Human Detection Research Articles

Highly Stretchable, Conductive, Adhesive, and Self-Powered Ionogel Sensor for Human Motion Detection, Signal Transmission, and Traffic Monitoring

Highly Stretchable, Conductive, Adhesive, and Self-Powered Ionogel Sensor for Human Motion Detection, Signal Transmission, and Traffic Monitoring

Improved YOLO-FastestV2 wheat spike detection model based on a multi-stage attention mechanism with a LightFPN detection head.

The number of wheat spikes has an important influence on wheat yield, and the rapid and accurate detection of wheat spike numbers is of great significance for wheat yield estimation and food security. Computer vision and machine learning have been widely studied as potential alternatives to human detection. However, models with high accuracy are computationally intensive and time consuming, and lightweight models tend to have lower precision. To address these concerns, YOLO-FastestV2 was selected as the base model for the comprehensive study and analysis of wheat sheaf detection. In this study, we constructed a wheat target detection dataset comprising 11,451 images and 496,974 bounding boxes. The dataset for this study was constructed based on the Global Wheat Detection Dataset and the Wheat Sheaf Detection Dataset, which was published by PP Flying Paddle. We selected three attention mechanisms, Large Separable Kernel Attention (LSKA), Efficient Channel Attention (ECA), and Efficient Multi-Scale Attention (EMA), to enhance the feature extraction capability of the backbone network and improve the accuracy of the underlying model. First, the attention mechanism was added after the base and output phases of the backbone network. Second, the attention mechanism that further improved the model accuracy after the base and output phases was selected to construct the model with a two-phase added attention mechanism. On the other hand, we constructed SimLightFPN to improve the model accuracy by introducing SimConv to improve the LightFPN module. The results of the study showed that the YOLO-FastestV2-SimLightFPN-ECA-EMA hybrid model, which incorporates the ECA attention mechanism in the base stage and introduces the EMA attention mechanism and the combination of SimLightFPN modules in the output stage, has the best overall performance. The accuracy of the model was P=83.91%, R=78.35%, AP= 81.52%, and F1 = 81.03%, and it ranked first in the GPI (0.84) in the overall evaluation. The research examines the deployment of wheat ear detection and counting models on devices with constrained resources, delivering novel solutions for the evolution of agricultural automation and precision agriculture.

HDetect-VS: Tiny Human Object Enhancement and Detection Based on Visual Saliency for Maritime Search and Rescue

Strong sun glint noise is an inevitable obstruction for tiny human object detection in maritime search and rescue (SAR) tasks, which can significantly deteriorate the performance of local contrast method (LCM)-based algorithms and cause high false alarm rates. For SAR tasks in noisy environments, it is more important to find tiny objects than localize them. Hence, considering background clutter and strong glint noise, in this study, a noise suppression methodology for maritime scenarios (HDetect-VS) is established to achieve tiny human object enhancement and detection based on visual saliency. To this end, the pixel intensity value distributions, color characteristics, and spatial distributions are thoroughly analyzed to separate objects from background and glint noise. Using unmanned aerial vehicles (UAVs), visible images with rich details, rather than infrared images, are applied to detect tiny objects in noisy environments. In this study, a grayscale model mapped from the HSV model (HSV-gray) is used to suppress glint noise based on color characteristic analysis, and large-scale Gaussian Convolution is utilized to obtain the pixel intensity surface and suppress background noise based on pixel intensity value distributions. Moreover, based on a thorough analysis of the spatial distribution of objects and noise, two-step clustering is employed to separate objects from noise in a salient point map. Experiments are conducted on the SeaDronesSee dataset; the results illustrate that HDetect-VS has more robust and effective performance in tiny object detection in noisy environments than other pixel-level algorithms. In particular, the performance of existing deep learning-based object detection algorithms can be significantly improved by taking the results of HDetect-VS as input.

Research on a Human Moving Object Detection Method Based on Gaussian Model and Deep Learning

In order to understand human motion object detection methods, the author proposes a research on human motion object detection method based on Gaussian model. Firstly, traditional Gaussian models are unable to detect complex scenes or slow moving targets. Therefore, an improved Gaussian model based moving object detection algorithm is proposed. Secondly, multiple Gaussian models are used to represent the features of each pixel in the moving target image, and based on the matching of each pixel in the image with the Gaussian model, it is considered as a background point. Conversely, it is based on the principle of the foreground, and the Gaussian model is updated. Finally, by updating the foreground model and calculating short-term stability indicators, the detection effect of moving targets is improved. By determining the Gaussian distribution and pixel relationship, new parameters are set to construct the background model and eliminate the impact caused by sudden changes in lighting. The experimental analysis results show that this method can effectively detect and track moving targets, with good noise resistance, high clarity, and an accuracy rate of up to 99%. Compared with traditional Gaussian model methods, the improved method can more effectively detect moving targets and has better robustness.

