LIANG Zhaodong,XIONG Xingguo,XU Dongpo,et al.An automatic method for freshwater fish species classification using shape and texture features[J].Journal of Nanjing Agricultural University,2021,44(3):576-585.[doi:10.7685/jnau.202009023]





An automatic method for freshwater fish species classification using shape and texture features
梁钊董1 熊兴国1 徐东坡2 陆明洲1 沈明霞1 张婉平3 童奇烈3
1. 南京农业大学人工智能学院, 江苏 南京 210031;
2. 中国水产科学研究院淡水渔业研究中心, 江苏 无锡 214128;
3. 杭州市渔政渔港渔船监督管理总站, 浙江 杭州 310008
LIANG Zhaodong1 XIONG Xingguo1 XU Dongpo2 LU Mingzhou1 SHEN Mingxia1 ZHANG Wanping3 TONG Qilie3
1. College of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210031, China;
2. Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences, Wuxi 214128, China;
3. Supervision and Administration Station of Hangzhou for Fishing Industry, Vessels, and Ports, Hangzhou 310008, China
machine visionclassification of freshwater fish speciestexturefeature dimension reductionrandom forest
[目的] 本文利用自行搭建的淡水鱼图像采集装置在淡水鱼捕捞现场采集鱼体图像,在图像数据集上提出一种实用的淡水鱼自动识别分类方法。[方法] 对14类淡水鱼图像进行预处理、自模板匹配操作,分割出完整的鱼体前景图像;提取鱼体全身、去尾鱼体及鱼尾的16维形状特征,利用灰度-梯度共生矩阵(GLGCM)、局部二值模式(LBP)及Gabor变换提取鱼体纹理特征,采用主成分分析法(PCA)分别筛选GLGCM纹理+形状、LBP纹理+形状和Gabor纹理+形状特征累积贡献率超过85%的特征组合。将降维前后特征集的70%和30%分别作为训练集和验证集,利用朴素贝叶斯、K近邻(KNN)、线性回归、决策树、随机森林、支持向量机(SVM)、梯度提升决策树(GBDT)7种机器学习方法训练淡水鱼品种分类器,并利用验证集数据分析对比各分类器的性能。[结果] 鱼体前景图像分割算法测试结果表明,本文提出的自模板匹配方法可在不建立大规模模板库的前提下,以99.79%的正确率分割鱼体图像。分类器性能验证及对比结果表明,基于GLGCM纹理+形状特征的随机森林分类器的淡水鱼识别精度最高,降维获得的5维GLGCM纹理+形状特征向量识别14种淡水鱼的正确率达到99.52%。[结论] 提出的自模板匹配方法可以在不构建庞大模板库的前提下实现鱼体前景区域的分割,基于筛选得到的GLGCM纹理+形状特征的随机森林分类器可用于自动识别淡水鱼品种。
[Objectives] An automatic method for freshwater fish species classification was proposed in this paper,based on the fish images collected by using a homemade fish image acquisition device. [Methods] A self-template matching method was developed to segment the fish body region from the binary image,which was obtained after the image pre-processing operations. Then the shape and texture features of the fish body were extracted,where the latter were obtained using three methods,including gray level-gradient co-occurrence matrix(GLGCM),local binary pattern(LBP),and Gabor filter. Principal component analysis(PCA) was carried out to transform a high dimensional feature set of GLGCM texture and shape,LBP texture and shape,or Gabor texture and shape into smaller ones that contained at least 85% of the information in the large feature set. 70% and 30% of the feature sets were used to train and validate species classification models established using seven machine learning methods[Naive Bayes,K-neares neighbors(KNN),linear regression,randow forest,decision tree,support vector machine(SVM),gradient boosting decision tree(GBDT)]. [Results] Fish body segmentation test indicated that the developed self-template matching method could segment the region on interest with a correct rate of 99.79%,which was achieved without the need of an extra fish body template library. Classification models validation results indicated that the random forest classifier using reduced dimensional GLGCM texture and shape feature performed best,and an average accuracy of 99.52% was achieved when it was used to classify all the 14 freshwater fish species. [Conclusions] The self-template matching method developed in this study could extract fish body region well without the need of an extra fish body template library. The reduced dimensional GLGCM texture and shape feature set was suitable for identifying the species of freshwater fish.


