Chao Zhang

Bio: Chao Zhang is an academic researcher from Anhui University. The author has contributed to research in topics: Computer science & Electrooculography. The author has an hindex of 7, co-authored 18 publications receiving 118 citations.

Driver drowsiness detection using multi-channel second order blind identifications

Abstract: It is well known that blink, yawn, and heart rate changes give clue about a human's mental state, such as drowsiness and fatigue. In this paper, image sequences, as the raw data, are captured from smart phones which serve as non-contact optical sensors. Video streams containing subject's facial region are analyzed to identify the physiological sources that are mixed in each image. We then propose a method to extract blood volume pulse and eye blink and yawn signals as multiple independent sources simultaneously by multi-channel second-order blind identification (SOBI) without any other sophisticated processing, such as eye and mouth localizations. An overall decision is made by analyzing the separated source signals in parallel to determine the driver's driving state. The robustness of the proposed method is tested under various illumination contexts and a variety of head motion modes. Experiments on 15 subjects show that the multi-channel SOBI presents a promising framework to accurately detect drowsiness by merging multi-physiological information in a less complex way.

DETRs with Hybrid Matching

Abstract: One-to-one set matching is a key design for DETR to establish its end-to-end capability, so that object detection does not require a hand-crafted NMS (non-maximum suppression) to remove duplicate detections. This end-to-end signature is important for the versatility of DETR, and it has been generalized to broader vision tasks. However, we note that there are few queries assigned as positive samples and the one-to-one set matching significantly reduces the training efficacy of positive samples. We propose a simple yet effective method based on a hybrid matching scheme that combines the original one-to-one matching branch with an auxiliary one-to-many matching branch during training. Our hybrid strategy has been shown to significantly improve accuracy. In inference, only the original one-to-one match branch is used, thus maintaining the end-to-end merit and the same inference efficiency of DETR. The method is named H-DETR, and it shows that a wide range of representative DETR methods can be consistently improved across a wide range of visual tasks, including DeformableDETR, PETRv2, PETR, and TransTrack, among others. The code is available at: https://github.com/HDETR

To Explore the Potentials of Independent Component Analysis in Brain-Computer Interface of Motor Imagery

Abstract: This paper is focused on the experimental approach to explore the potential of independent component analysis (ICA) in the context of motor imagery (MI)-based brain-computer interface (BCI). We presented a simple and efficient algorithmic framework of ICA-based MI BCI (ICA-MIBCI) for the evaluation of four classical ICA algorithms (Infomax, FastICA, Jade, and Sobi) as well as a simplified Infomax (sInfomax). Two novel performance indexes, self-test accuracy and the number of invalid ICA filters, were employed to assess the performance of MIBCI based on different ICA variants. As a reference method, common spatial pattern (CSP), a commonly-used spatial filtering method, was employed for the comparative study between ICA-MIBCI and CSP-MIBCI. The experimental results showed that sInfomax-based spatial filters exhibited significantly better transferability in session to session and subject to subject transfer as compared to CSP-based spatial filters. The online experiment was also introduced to demonstrate the practicability and feasibility of sInfomax-based MIBCI. However, four classical ICA variants, especially FastICA, Jade, and Sobi, performed much worse as compared to sInfomax and CSP in terms of classification accuracy and stability. We consider that conventional ICA-based spatial filtering methods tend to be overfitting while applied to real-life electroencephalogram data. Nevertheless, the sInfomax-based experimental results indicate that ICA methods have a great space for improvement in the application of MIBCI. We believe that this paper could bring forth new ideas for the practical implementation of ICA-MIBCI.

