Design and implementation of a parallel OTSU algorithm based on a ternary optical computer DOI
Kai Song,

Yongjun Sun,

Huaqiong Ma

et al.

Applied Optics, Journal Year: 2023, Volume and Issue: 62(36), P. 9591 - 9591

Published: Nov. 29, 2023

The OTSU method (OTSU) is considered to be the best algorithm for threshold selection in image segmentation. easy calculate and not affected by brightness contrast, so it widely used field of digital processing. Due slow running speed large storage space on traditional electronic computer, this research makes use advantages ternary optical computer (TOC), such as numerous data bits reconfigurable processor. Through analysis algorithm, we find part that can calculated parallel, put forward a parallel based TOC. first uses classifier classify number pixels. By reconstructing MSD adder multiplier, inter-class variance under each gray value maximum selected comparing variance. corresponding segmentation threshold. analyzing clock cycle bit resources then conducting comparative experimental verification, found computational TOC increased approximately 12% when handling images with same amount computation.

Language: Английский

Research on Fast Multi-Threshold Image Segmentation Technique Using Histogram Analysis DOI Open Access

Mingjin Xu,

Shaoshan Chen,

Xiaopeng Gao

et al.

Electronics, Journal Year: 2023, Volume and Issue: 12(21), P. 4446 - 4446

Published: Oct. 29, 2023

This paper investigates a method for the multi-threshold segmentation of grayscale imaging using local minimum points histogram curve as threshold. By smoothing and judging conditions, expected peaks valleys are identified, corresponding used thresholds to achieve fast image segmentation. Compared OTSU (maximum between-class variance) region growing method, this has less computational complexity. In recognition process solder pads with adhesion underfill in LED Chips, time is than one percent that method. The effect better it can images. Moreover, strong adaptability differences overall images, meeting requirements high UPH (Units Per Hour) industrial production lines.

Language: Английский

Citations

7
A trajectory planning method for a casting sorting robotic arm based on a nature-inspired Genghis Khan shark optimized algorithm DOI Creative Commons
Chengjun Wang,

Xingyu Yao,

Fan Ding

et al.

Mathematical Biosciences & Engineering, Journal Year: 2024, Volume and Issue: 21(2), P. 3364 - 3390

Published: Jan. 1, 2024

<abstract> <p>In order to meet the efficiency and smooth trajectory requirements of casting sorting robotic arm, we propose a time-optimal planning method that combines heuristic algorithm inspired by behavior Genghis Khan shark (GKS) segmented interpolation polynomials. First, basic model arm was constructed based on parameters, workspace is analyzed. A matrix formed combining cubic quintic polynomials using approach solve for 14 unknown parameters plan trajectory. To enhance smoothness in joint space, dynamic nonlinear learning factor introduced traditional Particle Swarm Optimization (PSO) algorithm. Four different biological behaviors, GKS, were simulated. Within premise time optimality, target function set effectively optimize within feasible space. Simulation verification performed after determining working tasks arm. The results demonstrated optimized achieved continuous velocity, while also optimizing overall runtime given constraints. comparison made between PSO an improved algorithm, revealing exhibited better convergence. Moreover, GKS showed decreased likelihood getting trapped local optima, thereby confirming effectiveness proposed algorithm.</p> </abstract>

Language: Английский

Citations

2
Particle Swarm Optimizer Variants for Multi-level Thresholding: Theory, Performance Enhancement and Evaluation DOI
Arunita Das, Buddhadev Sasmal, Krishna Gopal Dhal

et al.

Archives of Computational Methods in Engineering, Journal Year: 2024, Volume and Issue: 31(5), P. 3115 - 3150

Published: March 1, 2024

Language: Английский

Citations

2
Multi-threshold remote sensing image segmentation with improved ant colony optimizer with salp foraging DOI Creative Commons

Yunlou Qian,

Jiaqing Tu,

Gang Luo

et al.

Journal of Computational Design and Engineering, Journal Year: 2023, Volume and Issue: 10(6), P. 2200 - 2221

Published: Oct. 16, 2023

Abstract Remote sensing images can provide direct and accurate feedback on urban surface morphology geographic conditions. They be used as an auxiliary means to collect data for current geospatial information systems, which are also widely in city public safety. Therefore, it is necessary research remote images. we adopt the multi-threshold image segmentation method this paper segment research. We first introduce salp foraging behavior into continuous ant colony optimization algorithm (ACOR) construct a novel ACOR version based (SSACO). The original algorithm’s convergence ability avoid hitting local optima enhanced by behavior. In order illustrate key benefit, SSACO tested against 14 fundamental algorithms using 30 benchmark test functions IEEE CEC2017. Then, compared with other algorithms. experimental results examined from various angles, findings convincingly demonstrate main power of SSACO. performed comparison studies 12 between techniques several peer approaches benefits segmentation. Peak signal-to-noise ratio, structural similarity index, feature index evaluation demonstrated SSACO-based approach. excellent optimizer since seeks serve guide point reference

Language: Английский

Citations

6
A Partial Discharge Localization Method Based on the Improved Artificial Fish Swarms Algorithm DOI Creative Commons
Hao Qiang, Qun Wang, Hui Niu

et al.

