furthermore the Huffman implementation based on simple logic gates and without the use of memory tables increases more the speed and reduce the power consumption of the proposed system. Further analysis and comparison with existing state-of-the-art methods proved that the proposed method has better performance.

Face recognition has made significant progress in recent years. It has been successfully applied to many applications from security entrance systems for staff identification to visual surveillance. However, the performance of current systems are highly affected by factors such as subject’s head orientation, facial expressions, and face wear-on. In order to address this challenge, researchers usually resort to more training data or more complex recognition models. We believe the secret lies in using a more rich data source. Therefore, in this paper, we use a holoscopic 3D (H3D) imaging system. To show the effectiveness of the sensor, we use it with a simple back propagation neural network (BPNN) classifier. We illustrate the accuracy of the system through an experiment and show that it has high accuracy and is robust against challenges aforementioned.

With the rapid development of deep learning (DL) in recent years, automatic modulation recognition (AMR) with DL has achieved high accuracy. However, aiming to obtain higher classification accuracy, DL requires numerous training samples. In order to solve this problem, it is a challenge to study how to efficiently use DL for AMR in the case of few samples. In this letter, inspired by the capsule network (CapsNet), we propose a new network structure named AMR-CapsNet to achieve higher classification accuracy of modulation signals with fewer samples, and further analyze the adaptability of DL models in the case of few samples. The simulation results demonstrate that when 3% of the dataset is used to train and the signal-to-noise ratio (SNR) is greater than 2 dB, the overall classification accuracy of the AMR-CapsNet is greater than 80%. Compared with convolutional neural network (CNN), the classification accuracy is improved by 20%.

The EmoPain 2020 Challenge is the first international competition aimed at creating a uniform platform for the comparison of machine learning and multimedia processing methods of automatic chronic pain assessment from human expressive behaviour, and also the identification of pain-related behaviours. The objective of the challenge is to promote research in the development of assistive technologies that help improve the quality of life for people with chronic pain via real-time monitoring and feedback to help manage their condition and remain physically active. The challenge also aims to encourage the use of the relatively underutilised, albeit vital bodily expression signals for automatic pain and pain-related emotion recognition. This paper presents a description of the challenge, competition guidelines, bench-marking dataset, and the baseline systems' architecture and performance on the three sub-tasks: pain estimation from facial expressions, pain recognition from multimodal movement, and protective movement behaviour detection., Comment: 8 pages

Recently, video-based micro-gesture recognition with the data captured by holoscopic 3D (H3D) sensors is getting more and more attention, mainly because of their particular advantages to use a single aperture camera to embed the 3D information in 2D images. However, it is not easy to use the embedded 3D information in an efficient manner due to the special imaging principles of H3D sensors. In this paper, an efficient Pseudo View Points (PVP) based method is proposed to introduce the embedded 3D information in H3D images into a new micro-gesture recognition framework. Specifically, we obtain several pseudo view points based frames by composing all the pixels at the same position in each elemental image(EI) in the original H3D frames. This is a very efficient and robust step, and could mimic the real view points so as to represent the 3D information in the frames. Then, a new recognition framework based on 3D DenseNet and Bi-GRU networks is proposed to learn the dynamic patterns of different micro-gestures based on the representation of the pseudo view points. Finally, we perform a thorough comparison of the related benchmark, which demonstrates the effectiveness of our method and also reports a new state of the art performance.

Finger microgestures have been widely used in human computer interaction (HCI), particularly for interactive applications, such as virtual reality (VR) and augmented reality (AR) technologies, to provide immersive experience. However, traditional 2D image-based microgesture recognition suffers from low accuracy due to the limitations of 2D imaging sensors, which have no depth information. In this article, we proposed an innovative 3D microgesture recognition system based on a holoscopic 3D imaging sensor. Due to the lack of holoscopic 3D datasets, a comprehensive holoscopic 3D microgesture (HoMG) database is created and used to develop a robust 3D microgesture recognition method. Then, a fast algorithm is proposed to extract multiviewpoint images from one holoscopic image. Furthermore, we applied a CNN model with an attention-based residual block to each viewpoint image to improve the algorithm performance. Finally, bagging classification tree decision-level fusion is applied to combine the predictions. The experimental results demonstrate that the proposed method outperforms state-of-the-art methods and delivers a better accuracy than existing methods.

Immersive technology attempts to emulate a physical world through the means of a digital or simulated world. Micro-gestures are small variation actions on human hands defined by user that is one of the most convenient human action in immersive technology. Holoscopic 3D imaging uses bionics technology to capture spatial image in the pattern of “fly’s eye” and it has fruitful 3D cubic information compared to 2D images that can be used for high accurate micro-gesture controller systems. In this paper, a new micro-gesture recognition system based on holoscopic 3D imaging system is proposed for immersive applications. It is built on fast pre-processing, dynamic image feature extraction and a non-linear Support Vector Machine classifier. It is evaluated on the public Holoscopic Micro 3D Gesture (HoMG) dataset outperforming all the existing state-of-the-art methods on the same dataset.

Precisely describing the action inside of a video is a challenging task because the content of the video includes various objects, with different local motion information at different speed in the video frames. In this paper, a new video feature is proposed based on the spatial information of the objects in a frame, along with the motion information between one against multiple consecutive frames. Motion information between pixels at the same position in the whole video are all combined for a new Spatial Multi-Scale Motion History Histogram (SMMHH) dynamic descriptor. The detailed algorithm of the SMMHH was given and it is tested in both human action recognition and touch gesture recognition applications based on the public video datasets. Experimental results demonstrate its excellent performance compared to other traditional methods.