EVIDENCE FOR AUDITORY STIMULUS-SPECIFIC ADAPTATION BUT NOT DEVIANCE DETECTION IN LARVAL ZEBRAFISH BRAINS.

Animals receive a constant stream of sensory input, and detecting changes in this sensory landscape is critical to their survival. One signature of change detection in humans is the auditory mismatch negativity (MMN), a neural response to unexpected stimuli that deviate from a predictable sequence. This process requires the auditory system to adapt to specific repeated stimuli while remaining sensitive to novel input (stimulus-specific adaptation). MMN was originally described in humans, and equivalent responses have been found in other mammals and birds, but it is not known to what extent this deviance detection circuitry is evolutionarily conserved. Here we present the first evidence for stimulus-specific adaptation in the brain of a teleost fish, using whole-brain calcium imaging of larval zebrafish at single-neuron resolution with selective plane illumination microscopy. We found frequency-specific responses across the brain with variable response amplitudes for frequencies of the same volume, and created a loudness curve to model this effect. We presented an auditory 'oddball' stimulus in an otherwise predictable train of pure tone stimuli, and did not find a population of neurons with specific responses to deviant tones that were not otherwise explained by stimulus-specific adaptation. Further, we observed no deviance responses to an unexpected omission of a sound in a repetitive sequence of white noise bursts. These findings extend the known scope of auditory adaptation and deviance responses across the evolutionary tree, and lay groundwork for future studies to describe the circuitry underlying auditory adaptation at the level of individual neurons.

A dual-site fluorescent probe for the detection of γ-glutamyl transpeptidase activity and its application in garlic

Developing convenient γ-glutamyl transpeptidase (GGT) activity detection methods is of great significance for soaking Laba garlic and human diseases detection. A dual-site fluorescent probe (probe 1) was developed for detection the activity of GGT. Probe 1 could recognize GGT by the enzymatic hydrolysis of peptide bond by GGT. There has a linear relationship between the fluorescence intensity of probe 1 at 416 nm and the activity of GGT. And the color of the probe solution gradually changed from colorless to blue with the increase of GGT activity under 365 nm ultraviolet light. Importantly, it has a linear relationship between the activity of GGT and the blue (B) value of probe solution photo. Therefore, probes can serve as a convenient tool for detecting GGT activity. More importantly, the probe has been successfully applied to detect of GGT activity in garlic.

Clinical consequences of computer-aided colorectal polyp detection

Background and aimRandomised trials show improved polyp detection with computer-aided detection (CADe), mostly of small lesions. However, operator and selection bias may affect CADe’s true benefit. Clinical outcomes of increased...

Classification of Fashion Models' Walking Styles Using Publicly Available Data, Pose Detection Technology, and Multivariate Analysis: From Past to Current Trendy Walking Styles.

Understanding past and current trends is crucial in the fashion industry to forecast future market demands. This study quantifies and reports the characteristics of the trendy walking styles of fashion models during real-world runway performances using three cutting-edge technologies: (a) publicly available video resources, (b) human pose detection technology, and (c) multivariate human-movement analysis techniques. The skeletal coordinates of the whole body during one gait cycle, extracted from publicly available video resources of 69 fashion models, underwent principal component analysis to reduce the dimensionality of the data. Then, hierarchical cluster analysis was used to classify the data. The results revealed that (1) the gaits of the fashion models analyzed in this study could be classified into five clusters, (2) there were significant differences in the median years in which the shows were held between the clusters, and (3) reconstructed stick-figure animations representing the walking styles of each cluster indicate that an exaggerated leg-crossing gait has become less common over recent years. Accordingly, we concluded that the level of leg crossing while walking is one of the major changes in trendy walking styles, from the past to the present, directed by the world's leading brands.