Energies, Journal Year: 2023, Volume and Issue: 16(6), P. 2928 - 2928

Published: March 22, 2023

Accurate localization of partial discharge in GIS equipment remains a key focus daily maintenance for substations, which can be achieved through advanced detection and location techniques, as well regular testing the equipment. However, there is currently an issue with low accuracy algorithm. Aiming at problems precision local optimization swarm intelligence algorithm system equipment, this paper proposes 3D based on time difference arrival (TDOA) model improved artificial fish (IAFSA). By introducing investigation behaviour bee colony(ABC) into swarms (AFSA), more efficient to jump out extremum, enhance performance, improve global search ability overcome premature convergence. Furthermore, precise positioning dynamic parameters. The results function show that IAFSA significantly superior AFSA particle (PSO) terms stability. When applied experiments, maximum relative error less than 2.5%. This validates proposed method achieve high-precision localization, has good engineering application value, provides strong support safe operation

Language: Английский

Citations

5
Hybrid-Supervised-Learning-Based Automatic Image Segmentation for Water Leakage in Subway Tunnels DOI Creative Commons
Dongwei Qiu, Haorong Liang, Zhilin Wang

et al.

Applied Sciences, Journal Year: 2022, Volume and Issue: 12(22), P. 11799 - 11799

Published: Nov. 20, 2022

Quickly and accurately identifying water leakage is one of the important components health monitoring subway tunnels. A mobile vision measurement system consisting several high-resolution, industrial, charge-coupled device (CCD) cameras placed on trains to implement structural in Through image processing technology proposed this paper, areas tunnels can be found repaired real time. lightweight automatic segmentation approach using hybrid-supervised-deep-learning proposed. This consists weakly supervised learning Water Leakage-CAM fully WRDeepLabV3+. The used for labeling data. WRDeepLabV3+ accurate identification Compared with other end-to-end semantic networks, hybrid-supervised more completely segment region when dealing complex environments. paper achieves highest MIoU 82.8% experimental dataset, which 6.4% higher than second. efficiency also 25% second significantly outperforms deep approaches.

Language: Английский

Citations

7
Multi-threshold image segmentation using new strategies enhanced whale optimization for lupus nephritis pathological images DOI
Jinge Shi, Yi Chen, Chaofan Wang

et al.

Displays, Journal Year: 2024, Volume and Issue: 84, P. 102799 - 102799

Published: July 20, 2024

Language: Английский

Citations

1
Lightweight mask R-CNN for instance segmentation and particle physical property analysis in multiphase flow DOI

Ming-Xiang He,

Kexin He,

Qingshan Huang

et al.

Powder Technology, Journal Year: 2024, Volume and Issue: unknown, P. 120366 - 120366

Published: Oct. 1, 2024

Language: Английский

Citations

1
Research on Wavelet Transform Modulus Maxima and OTSU in Edge Detection DOI Creative Commons
Ning You, Libo Han, Yuming Liu

et al.

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(7), P. 4454 - 4454

Published: March 31, 2023

During routine bridge maintenance, edge detection allows the partial condition of to be viewed. However, many methods often have unsatisfactory performances when dealing with images complex backgrounds. Moreover, processing involves manual selection thresholds, which can result in repeated testing and comparisons. To address these problems this paper, wavelet transform modulus maxima method is used detect target image, then threshold value image determined automatically according OTSU remove pseudo-edges. Thus, real edges detected. The results show that information entropy SSIM are highest compared commonly Canny Laplace algorithms, means quality optimal. more fully illustrate advantages backgrounds were detected algorithms paper still In addition, automatic thresholds saves operator’s effort improves efficiency. Thanks combined use above two methods, efficiency significantly improved, has a good application engineering practice.

Language: Английский

Citations

2
Research on Wavelet Transform Modulus Maxima and OTSU in Edge Detection DOI Open Access
Ning You, Libo Han, Yuming Liu

et al.

Published: Sept. 21, 2023

During routine bridge maintenance, edge detection allows the partial condition of to be viewed. However, many methods often have unsatisfactory performances when dealing with images complex backgrounds. Moreover, processing involves manual selection thresholds, which can result in repeated testing and comparisons. To address these problems this paper, wavelet transform modulus maxima method is used detect target image, then threshold value image determined automatically according OTSU remove pseudo-edges. Thus, real edges detected. The results show that information entropy SSIM are highest compared commonly Canny Laplace algorithms, means quality optimal. more fully illustrate advantages backgrounds were detected algorithms paper still In addition, automatic thresholds saves operator&rsquo;s effort improves efficiency. Thanks combined use above two methods, efficiency significantly improved, has a good application engineering practice.

Language: Английский

Citations

2