基于机器视觉的水下河蟹识别方法
|
摘要
为了探测河蟹在池塘中的数量及分布情况,为自动投饵船提供可靠的数据反馈,提出了基于机器视觉的水下河蟹识别方法。该方法通过在投饵船下方安装摄像头进行河蟹图像实时采集,针对水下光线衰减大、视野模糊等特点,采用优化的Retinex算法提高图像对比度,增强图像细节,修改基于深度卷积神经网络YOLO V3的输入输出,并采用自建的数据集对其进行训练,实现了对水下河蟹的高精度识别。实验所训练的YOLO V3模型在测试集上的平均精度均值达86. 42%,对水下河蟹识别的准确率为96. 65%,召回率为91. 30%。实验对比了多种目标检测算法,仅有YOLO V3在识别准确率和识别速率上均达到较高水平。在同一硬件平台上YOLO V3的识别速率为10. 67 f/s,优于其他算法,具有较高的实时性和应用价值。
In order to detect underwater crabs,a detection method based on machine vision was proposed,which can produce necessary feedback data on the number and distribution of crabs to automatic bait casting boat in real time so that the boat can cast baits precisely. An underwater camera with LED light installed at the bottom of the boat was used to capture images of crabs. These images taken under water were enhanced by optimized retinex filter firstly,which can make images clearer and enhance the details in images. Then,a dataset included original images captured underwater,captured from a laboratory and downloaded from webs was built. There were totally 3500 labelled original images in the dataset. The dataset was augmented and divided into training and test datasets. Finally,a deep convolution neural network,YOLO V3 was trained to detect the crabs by training dataset. The mean average precision of trained network reached 86. 42% on test dataset, the detection precision for underwater crabs was 96. 65% and the recall ratio was 91. 30%. Compared with crabs with big size,the crabs with small size were more difficult to be detected. Compared with other methods for object detection,YOLO V3 can reach a high level of both recognition precision and speed. The recognition speed of the proposed method was 10. 67 f/s,which was higher than that of other methods on the same hardware platform. Therefore,the proposed method was real time and had application values.
In order to detect underwater crabs,a detection method based on machine vision was proposed,which can produce necessary feedback data on the number and distribution of crabs to automatic bait casting boat in real time so that the boat can cast baits precisely. An underwater camera with LED light installed at the bottom of the boat was used to capture images of crabs. These images taken under water were enhanced by optimized retinex filter firstly,which can make images clearer and enhance the details in images. Then,a dataset included original images captured underwater,captured from a laboratory and downloaded from webs was built. There were totally 3500 labelled original images in the dataset. The dataset was augmented and divided into training and test datasets. Finally,a deep convolution neural network,YOLO V3 was trained to detect the crabs by training dataset. The mean average precision of trained network reached 86. 42% on test dataset, the detection precision for underwater crabs was 96. 65% and the recall ratio was 91. 30%. Compared with crabs with big size,the crabs with small size were more difficult to be detected. Compared with other methods for object detection,YOLO V3 can reach a high level of both recognition precision and speed. The recognition speed of the proposed method was 10. 67 f/s,which was higher than that of other methods on the same hardware platform. Therefore,the proposed method was real time and had application values.
引文
[1]孙月平,赵德安,洪剑青,等.河蟹养殖船载自动均匀投饵系统设计及效果试验[J].农业工程学报,2015,31(11):31-39.SUN Yueping,ZHAO Dean,HONG Jianqing,et al. Design of automatic and uniform feeding system carried by workboat and effect test for raising river crab[J]. Transactions of the CSAE,2015,31(11):31-39.(in Chinese)
[2]赵德安,罗吉,孙月平,等.河蟹养殖自动作业船导航控制系统设计与测试[J].农业工程学报,2016,32(11):181-188.ZHAO Dean,LUO Ji,SUN Yueping,et al. Design and experiment of navigation control system of automatic operation boat in river crab culture[J]. Transactions of the CSAE,2016,32(11):181-188.(in Chinese)
[3] JIANG Yu,LI Zhenbo,FANG Jingjing,et al. Automatic video tracking of Chinese mitten crabs based on the particle filter algorithm using a biologically constrained probe and resampling[J]. Computers and Electronics in Agriculture,2014,106:111-119.