DFSTrack: Dual-stream fusion Siamese network for human pose tracking in videos

Human pose tracking is a challenging task that involves estimating the human pose and tracking it across multiple frames in a video sequence. In recent years, deep learning-based methods have made significant progress in this field, achieving state-of-the-art performance. However, due to complex background and occlusion among people missed detection and incorrect association matching are still the challenging problems. To address these issues, we adopt a top-down framework to perform human pose tracking in the paper. We propose a human detection prediction recovery module (HDP module) to recover missed detection, and propose a dual-stream fusion Siamese network for human matching (DFSTrack). Specifically, we design a residual graph convolutional block (RGCN block) for spatial position encoding of human keypoints, and use spatial self-attention and temporal cross-attention to design a dual-stream spatial–temporal fusion transformer (DST Transformer). The graph convolutional block and transformer are cascaded to simultaneously obtain information on the spatial and temporal positions of human keypoints, allowing the Siamese network to solve the erroneous human matching. Experimental results on the PoseTrack17 dataset, PoseTrack18 dataset and PoseTrack21 dataset demonstrate that our proposed method outperforms state-of-the-art methods on human pose tracking tasks. Our code and pretrained models are available at https://github.com/yhtian2023/DFSTrack.

Human Detection and Counting using YOLOV4

Object detection is a fundamental problem in computer vision that involves detecting and localizing objects within an image. In recent years, convolutional neural networks (CNNs) have become the state-of-the-art approach for object detection tasks. YOLO (You Only Look Once) is a popular real-time object detection system that uses a single CNN to predict the bounding boxes and class probabilities for objects in an image. In this paper, we present a human detection and counting system using YOLO. Our system is trained on a large dataset of annotated images and achieves high accuracy and real-time performance. We evaluate our system on a public dataset and demonstrate its effectiveness in various scenarios. This paper focuses on developing a system for human detection and counting using YOLOv4 and Flask API. The system aims to address the increasing importance of human detection and counting in various fields such as social distancing, crowd management, and surveillance.

Efficient Fabrication of Bioinspired Flexible Pressure Sensors via Electrohydrodynamic Jet Printing Method.

Bioinspired microdevices have made significant strides in various applications including human motion and health detection. However, facile and highly efficient fabrication approach of flexible pressure sensors remains a great challenge. Herein, inspired by the gecko's foot structure, a flexible pressure sensor with microdomes structure is fabricated by tip-assisted on-demand electrohydrodynamic jet (EHD-jet) printing method. Ascribed to the interlocking electrodes with microdome structure, 3Ddeformation rates are substantially enlarged. When the microdromes structure is under pressure, the resistivity of carbon nanotubes film coated on the surface of microdomes structure will change remarkably. By using the combined effect of assisted tip and ring focusing electrode, the influence and constraints on microstructure fabrication caused by substrate material and morphology are minimized. The desired uniform structures can be adjusted rapidly by changing the printing parameters and liquid properties. High length-height ratio (0.64) of microdomes enhances sensitivity, with minimum detection limit is 2Pa and response time is 40ms. Finally, the bionic flexible sensor indicated excellent performance in capable of detecting pressure, sound vibrations and human motion. This work presents a new method for high-efficiency fabrication micro-nano patterns for flexible sensors inspired, which could be used in wearable tech and health monitoring.

All fiber Mach-Zehnder interferometer immunosensor based on rabbit anti-human IgG and protein a for human IgG detection

Antibodies produced by the immune system, including immunoglobin G (IgG), are among the important biological biomarkers. Optical fibers are ideal for fabricating biosensors because of their flexibility, weight, biocompatibility, electromagnetic interference immunity, etc. This study proposes a multimode-single mode-multimode (MSM) Mach-Zehnder interferometer immunosensor based on rabbit anti-human IgG and protein A (PA) for human IgG detection. When IgG is trapped by antibodies located on the surface of the fiber, the refractive index (RI) changes, and the wavelength redshift occurs in the transmission spectrum of the interference pattern. Here, for binding the antibody to the pre coated PA surface, a rabbit anti-human IgG has been used, which has stronger binding to PA and as a result higher stability than the goat anti-human IgG on the optical fiber surface. By optimizing the optical configuration and biolayer material, a limit of detection of 0.067 μg/ml and sensitivity of 0.14 nm.ml/μg could be achieved.

