Manipulator trajectory tracking based on adaptive sliding mode control DOI
Haoyi Zhao, Bo Tao, Ruyi Ma

et al.

Concurrency and Computation Practice and Experience, Journal Year: 2022, Volume and Issue: 34(21)

Published: May 24, 2022

Abstract A manipulator is a complex electromechanical system that nonlinear, strongly coupled, and uncertain. Achieving its precise high‐quality trajectory control difficult. Sliding mode (SMC) one of the common methods for manipulators. However, discontinuities in SMC can cause jitter vibration system, leading to reduction performance system. For self‐adaptive capability problem SMC, Dobot magician treated as research object this article. The dynamics equations are established by Lagrange method, simplified model constructed. method sliding proposed. Self‐adaptive parameters added achieve adjustment parameters. In MATLAB/Simulink simulation environment analysis show has better self‐tuning ability tracking than traditional weakens phenomenon existing SMC.

Language: Английский

Low-Light Mine Image Enhancement Algorithm Based on Improved Retinex DOI Creative Commons
Feng Tian, Mengjiao Wang, Xiaopei Liu

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(5), P. 2213 - 2213

Published: March 6, 2024

Aiming at solving the problems of local halo blurring, insufficient edge detail preservation, and serious noise in traditional image enhancement algorithms, an improved Retinex algorithm for low-light mine is proposed. Firstly, HSV color space, hue component remains unmodified, multi-scale guided filtering are combined to estimate illumination reflection components from brightness component. Secondly, equalized using Weber–Fechner law, contrast limited adaptive histogram equalization (CLAHE) fused with denoising Then, saturation adaptively stretched. Finally, it converted back RGB space obtain enhanced image. By comparing single-scale (SSR) (MSR) algorithm, mean, standard deviation, information entropy, average gradient, peak signal-to-noise ratio (PSNR), structural similarity (SSIM) by 50.55%, 19.32%, 3.08%, 28.34%, 29.10%, 22.97%. The experimental dates demonstrate that improves brightness, prevents artifacts while retaining details, reduces effect noise, provides some theoretical references enhancement.

Language: Английский

Citations

8
Detection method of wheat spike improved YOLOv5s based on the attention mechanism DOI Creative Commons
Hecang Zang, Yanjing Wang,

Linyuan Ru

et al.

Frontiers in Plant Science, Journal Year: 2022, Volume and Issue: 13

Published: Sept. 28, 2022

In wheat breeding, spike number is a key indicator for evaluating yield, and the timely accurate acquisition of great practical significance yield prediction. actual production; method using an artificial field survey to count spikes time-consuming labor-intensive. Therefore, this paper proposes based on YOLOv5s with improved attention mechanism, which can accurately detect small-scale better solve problems occlusion cross-overlapping spikes. This introduces efficient channel module (ECA) in C3 backbone structure network model; at same time, global mechanism (GAM) inserted between neck head structure; be more Effectively extract feature information suppress useless information. The result shows that accuracy model reached 71.61% task number, was 4.95% higher than standard had counting accuracy. YOLOv5m have similar parameters, while RMSE MEA are reduced by 7.62 6.47, respectively, performance YOLOv5l. improves its applicability complex environments provides technical reference automatic identification numbers estimation. Labeled images, source code, trained models available at: https://github.com/228384274/improved-yolov5.

Language: Английский

Citations

37
Biomimetic Vision for Zoom Object Detection Based on Improved Vertical Grid Number YOLO Algorithm DOI Creative Commons
Xinyi Shen, Guolong Shi, Huan Ren

et al.

Frontiers in Bioengineering and Biotechnology, Journal Year: 2022, Volume and Issue: 10

Published: May 20, 2022

With the development of bionic computer vision for images processing, researchers have easily obtained high-resolution zoom sensing images. The drones equipped with high-definition cameras has greatly increased sample size and image segmentation target detection are important links during process information. As biomimetic remote usually prone to blur distortion in imaging, transmission processing stages, this paper improves vertical grid number YOLO algorithm. Firstly, light shade a were abstracted, grey-level cooccurrence matrix extracted feature parameters quantitatively describe texture characteristics image. Simple Linear Iterative Clustering (SLIC) superpixel method was used achieve light/dark scenes, saliency area obtained. Secondly, model segmenting dark scenes established made dataset meet recognition standard. Due refraction passing through lens other factors, difference contour boundary value between pixel background would make it difficult detect target, pixels main part separated be sharper edge detection. Thirdly, algorithm an improved proposed real time on processed array. adjusted aspect ratio modified grids network structure by using 20 convolutional layers five maximum aggregation layers, which more accurately adapted "short coarse" identified object information density. Finally, comparison mainstream algorithms different environments, test results aid showed that high spatial resolution images, higher accuracy than had real-time performance accuracy.