[1] 王坤殿. 淡水鱼种类识别与重量在线检测方法研究及装备设计[D]. 武汉:华中农业大学,2015. Wang K D. Study of Freshwater fish species identification and on-line weight detection and equipment design[D]. Wuhan:Huazhong Agricultural University,2015(in Chinese with English abstract).
[2] 李玲,宗力,王玖玖,等. 大宗淡水鱼加工前处理技术和装备的研究现状及方向[J]. 渔业现代化,2010,37(5):43-46,71. Li L,Zong L,Wang J J,et al. Research status and development trend of massive freshwater fish pre-treatment processing technology and equipment[J]. Fishery Modernization,2010,37(5):43-46,71(in Chinese with English abstract).
[3] 张志强. 基于机器视觉的淡水鱼品种识别及重量预测研究[D]. 武汉:华中农业大学,2011. Zhang Z Q. Research on freshwater fish species recognition and weight prediction based on machine vision[D]. Wuhan:Huazhong Agricultural University,2011(in Chinese with English abstract).
[4] White D J,Svellingen C,Strachan N J C. Automated measurement of species and length of fish by computer vision[J]. Fisheries Research,2006,80(2/3):203-210.
[5] Alsmadi M,Omar K B,Noah S A,et al. A hybrid memetic algorithm with back-propagation classifier for fish classification based on robust features extraction from PLGF and shape measurements[J]. Information Technology Journal,2011,10(5):944-954.
[6] Hu J,Li D L,Duan Q L,et al. Fish species classification by color,texture and multi-class support vector machine using computer vision[J]. Computers and Electronics in Agriculture,2012,88:133-140.
[7] 吴一全,殷骏,戴一冕,等. 基于蜂群优化多核支持向量机的淡水鱼种类识别[J]. 农业工程学报,2014,30(16):312-319. Wu Y Q,Yin J,Dai Y M,et al. Identification method of freshwater fish species using multi-kernel support vector machine with bee colony optimization[J]. Transactions of the Chinese Society of Agricultural Engineering,2014,30(16):312-319(in Chinese with English abstract).
[8] 涂兵,王锦萍,王思成,等. 基于背部轮廓相关系数算法的淡水鱼种类识别研究[J]. 计算机工程与应用,2016,52(16):162-166. Tu B,Wang J P,Wang S C,et al. Research on identification of freshwater fish species based on fish back contour correlation coefficient[J]. Computer Engineering and Applications,2016,52(16):162-166(in Chinese with English abstract).
[9] 谢忠红,郭小清,程碧云,等. 基于多特征的淡水鱼种类识别研究[J]. 扬州大学学报(农业与生命科学版),2016,37(3):71-77. Xie Z H,Guo X Q,Cheng B Y,et al. Species recognition of fishes based on multiple features[J]. Journal of Yangzhou University(Agricultural and Life Science Edition),2016,37(3):71-77(in Chinese with English abstract).
[10] Mala C,Sridevi M. Multilevel threshold selection for image segmentation using soft computing techniques[J]. Soft Computing,2016,20(5):1793-1810.
[11] Mahalakshmi T,Muthaiah R,Swaminathan P. An overview of template matching technique in image processing[J]. Research Journal of Applied Sciences,Engineering and Technology,2012,4(24):5469-5473.
[12] Suzuki S,Be K. Topological structural analysis of digitized binary images by border following[J]. Computer Vision,Graphics,and Image Processing,1985,30(1):32-46.
[13] Andalo F A,Miranda P A V,Torres R D S,et al. Shape feature extraction and description based on tensor scale[J]. Pattern Recognition,2010,43(1):26-36.
[14] Mohanaiah P,Sathyanarayana P,Gurukumar L. Image texture feature extraction using GLCM approach[J]. International Journal of Scientific and Research Publications,2013,3(5):1-5.