Non-covalently crosslinked networks MXene-doped poly(eutectic) conductive elastomers with antimicrobial, self-healing, tunable mechanical properties, and wide temperature durability

Poly(eutectic) elastomers, as an emerging category of flexible electronic materials, have garnered growing interest for their cost-effectiveness, non-volatility, and biocompatibility. Nonetheless, they encounter challenges including limited functionalities, sensitivity, durability, and mechanical properties. This study involved synthesizing various MXene-doped poly(eutectic) elastomers using robust acrylamide and soft acrylic acid as organic monomers, minor MXene nanosheets as inorganic fillers and physical crosslinkers, and a rapid in-situ photopolymerization method within 2 min. Owing to the synergistic effect of thermodynamic self-assembly, high-density hydrogen bonding, and cation-dipole interactions in the matrix networks, the resulting non-covalently crosslinked elastomers exhibited remarkable mechanical strength (up to ∼ 16.2 MPa) and impressive tensile properties (nearly 700 %), along with rapid electrical self-healing (within 0.046 s) and the ability to withstand loads up to 2500 times their weight following self-healing without fracturing. Notably, these elastomers also demonstrated excellent microstrain (1 %–10 %) sensitivity with a gauge factor of 2.02, wide temperature tolerance (−20 °C ∼ 60 °C), outstanding harsh temperature durability (−20 °C) exceeding 22,000 cycles, and antibacterial properties. Furthermore, these elastomers-assembled strain sensors displayed real-time and rapid response in human motion detection and wireless signal transmission, presenting a promising approach for designing functionally integrated elastomeric materials.

High-performance optical fiber differential urea sensing system for trace urea concentration detection with enhanced sensitivity using liquid crystals

A compact optical fiber differential sensing system consisting of microfiber interferometer (MFI) functionalized urease combined with stearic acid (SA)-doped 4-cyano-4′-pentylbiphenyl (5CB) and two fiber Bragg gratings (FBGs) for urea detection has been proposed and experimentally demonstrated. The SA-doped 5CB plays an important role of sensing signal amplification in urease specific recognition of urea. The hydrolysis reaction between urea and urease at the microfiber surface causes the deprotonation and self-assembly of SA, which further induces the reorientation of 5CB from planar alignment to homeotropic alignment. This process can be captured and transduced into the macroscopic wavelength shift of the MFI and power difference of two FBGs in the differential sensing system. The experimental results show that the differential sensing system can effectively detect urea in artificial urine, where the sensitivity of MFI is 3.85 nm/mM with a detection limit of 0.03 mM and the power difference sensitivity of two FBGs is 15.05 dB/mM with a detection limit of 0.01 mM within low concentration range of 0.01–0.1 mM. Furthermore, the differential sensing system has good selectivity, affordability, and temperature self-compensation. Therefore, the differential sensing system can be used as a sensitive method for urea detection and has potential applications in human health detection.

Unsupervised exceptional human action detection from repetition of human assembling tasks using entropy signal clustering

Unsupervised exceptional human action detection from repetition of human assembling tasks using entropy signal clustering

Detection and intelligent optimization of blood glucose signals based on MXene-assisted sensors

Two-dimensional materials exhibit unique properties that make them ideal for electronic sensing applications, particularly in the field of biosensors. In this study, a Ti3C2Tx MXene/graphene composite was investigated for its potential use in non-invasive blood glucose detection. The strong electrostatic interaction between MXene and graphene nanosheets effectively reduces the self-stacking of both components, resulting in the formation of a porous interlayer structure with enlarged interlayer spacing. This unique structure creates an efficient nanoscale transport channel, which increases the number of active sites and shortens ion diffusion paths. Consequently, the electrochemical properties of the composite membrane are enhanced, making it a promising candidate for sensing element applications. As the human blood glucose level is primarily determined by the glucose concentration in the blood, the concentration of glucose solution serves as the foundation for noninvasive blood glucose concentration detection. In this work, a prediction model was established using the partial least squares fitting method to correlate the electrochemical fingerprinting data with the glucose aqueous solution concentration, and the reliability of the model was verified. To address the limitations of the partial least squares method, which lacks nonlinear fitting capabilities, an RBF (radial basis function) network model was employed. Finally, a genetic algorithm was utilized to perform global optimization of the training parameters of the RBF network, and the quantitative relationship between the NIR spectral data and glucose concentration was validated. The findings of this study provide a theoretical basis for the development of equipment and the improvement of measurement accuracy in the field of human noninvasive glucose detection.

Flexible staircase triboelectric nanogenerator for motion monitoring and gesture recognition

Triboelectric sensors, which are used in human-computer interaction and condition monitoring, have attracted tremendous scientific interest owing to their broad applicability, high energy utilization, and high accuracy detection. Here, a staircase-style flexible triboelectric sensor assembled by the antimony-doped tin oxide (ATO) functionalized flexible polyvinylidene fluoride polymer (PVDF) composite film is first proposed for gesture recognition and human motion detection. The modified composite film ATO/PVDF with high polarization and large permittivity improves the electrical output of the triboelectric nanogenerator (TENG). Compared with the pure PVDF, the modified PVDF film-based TENG could substantially enhance power density by 11.5 times. The peak voltage and power density of the flexible staircase-style TENG achieve 70 V and 0.58 W/m2, respectively. The flexible triboelectric sensor can be attached to the finger or wrist to obtain movement information through positive/negative signals. A set of rules between gestures and corresponding letters and numbers is defined and a gesture recognition system using deep learning algorithms is developed. The flexible triboelectric sensor realizes human body part bending, gesture, and motion state recognition, offering potential solutions and novel possibilities in smart sensing and human-computer interaction.