[4]王勇平,聂余满,谢成军,等.基于机器视觉的养殖鱼群智能投饵系统设计与研究[J].仪表技术,2015(1):1-4.WANG Yongping,NIE Yuman,XIE Chengjun,et al. Design and research of intelligent feeding system for farmed fish based on machine vision[J]. Instrumentation Technology,2015(1):1-4.(in Chinese)
[5]胡利永,魏玉艳,郑堤,等.基于机器视觉技术的智能投饵方法研究[J].热带海洋学报,2015,34(4):90-95.HU Liyong,WEI Yuyan,ZHENG Di,et al. Research on intelligent bait casting method based on machine vision technology[J]. Journal of Tropical Oceanography,2015,34(4):90-95.(in Chinese)
[6]马国强,田云臣,李晓岚. K-均值聚类算法在海水背景石斑鱼彩色图像分割中的应用[J].计算机应用与软件,2016,33(5):192-195.MA Guoqiang,TIAN Yuncheng,LI Xiaolan. Application of K-means clustering algorithm in colour image segmentation of grouper in seawater background[J]. Computer Application and Software,2016,33(5):192-195.(in Chinese)
[7]熊俊涛,刘振,林睿,等.自然环境下树上绿色芒果的无人机视觉检测技术[J/OL].农业机械学报,2018,49(11):23-29.XIONG Juntao,LIU Zhen,LIN Rui,et al. Unmanned aerial vehicle vision detection technology of green mango on tree in natural environment[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(11):23-29. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20181103&flag=1&jour nal_id=jcsam. DOI:10. 6041/j.issn. 1000-1298. 2018. 11. 003.(in Chinese)
[8]杨洋,张亚兰,苗伟,等.基于卷积神经网络的玉米根茎精确识别与定位研究[J/OL].农业机械学报,2018,49(10):46-53.YANG Yang,ZHANG Yalan,MIAO Wei,et al. Accurate identification and location of corn rhizome based on faster R-CNN[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(10):46-53. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20181006&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2018. 10. 006.(in Chinese)
[9]郑一力,张露.基于迁移学习的卷积神经网络植物叶片图像识别方法[J/OL].农业机械学报,2018,49(增刊):354-359.ZHENG Yili,ZHANG Lu. Plant leaf image recognition method based on transfer learning with convolutional neural networks[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(Supp.):354-359. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=2018s047&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2018. S0. 047.(in Chinese)
[10]薛金林,闫嘉,范博文.多类农田障碍物卷积神经网络分类识别方法[J/OL].农业机械学报,2018,49(增刊):35-41.XUE Jinlin,YAN Jia,FAN Bowen. Classification and identification method of multiple kinds of farm obstacles based on convolutional neural network[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(Supp.):35-41. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=2018s006&flag=1&journal_id=jcsam.DOI:10. 6041/j. issn. 1000-1298. 2018. S0. 006.(in Chinese)
[11]姜红花,王鹏飞,张昭,等.基于卷积网络和哈希码的玉米田间杂草快速识别方法[J/OL].农业机械学报,2018,49(11):30-38.JIANG Honghua,WANG Pengfei,ZHANG Zhao,et al. Fast identification of field weeds based on deep convolutional network and binary Hash code[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(11):30-38. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20181104&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2018. 11. 004.(in Chinese)
[12]傅隆生,冯亚利,ELKAMIL T,等.基于卷积神经网络的田间多簇猕猴桃图像识别方法[J].农业工程学报,2018,34(2):205-211.FU Longsheng,FENG Yali,ELKAMIL T,et al. Image recognition method of multi-cluster kiwifruit in field based on convolutional neural networks[J]. Transactions of the CSAE,2018,34(2):205-211.(in Chinese)
[13]熊俊涛,刘振,汤林越,等.自然环境下绿色柑橘视觉检测技术研究[J/OL].农业机械学报,2018,49(4):45-52.XIONG Juntao,LIU Zhen,TANG Linyue,et al. Visual detection technology of green citrus under natural environment[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(4):45-52. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20180405&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2018. 04. 005.(in Chinese)
[14]薛月菊,黄宁,涂淑琴,等.未成熟芒果的改进YOLOv2识别方法[J].农业工程学报,2018,34(7):173-179.XUE Yueju,HUANG Ning,TU Shuqin,et al. Immature mango detection based on improved YOLOv2[J]. Transactions of the CSAE,2018,34(7):173-179.(in Chinese)
[15] XIAO Limei,LI Ce,WU Zonghe,et al. An enhancement method for X-ray image via fuzzy noise removal and homomorphic filtering[J]. Neurocomputing,2016,195:56-64.
