遮挡和光照变化下的单鱼目标跟踪

doi:10.3969/j.issn.1000-1158.2021.02.07

Abstract
Figure/Table
References (17)
Related Citation (15)

Download: PDF (4493 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Fish tracking is the basis of fish behavior analysis, when tracking fish in water, the target fish will be difficult to track because of the change of posture and the influence of the occlusion or light of the surrounding fish or object. In order to solve the above problem, a method of tracking fish body by combining Mobilenet-SSD(SSD, single shot multibox detector) with Dlib association tracker was proposed. The fish body in video is accurately detected by SSD algorithm, and then the information is input into Dlib association tracker, which makes the tracking object location more accurate and improves the robustness of fish body tracking when the movement of fish in water occured and the light changes. The experimental results show that the performance of the mentioned method in fish tracking video is obviously better than that of other algorithms, the tracking success rate in different environments is more than 90%.

Key words： metrology computer vision single fish target tracking mobilenet_SSD Dlib image processing

Received: 05 September 2019 Published: 18 February 2021

PACS:

TB96

Fund:;Postdoctoral Science Foundation of Hebei Province

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	CHENG Shu-hong
	WANG Ying

Cite this article:

CHENG Shu-hong,WANG Ying. Single Fish Tracking under Occlusion and Light Change[J]. Acta Metrologica Sinica, 2021, 42(2): 171-177.

URL:

http://jlxb.china-csm.org:81/Jwk_jlxb/EN/10.3969/j.issn.1000-1158.2021.02.07 OR http://jlxb.china-csm.org:81/Jwk_jlxb/EN/Y2021/V42/I2/171

［1］何佳, 黄志涛, 宋协法, 等. 基于计算机视觉技术的水产养殖中鱼类行为识别与量化研究进展［J］. 渔业现代化, 2019, 46 (3): 7-14.
He J, Huang Z T, Song X F, et al. Research progress on recognition and quantification of fish behavior in aquaculture based on computer vision technology［J］. Fishery Modernization, 2019, 46 (3): 7-14.
［2］李永亮, 李健, 李桂莲, 等. 发光菌法测定医疗机构废水生物毒性［J］. 计量学报, 2017, 38 (2): 238-241.
Li Y L, Li J, Li G L, et al. Determination of biological toxicity of wastewater from medical institutions by light-emitting bacteria method［J］. Acta Metrologica Sinica, 2017, 38 (2): 238-241.
［3］Baatrup E. Medication in the environment affects feeding behavior of fish［EB/OL］. http://dx.doi.org/10.1002/etc.4474, 2019-03-18.
［4］Ma H, Tsai T F, Liu C C . Real-time monitoring of water quality using temporal trajectory of live fish［J］. Expert Systems with Applications, 2010, 37 (7): 5158-5171.
［5］卢焕达, 刘鹰, 范良忠. 基于计算机视觉的鱼类行为自动监测系统设计与实现［J］. 渔业现代化, 2011, 38 (1): 19-23.
Lu H D, Liu Y, Fan L Z. Design and implementation of automatic Fish behavior Monitoring system based on computer Vision［J］. Fishery Modernization, 2011, 38 (1): 19-23.
［6］程淑红, 张仕军, 李雷华, 等. 基于鱼群运动特征和XGBoost的异常水质监测［J］. 计量学报, 2018, 39 (4): 572-577.
Cheng S H, Zhang S J, Li L H, et al. A water quality monitoring based on fish movement characteristics and XGBoost［J］. Acta Metrologica Sinica, 2018, 39 (4): 572-577.
［7］Henriques J F, Caseiro R, Martins P, et al. High-speed tracking with kernelized correlation filters［J］. IEEE Transactions on Pattern Analysis and Machine Intelligen-ce, 2015, 37 (3): 583-596.
［8］Danelljan M, Hager G, Khan F S, et al. Accurate scale estimation for robustvisual tracking［J］. British Machine Vision Conference, 2014, 65: 1-65. 11.
［9］Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector［C］//European conference on compu-ter vision. Springer, Cham, 2016.
［10］Howard A G, Zhu M, Chen B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications ［EB/OL］.https: //arxiv. org/abs/1704. 04861, 2017-04-17.
［11］Danelljan M, Hger G, Khan F S, et al. Accurate Scale Estimation for Robust Visual Tracking［C］// British Machine Vision Conference, 2014.
［12］Alexandrov G N, Dardeer M M. Long distance transmi-ssion line with controlled shunt reactors (CSRT)［C］// IEEE. 12th International Middle-East Power System Conference.2008: 294-299.
［13］Henriques J F, Caseiro R, Martins P, et al. High-speed tracking with kernelized correlation filters［J］. IEEE Transactions on Pattern Analysis and Machine Intellig-ence, 2015, 37 (3): 583-596.
［14］Bolme D S, Beveridge J R, Draper B A, et al. Visual object tracking using adaptive correlation filters［C］//IEEE. 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2010: 2544-2550.
［15］Kalal Z, Mikolajczyk K, Matas J. Tracking-learning-detection［J］. IEEE transactions on pattern analysis and machine intelligence, 2012, 34 (7): 1409.
［16］Grabner H, Grabner M, Bischof H. Real-time tracking via on-line boosting［C］//British Machine Vision Association. Proceedings of the 2006 British Machine vision conference. Edinburgy, United kingaom, 2006: 47-56.
［17］于德鑫, 曹晓杰, 杨敏, 等. 基于Kalman滤波与样本加权的压缩感知跟踪算法［J］. 智能计算机与应用, 2019, (5): 1-5.
Yu D X, Cao X J, Yang M, et al. Compressed tracking based on kalman filter and sample weighting［J］. Intelligent Computer and Applications, 2019,(5): 1-5.