Automate fry counting using computer vision and multi-class least squares support vector machine

CORC > 海洋研究所 > 中国科学院海洋研究所 > 海洋生态与环境科学重点实验室

	Automate fry counting using computer vision and multi-class least squares support vector machine
	Fan, Liangzhong 1; Liu, Ying2 ; Liu, Y
刊名	AQUACULTURE
	2013-03-04
卷号	380 页码:91-98
关键词	Back propagation neural network Least squares support vector machine Computer vision Fry counting
ISSN号	0044-8486
通讯作者	Liu, Y
中文摘要	In this paper, an approach based on geometric features to count overlapping fry fish is presented. Back propagation neural network (BPNN) and least squares support vector machine (LS-SVM) were used to construct classification models. 19 video clips with fish numbers varying from 10 to 100 were captured by a computer vision system. A total of 600 sub-images with overlapping fish were randomly selected, 300 images were used as a training set to create a calibration model, and remaining images were used to verify the model. 7 geometric features (area, perimeter, convex area, bounding box width, bounding box height, skeleton length, endpoint number) were obtained from the overlapping fish images. Results indicate that the best performance with about 98.73% of the average counting accuracy rate is achieved by LS-SVM model, which is better than the performance of BPNN model. The combined multiple geometric features coupled with an LS-SVM classifier is a highly accurate way for fry fish counting. (C) 2012 Published by Elsevier B.V.
英文摘要	In this paper, an approach based on geometric features to count overlapping fry fish is presented. Back propagation neural network (BPNN) and least squares support vector machine (LS-SVM) were used to construct classification models. 19 video clips with fish numbers varying from 10 to 100 were captured by a computer vision system. A total of 600 sub-images with overlapping fish were randomly selected, 300 images were used as a training set to create a calibration model, and remaining images were used to verify the model. 7 geometric features (area, perimeter, convex area, bounding box width, bounding box height, skeleton length, endpoint number) were obtained from the overlapping fish images. Results indicate that the best performance with about 98.73% of the average counting accuracy rate is achieved by LS-SVM model, which is better than the performance of BPNN model. The combined multiple geometric features coupled with an LS-SVM classifier is a highly accurate way for fry fish counting. (C) 2012 Published by Elsevier B.V.
学科主题	Fisheries ; Marine & Freshwater Biology
WOS标题词	Science & Technology ; Life Sciences & Biomedicine
类目[WOS]	Fisheries ; Marine & Freshwater Biology
研究领域[WOS]	Fisheries ; Marine & Freshwater Biology
关键词[WOS]	FISH COUNTER ; CLASSIFICATION ; SALMON
收录类别	SCI
原文出处	10.1016/j.aquaculture.2012.10.016
语种	英语
WOS记录号	WOS:000314642900015
公开日期	2014-07-17
内容类型	期刊论文
源URL	[http://ir.qdio.ac.cn/handle/337002/16460]
专题	海洋研究所_海洋生态与环境科学重点实验室海洋研究所_海洋生物技术研发中心
通讯作者	Liu, Y
作者单位	1.Zhejiang Univ, Ningbo Inst Technol, Ningbo 315100, Zhejiang, Peoples R China 2.Chinese Acad Sci, Inst Oceanol, Qingdao 266071, Peoples R China
推荐引用方式 GB/T 7714	Fan, Liangzhong,Liu, Ying,Liu, Y. Automate fry counting using computer vision and multi-class least squares support vector machine[J]. AQUACULTURE,2013,380:91-98.
APA	Fan, Liangzhong,Liu, Ying,&Liu, Y.(2013).Automate fry counting using computer vision and multi-class least squares support vector machine.AQUACULTURE,380,91-98.
MLA	Fan, Liangzhong,et al."Automate fry counting using computer vision and multi-class least squares support vector machine".AQUACULTURE 380(2013):91-98.