Hf3C2O2 MXene: A promising NH3 gas sensor with high selectivity/sensitivity and fast recover time at room temperature

Owing to metallic conductivity, tunable surface chemistry, mechanical flexibility, high surface area and chemical stability, MXenes have emerged as promising candidates for room-temperature gas sensors. In this study, gas sensing performances of the newly synthesized Hf3C2 MXene with oxygen termination toward a wide range of common molecules (NH3, H2O, H2S, H2, CO2, CO, CH4, N2 and NO) were investigated by first-principles density functional calculations. With the most negative adsorption energy, the highest charge transfer and the shortest adsorption distance, the adsorption of NH3 among the investigated gases causes the most significant changes in electrical conductivity and work function of Hf3C2O2 substrate, thus holds the biggest gas response. Only the NH3 exhibits strong chemical interactions with Hf3C2O2, and other gas molecules are weakly chemisorbed or physisorbed. All these indicate that Hf3C2O2 exhibits high selectivity and sensitivity to NH3. The adsorption energy of NH3 falls within the appropriate range for room-temperature gas sensors, enabling Hf3C2O2 to maintain NH3 sensitivity while also achieving a fast recovery time. Furthermore, giving that human exhaled breath does not affect, and may even enhance, the selectivity and sensitivity of Hf3C2O2 towards NH3, Hf3C2O2 can serve effectively as an NH3 sensor in human disease detection.

Self-healing and reprocessable biobased non-isocyanate polyurethane elastomer with dual dynamic covalent adaptive network for flexible strain sensor

Owing to the nontoxicity of monomers, non-isocyanate polyurethane elastomers have attracted considerable attention, yet the synthesis and functionalization are still challenging. Herein, self-healing and reprocessable biobased non-isocyanate polyurethane (NIPU) elastomers with dual dynamic covalent adaptive network (DDCAN) by disulfide bonds and dynamic imine bonds were synthesized as substrate for the construction of flexible strain sensor with MXene as conductive substance. The tensile strength and elongation at break of the NIPU elastomer were 3.22 MPa and 234 %, respectively. Thanks to the formation of DDCAN, the NIPU elastomer showed high healing efficiency of 97.5 % after healing at room temperature for 24 h. Interestingly, the NIPU elastomer possessed superior reprocessing capability and the tensile strength after three reprocessing cycles reached 136.1 % of the original one, which was attributed to the increase of crosslinking points caused by topological rearrangement. In addition, the NIPU elastomer-based flexible strain sensor exhibited fast response (response time = 60 ms) and excellent repeatability (1000 bending-releasing cycles) and was successfully applied for human motion detection and speech recognition. The findings in this work conceivably stand out as a new methodology for the preparation of biobased and functional elastomers, which will greatly promote the sustainable development and application of flexible electronics.

Advanced Millimeter-Wave Radar System for Real-Time Multiple-Human Tracking and Fall Detection.

This study explored an indoor system for tracking multiple humans and detecting falls, employing three Millimeter-Wave radars from Texas Instruments. Compared to wearables and camera methods, Millimeter-Wave radar is not plagued by mobility inconveniences, lighting conditions, or privacy issues. We conducted an initial evaluation of radar characteristics, covering aspects such as interference between radars and coverage area. Then, we established a real-time framework to integrate signals received from these radars, allowing us to track the position and body status of human targets non-intrusively. Additionally, we introduced innovative strategies, including dynamic Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering based on signal SNR levels, a probability matrix for enhanced target tracking, target status prediction for fall detection, and a feedback loop for noise reduction. We conducted an extensive evaluation using over 300 min of data, which equated to approximately 360,000 frames. Our prototype system exhibited a remarkable performance, achieving a precision of 98.9% for tracking a single target and 96.5% and 94.0% for tracking two and three targets in human-tracking scenarios, respectively. Moreover, in the field of human fall detection, the system demonstrates a high accuracy rate of 96.3%, underscoring its effectiveness in distinguishing falls from other statuses.