It is evident that surface electromyography (sEMG) based human-machine interfaces (HMI) have inherent difficulty in predicting dexterous musculoskeletal movements such as finger motions. This paper an attempt to investigate a plausible alternative sEMG, ultrasound-driven HMI, for motion recognition due its characteristic of detecting morphological changes deep muscles and tendons. A multi-channel A-mode ultrasound lightweight device adopted evaluate the performance recognition; experiment...

Motions of the fingers are complex since hand grasping and manipulation conducted by spatial temporal coordination forearm muscles tendons. The dominant methods based on surface electromyography (sEMG) could not offer satisfactory solutions for finger motion classification due to its inherent nature measuring electrical activity motor units at skin's surface. In order recognize morphological changes accurate prediction, ultrasound imaging is employed investigate feasibility detecting...

Summary With the continuous development of sensor technology, acquisition cost RGB‐D images is getting lower and lower, gesture recognition based on depth Red‐Green‐Blue (RGB) has gradually become a research direction in field pattern recognition. However, most current processing methods for are relatively simple, ignoring relationship influence between its two modes, unable to make full use correlation factors different modes. In view above problems, this paper optimizes effect information...

Abstract The polycyclic aromatic hydrocarbon (PAH) concentrations in total suspended particulate matter (TSP) samples collected from October, 2021 to September, 2022 were analyzed clarify the pollution characteristics and sources of 16 PAHs atmospheric TSP Bengbu City. ρ(PAHs) ranged 1.71 43.85 ng/m 3 higher detected winter, followed by spring, autumn, summer. positive matrix factorization analysis revealed that, spring summer, PAH was caused mainly industrial emissions, gasoline diesel fuel...

It is evident that user training significantly affects performance of pattern-recognition-based myoelectric prosthetic device control. Despite plausible classification accuracy on offline datasets, online usually suffers from the changes in physiological conditions and electrode displacement. The ability generating consistent electromyographic (EMG) patterns can be enhanced via proper strategies order to improve performance. This study proposes a clustering-feedback strategy provides...

Abstract Robot grasping technology is a hot spot in robotics research. In relatively fixed industrialized scenarios, using robots to perform grabbing tasks efficient and lasts long time. However, an unstructured environment, the items are diverse, placement posture random, multiple objects stacked occluded each other, which makes it difficult for robot recognize target when grasped grasp method complicated. Therefore, we propose accurate, real‐time detection based on convolutional neural...

Summary Gesture recognition has always been a research hotspot in the field of human‐computer interaction. Its purpose is to realize natural interaction with machine by recognizing semantics expressed gesture. In process gesture recognition, occlusion an inevitable problem. some or even all features will be lost due gesture, resulting wrong unrecognizability Therefore, it great significance study under occlusion. The single shot multibox detector (SSD) algorithm analyzed, and front‐end...

The ability to predict wrist and hand motions simultaneously is essential for natural controls of protheses. In this paper, we propose a novel method that includes subclass discriminant analysis (SDA) principal component the simultaneous prediction rotation (pronation/supination) finger gestures using wearable ultrasound. We tested on eight with concurrent rotations. Results showed SDA was able achieve accurate classification both rotations under dynamic When grouping into three subclasses,...

Global temporal information and local semantic are essential cues for high-performance online object detection in videos. However, despite their promising accuracy most cases, state-of-the-art approaches have following two limitations: invalid background/scale suppression inadequate mining between frames. Many jobs currently focus on learning based a single frame. In this article, we propose an attentional global–local network; is one of the first attempts to fully use both types Attention...

Fine multifunctional prosthetic hand manipulation requires precise control on the pinch-type and corresponding force, it is a challenge to decode both aspects from myoelectric signals. This paper proposes an attribute-driven granular model (AGrM) under machine-learning scheme solve this problem. The utilizes additionally captured attribute as latent variable for supervised granulation procedure. It was fulfilled EMG-based classification fingertip force grand prediction. In experiments, 16...

Hand detection is a crucial technology for space human-robot interaction (SHRI), and the awareness of hand identities particularly critical. However, most advanced works have three limitations: 1) low accuracy small-size objects; 2) insufficient temporal feature modeling between frames in videos; 3) inability real-time detection. In article, detector (called TA-RSSD) proposed based on SSD spatiotemporal long short-term memory (ST-LSTM) SHRI applications. Next, online tubelet analysis,...

Because large numbers of artworks are preserved in museums and galleries, much work must be done to classify these works into genres, styles artists. Recent technological advancements have enabled an increasing number digitized. Thus, it is necessary teach computers analyze (e.g., annotate) art assist people performing such tasks. In this study, we tested 7 different models on 3 datasets under the same experimental setup compare their classification performances when either using or not...

Surface electromyography (sEMG) non-invasively captures the electrical activity generated by muscle contractions, offering valuable insights into motion intentions. While sEMG has been widely applied to general gesture recognition in rehabilitation, there limited exploration of specific, intricate daily tasks, such as pouring action. Pouring is a common yet complex movement requiring precise coordination and control, making it an ideal focus for rehabilitation studies. This research proposes...

How do humans recognize an action or interaction in the real world? Due to diversity of viewing perspectives, it is a challenge for identify regular activity when they observe from uncommon perspective. We argue that discriminative spatiotemporal information remains essential cue human recognition. Most existing skeleton-based methods learn optimal representation based on human-crafted criterion requires many labeled data and much effort. This article introduces adaptive neural networks...

While myoelectric pattern recognition is a prevailing way for gesture recognition, the inherent nonstationarity of electromyography signals hinders its long-term application. This study aims to prove hypothesis that morphological information muscle contraction detected by ultrasound image potentially suitable use. A set ultrasound-based algorithms are proposed realize robust hand over multiple days, with user training only at first day. markerless calibration algorithm presented position...

The learning of inter-day representation electromyographic (EMG) signals across multiple days remains a challenging topic and not fully accommodated yet. This study aims to improve the hand motion classification accuracy via convolutional neural network (CNN)-based data feature fusion. An EMG database (ISRMyo-I) was recorded from six subjects on 10 low density electrode setting. investigated CNNs’ capability learning, found that output first connected layer (CNNFeats) decent supplement set...

Pattern recognition algorithms have been applied in the surface electromyography (sEMG) based hand motion for their promising accuracy.Research on proposing new features, improving classifiers and combinations has extensively conducted past decade.Meanwhile, feature projection methodology, routinely exploited between phases of extraction classification.Meanwhile, limited publications seen addressing selection, which is a vital alternative dimensionality reduction pattern recognition.Recent...

Small object detection is the main challenge for image of unmanned aerial vehicles (UAVs), especially with small pixel ratios and blurred boundaries. In this article, a one-stage detector (SF-SSD) proposed new spatial cognition algorithm. The deconvolution operation introduced to feature fusion module, which enhances representation shallow features. These more representative features prove effective small-scale detection. Empowered by method, deep model can redetect objects less-reliable...

Given the dense population on university campuses, indoor and outdoor airborne bacterial contamination may lead to rapid spread of diseases in a environment. However, there are few studies characteristics pathogenic communities different sites campus. In this study, we collected particulate matter samples from locations at Bengbu City, Anhui Province, China, analyzed community bacteria using high-throughput sequencing technique. The results showed that composition dominant was consistent...