A study on medical Internet of Things and Big Data in personalized healthcare system

Abstract: Personalized healthcare systems deliver e-health services to fulfill the medical and assistive needs of the aging population Internet of Things (IoT) is a significant advancement in the Big Data era, which supports many real-time engineering applications through enhanced services Analytics over data streams from IoT has become a source of user data for the healthcare systems to discover new information, predict early detection, and makes decision over the critical situation for the improvement of the quality of life In this paper, we have made a detailed study on the recent emerging technologies in the personalized healthcare systems with the focus towards cloud computing, fog computing, Big Data analytics, IoT and mobile based applications We have analyzed the challenges in designing a better healthcare system to make early detection and diagnosis of diseases and discussed the possible solutions while providing e-health services in secure manner This paper poses a light on the rapidly growing needs of the better healthcare systems in real-time and provides possible future work guidelines

Motor Imagery EEG Signals Decoding by Multivariate Empirical Wavelet Transform-Based Framework for Robust Brain–Computer Interfaces

Abstract: The robustness and computational load are the key challenges in motor imagery (MI) based on electroencephalography (EEG) signals to decode for the development of practical brain-computer interface (BCI) systems. In this study, we propose a robust and simple automated multivariate empirical wavelet transform (MEWT) algorithm for the decoding of different MI tasks. The main contributions of this study are four-fold. First, the multiscale principal component analysis method is utilized in the preprocessing module to obtain robustness against noise. Second, a novel automated channel selection strategy is proposed and then is further verified with comprehensive comparisons among three different strategies for decoding channel combination selection. Third, a sub-band alignment method by utilizing MEWT is adopted to obtain joint instantaneous amplitude and frequency components for the first time in MI applications. Four, a robust correlation-based feature selection strategy is applied to largely reduce the system complexity and computational load. Extensive experiments for subject-specific and subject independent cases are conducted with the three-benchmark datasets from BCI competition III to evaluate the performances of the proposed method by employing typical machine-learning classifiers. For subject-specific case, experimental results show that an average sensitivity, specificity and classification accuracy of 98% was achieved by employing multilayer perceptron neural networks, logistic model tree and least-square support vector machine (LS-SVM) classifiers, respectively for three datasets, resulting in an improvement of upto 23.50% in classification accuracy as compared with other existing method. While an average sensitivity, specificity and classification accuracy of 93%, 92.1% and 91.4% was achieved for subject independent case by employing LS-SVM classifier for all datasets with an increase of up to 18.14% relative to other existing methods. Results also show that our proposed algorithm provides a classification accuracy of 100% for subjects with small training size in subject-specific case, and for subject independent case by employing a single source subject. Such satisfactory results demonstrate the great potential of the proposed MEWT algorithm for practical MI EEG signals classification.

A Review on Human-Computer Interaction and Intelligent Robots

Abstract: In the field of artificial intelligence, human–computer interaction (HCI) technology and its related intelligent robot technologies are essential and interesting contents of research. From the perspective of software algorithm and hardware system, these above-mentioned technologies study and try to build a natural HCI environment. The purpose of this research is to provide an overview of HCI and intelligent robots. This research highlights the existing technologies of listening, speaking, reading, writing, and other senses, which are widely used in human interaction. Based on these same technologies, this research introduces some intelligent robot systems and platforms. This paper also forecasts some vital challenges of researching HCI and intelligent robots. The authors hope that this work will help researchers in the field to acquire the necessary information and technologies to further conduct more advanced research.

Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection

Abstract: In this paper, we present an open-set object detector, called Grounding DINO, by marrying Transformer-based detector DINO with grounded pre-training, which can detect arbitrary objects with human inputs such as category names or referring expressions. The key solution of open-set object detection is introducing language to a closed-set detector for open-set concept generalization. To effectively fuse language and vision modalities, we conceptually divide a closed-set detector into three phases and propose a tight fusion solution, which includes a feature enhancer, a language-guided query selection, and a cross-modality decoder for cross-modality fusion. While previous works mainly evaluate open-set object detection on novel categories, we propose to also perform evaluations on referring expression comprehension for objects specified with attributes. Grounding DINO performs remarkably well on all three settings, including benchmarks on COCO, LVIS, ODinW, and RefCOCO/+/g. Grounding DINO achieves a $52.5$ AP on the COCO detection zero-shot transfer benchmark, i.e., without any training data from COCO. It sets a new record on the ODinW zero-shot benchmark with a mean $26.1$ AP. Code will be available at \url{https://github.com/IDEA-Research/GroundingDINO}.