Language: Английский

Citations

36
Multi-Objective Location and Mapping Based on Deep Learning and Visual Slam DOI Creative Commons
Ying Sun, Jun Hu, Juntong Yun

et al.

Sensors, Journal Year: 2022, Volume and Issue: 22(19), P. 7576 - 7576

Published: Oct. 6, 2022

Simultaneous localization and mapping (SLAM) technology can be used to locate build maps in unknown environments, but the constructed often suffer from poor readability interactivity, primary secondary information map cannot accurately grasped. For intelligent robots interact meaningful ways with their environment, they must understand both geometric semantic properties of scene surrounding them. Our proposed method not only reduce absolute positional errors (APE) improve positioning performance system also construct object-oriented dense point cloud output model each object reconstruct indoor scene. In fact, eight categories objects are for detection using coco weights our experiments, most actual reconstructed theory. Experiments show that number points is significantly reduced. The average error Technical University Munich (TUM) datasets very small. camera reduced introduction constraints, improved. At same time, algorithm segment environment high accuracy.

Language: Английский

Citations

26
Target Detection Based on Two-Stream Convolution Neural Network With Self-Powered Sensors Information DOI
Li Huang, Xiang Zhao, Juntong Yun

et al.

IEEE Sensors Journal, Journal Year: 2022, Volume and Issue: 23(18), P. 20681 - 20690

Published: Nov. 16, 2022

With the rapid development of artificial intelligence, a neural network is widely used in various fields. The target detection algorithm mainly based on network, but accuracy greatly related to complexity scene and texture. A RGB-D image from perspective lightweight model integration depth map overcome weak environmental illumination with self-powered sensors information proposed. This article analyzes structure YOLOv4 MobileNet, compares variation parameter numbers between depthwise separable convolution convolutional networks, combines advantages MobileNetv3 network. main three effective feature layers replaced by for initial layer extraction strengthen At same time, standard models are convolution. proposed method compared YOLOv4-MobileNetv3 this article, experimental results show that retains its original accuracy, size about 23% model, processing speed 42% higher than can still reach 83% environment poor lighting conditions.

Language: Английский

Citations

21
A Universal Optoelectronic Imaging Platform with Wafer-scale Integration of Two-dimensional Semiconductors DOI Creative Commons
Xinyu Wang, Dié Wang, Yuchen Tian

et al.

Chip, Journal Year: 2024, Volume and Issue: 3(4), P. 100107 - 100107

Published: Aug. 8, 2024

Photodetectors (PDs) are crucial in modern society, as they enable the detection of a diverse range light-based signals. With exponential increase their development, materials being used to create wide PDs that play critical roles enabling various applications and technologies. Image sensor technology has been hindered due lack universal system can integrate all types with silicon-based readout integrated circuits (ROICs). To address this issue, we conducted experiments using two-dimensional an example. We fabricated high-performance MoS2/MoTe2-based photodetectors identified most suitable ROICs pair them. This established solid foundation for further research field image sensors. developed implemented versatile testing uses printed circuit board connect PD ROIC. After ROIC generates sampled signal, it is collected processed by algorithms, which overcome device uniformity limitations produce high-quality visible naked eye. be made from different materials, making highly convenient development types. robust new platform expected spur innovation advancements rapidly developing field.

Language: Английский

Citations

4
An infrared image enhancement method based on Multi-Channel Feature Fusion Network DOI
Boyuan Chen, Yahao Song, Gang Yang

et al.

Optik, Journal Year: 2025, Volume and Issue: unknown, P. 172306 - 172306

Published: March 1, 2025

Language: Английский

Citations

0
Review on applications of computer vision techniques for pipeline inspection DOI
Yiming Liu, Seyed Amirhossein Moghaddas, Shuomang Shi

et al.

Measurement, Journal Year: 2025, Volume and Issue: unknown, P. 117370 - 117370

Published: March 1, 2025

Language: Английский

Citations

0
Application of image enhancement in the auxiliary diagnosis of oral potentially malignant disorders DOI
Jiaqi Wang,

Jiawang Liu,

Y. Liu

et al.

Clinical Oral Investigations, Journal Year: 2025, Volume and Issue: 29(5)

Published: April 25, 2025

Language: Английский

Citations

0
Infrared and visible image fusion network based on low-light image enhancement and attention mechanism DOI
Jinbo Lu,

Zhen Pei,

Jinling Chen

et al.

Signal Image and Video Processing, Journal Year: 2025, Volume and Issue: 19(6)

Published: May 3, 2025

Language: Английский

Citations

0