Copyright © 2019 The authors. Ultrasonic Testing (UT) is one of the most important technologies in Non-Detective Testing (NDT) methods. Recently, Barker code and Golay code pairs as coded excitation signals have been applied in ultrasound imaging system with improved quality. However, the signal-to-noise ratio (SNR) of existing UT system based on Barker code or Golay code can be influenced under high high attenuation materials or noisy conditions. In this paper, we apply the convolution of Barker and Golay codes as coded excitation signals for low voltage UT devices that combines the advantages of Barker code and Golay code together. There is no need to change the hardware of UT system in this method. The proposed method has been analyzed theoretically and then in extensive simulations. The experimental results demonstrated that the main lobe level of the code produced by convolution of Barker code and Golay code pairs is much higher than the simple pulse and the main lobe of the combined code is higher than the traditional Barker code, sidelobe is the same as the baker code that constitutes this combined code. So the peak sidelobe level (PSL) of the combined code is lower than the traditional Barker code. Equipped with this, UT devices can be applied in low voltage situations.

Active Shape Model (ASM) has been successfully applied in the segmentation of Diffusion Tensor Magnetic Resonance Image (DT-MRI, referred to as DTI) of brain. However, due to multiple anatomical structure types, irregular shapes, small gray-scale and large amount of these images, perfect segmentation performance could not be achieved. Especially, it is sensitive to initial values with high computational complexity. In this paper, we introduce the gray information of multiple atlases and the prior information of target shapes into the ASM and propose the Multi-Atlas Active Shape Model (referred to as MA-ASM) approach for DTI segmentation. It was evaluated in a manually labeled database with 7 Region of Interest (ROI)s for each of 20 subjects. In comparison with the state of art method of STAPLE (Simultaneous Truth Performance Level Estimation), the proposed algorithm was closer to the manual segmentation shape by subjective visual effects, and had higher overlap rates and lower error detection rates on quantitative analysis than STAPLE.

In this report first, the threat analysis for the IoRL system and the specific use cases it is applied to are assessed. Based on the obtained results an Integrated Security Framework (ISF) is introduced as a countermeasure and its architecture is described in details.

This report describes the system hardware and software architecture of the radio light receiver. The software architecture and building blocks of layer1, layer 2 and layer 3 of the radio light receiver including the intra handover are also described in this report. This report provides also the details of the radio light receiver interface of both mmW Tx/Rx module and VLC receiver. This report also describes the details of the protocols for the IoRL positioning solution. Finally, the consumer products’ design and interfaces are presented in this report.

This report describes the system hardware architecture of the remote radio light head and UE to support IoRL software structure for Access Layer, Network Layer, Orchestration Layer and Service Layer. The software architecture of Infrastructure Layer, Network Layer, Orchestration Layer and Service Layer is also described in this report. This report provides also the details of IoRL position sensing architecture. Finally, the light system electro-mechanical architecture and consumer products electro-mechanical architecture is presented in this report.

Social touch is an important form of social interaction. In Human Robot Interaction (HRI), touch can provide additional information to other modalities, such as audio, visual. One of the application is the robot therapy that has great social significance. In this paper, an ensemble classifier based on threeway decisions is proposed to recognize touch gestures. Firstly, features are extracted from on six perspectives and four classifiers are constructed on different scales with different pre-processing methods. . Then an ensemble classifier is used to combine the four classifiers to classify the gestures. The proposed method is tested on the public Corpus of Social Touch (Cost) dataset. The experiments results not only verify the validity of our method but also show the better accuracy of our ensemble classifier.

Micro-expression is a subtle and involuntary facial expression that may reveal the hidden emotion of human beings. Spotting micro-expression means to locate the moment when the microexpression happens, which is a primary step for micro-expression recognition. Previous work in microexpression expression spotting focus on spotting micro-expression from short video, and with hand-crafted features. In this paper, we present a methodology for spotting micro-expression from long videos. Specifically, a new convolutional neural network named as SMEConvNet (Spotting Micro-Expression Convolutional Network) was designed for extracting features from video clips, which is the first time that deep learning is used in micro-expression spotting. Then a feature matrix processing method was proposed for spotting the apex frame from long video, which uses a sliding window and takes the characteristics of micro-expression into account to search the apex frame. Experimental results demonstrate that the proposed method can achieve better performance than existing state-of-art methods.

Recently, real-time facial expression recognition has attracted more and more research. In this study, an automatic facial expression real-time system was built and tested. Firstly, the system and model were designed and tested on a MATLAB environment followed by a MATLAB Simulink environment that is capable of recognizing continuous facial expressions in real-time with a rate of 1 frame per second and that is implemented on a desktop PC. They have been evaluated in a public dataset, and the experimental results were promising. The dataset and labels used in this study were made from videos, which were recorded twice from five participants while watching a video. Secondly, in order to implement in real-time at a faster frame rate, the facial expression recognition system was built on the field-programmable gate array (FPGA). The camera sensor used in this work was a Digilent VmodCAM — stereo camera module. The model was built on the AtlysTM Spartan-6 FPGA development board. It can continuously perform emotional state recognition in real-time at a frame rate of 30. A graphical user interface was designed to display the participant’s video in real-time and two-dimensional predict labels of the emotion at the same time.