[15] Hubara I,Courbariaux M,Soudry D,et al. Quantized neural network:training neural networks with low precision weights and activations[J]. Journal of Machine Learning Research,2017,18:1-30.
[16] 洪继光. 灰度-梯度共生矩阵纹理分析方法[J]. 自动化学报,1984,10(1):22-25. Hong J G. Gray level-gradient cooccurrence matrix texture analysis method[J]. Acta Automatica Sinica,1984,10(1):22-25(in Chinese with English abstract).
[17] 秦立峰,何东健,宋怀波. 词袋特征PCA多子空间自适应融合的黄瓜病害识别[J]. 农业工程学报,2018,34(8):200-205. Qin L F,He D J,Song H B. Bag of words feature multi-PCA subspace adaptive fusion for cucumber diseases identification[J]. Transactions of the Chinese Society of Agricultural Engineering,2018,34(8):200-205(in Chinese with English abstract).
[18] 丁天华,卢伟,张超,等. 基于MUSIC功率谱和CPNN的鸡蛋散黄无损检测方法[J]. 南京农业大学学报,2015,38(6):1009-1015. DOI:10.7685/j.issn.1000-2030.2015.06.021. Ding T H,Lu W,Zhang C,et al. A nondestructive detection method of scattered eggs based on MUSIC power spectrum and CPNN[J]. Journal of Nanjing Agricultural University,2015,38(6):1009-1015(in Chinese with English abstract).
[19] Ahonen T,Hadid A,Pietik?inen M. Face description with local binary patterns:application to face recognition[C].//IEEE Transactions on Pattern Analysis and Machine Intelligence,2006:2037-2041.
[20] 高梓瑞. Gabor滤波器在纹理分析中的应用研究[D]. 武汉:武汉理工大学,2012. Gao Z R. The research of the Gabor filter in the texture analysis applications[D]. Wuhan:Wuhan University of Technology,2012(in Chinese with English abstract).
[21] AlBeladi A A,Muqaibel A H. Evaluating compressive sensing algorithms in through-the-wall radar via F1-score[J]. International Journal of Signal and Imaging Systems Engineering,2018,11(3):164-171.
[22] 米爱中,张盼. 一种基于混淆矩阵的分类器选择方法[J]. 河南理工大学学报(自然科学版),2017,36(2):116-121. Mi A Z,Zhang P. A method of classifier selection based on confusion matrix[J]. Journal of Henan Polytechnic University(Natural Science Edition),2017,36(2):116-121(in Chinese with English abstract).


 RAO Hong-hui,JI Chang-ying.Study on spray control to aim toward crops rows based on machine vision[J].Journal of Nanjing Agricultural University,2007,30(3):120.[doi:10.7685/j.issn.1000-2030.2007.01.024]
 CHEN Shijin,DING Dong,LI Bo,et al.Research on relationship between beef connective tissue features and tenderness by computer vision technology[J].Journal of Nanjing Agricultural University,2016,39(3):865.[doi:10.7685/jnau.201511016]
 ZHANG Chi,SHEN Mingxia,LIU Longshen,et al.Newborn piglets recognition method based on machine vision[J].Journal of Nanjing Agricultural University,2017,40(3):169.[doi:10.7685/jnau.201602017]
 GAO Qi,LI Yinian,JIN Dian,et al.Measurement method for quantitative characteristics of maize based on single image[J].Journal of Nanjing Agricultural University,2018,41(3):562.[doi:10.7685/jnau.201706028]
 QI Chao,XU Jiaqi,LIU Chao,et al.Automatic classification of chicken carcass weight based on machine vision and machine learning technology[J].Journal of Nanjing Agricultural University,2019,42(3):551.[doi:10.7685/jnau.201808013]
 DING Jing,SHEN Mingxia,LIU Longshen,et al.Automatic recognition of diarrhea in weaned piglets based on machine vision[J].Journal of Nanjing Agricultural University,2020,43(3):969.[doi:10.7685/jnau.201908003]


更新日期/Last Update: 1900-01-01