[16] GOMES D L,PAIVA A C,SILVA A C,et al. Augmented visualization using homomorphic filtering and Haar-based natural markers for power systems substations[J]. Computers in Industry,2018,97:67-75.
[17] ZHANG Shu,WANG Ting,DONG Junyu,et al. Underwater image enhancement via extended multi-scale Retinex[J].Neurocomputing,2017,245:1-9.
[18] XIE Kun,PAN Wei,XU Suxia. An underwater image enhancement algorithm for environment recognition and robot navigation[J]. Robotics,2018,7(1):14.
[19] REDMON J,DIVVALA S,GIRSHICK R,et al. You only look once:unified,real-time object detection[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2016:779-788.
[20] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Deep residual learning for image recognition[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2016:770-778.
[21] REN Shaoping,HE Kaiming,GIRSHICK R,et al. Faster R-CNN:towards real-time object detection with region proposal networks[C]∥Advances in Neural Information Processing Systems,2015:91-99.
[22] WANG Yiran,ZHU Xiaolin,WU Bo. Automatic detection of individual oil palm trees from UAV images using HOG features and an SVM classifier[J]. International Journal of Remote Sensing,2018,39:1-15.
[2]赵德安,罗吉,孙月平,等.河蟹养殖自动作业船导航控制系统设计与测试[J].农业工程学报,2016,32(11):181-188.ZHAO Dean,LUO Ji,SUN Yueping,et al. Design and experiment of navigation control system of automatic operation boat in river crab culture[J]. Transactions of the CSAE,2016,32(11):181-188.(in Chinese)
[3] JIANG Yu,LI Zhenbo,FANG Jingjing,et al. Automatic video tracking of Chinese mitten crabs based on the particle filter algorithm using a biologically constrained probe and resampling[J]. Computers and Electronics in Agriculture,2014,106:111-119.
[4]王勇平,聂余满,谢成军,等.基于机器视觉的养殖鱼群智能投饵系统设计与研究[J].仪表技术,2015(1):1-4.WANG Yongping,NIE Yuman,XIE Chengjun,et al. Design and research of intelligent feeding system for farmed fish based on machine vision[J]. Instrumentation Technology,2015(1):1-4.(in Chinese)
[5]胡利永,魏玉艳,郑堤,等.基于机器视觉技术的智能投饵方法研究[J].热带海洋学报,2015,34(4):90-95.HU Liyong,WEI Yuyan,ZHENG Di,et al. Research on intelligent bait casting method based on machine vision technology[J]. Journal of Tropical Oceanography,2015,34(4):90-95.(in Chinese)
[6]马国强,田云臣,李晓岚. K-均值聚类算法在海水背景石斑鱼彩色图像分割中的应用[J].计算机应用与软件,2016,33(5):192-195.MA Guoqiang,TIAN Yuncheng,LI Xiaolan. Application of K-means clustering algorithm in colour image segmentation of grouper in seawater background[J]. Computer Application and Software,2016,33(5):192-195.(in Chinese)
[7]熊俊涛,刘振,林睿,等.自然环境下树上绿色芒果的无人机视觉检测技术[J/OL].农业机械学报,2018,49(11):23-29.XIONG Juntao,LIU Zhen,LIN Rui,et al. Unmanned aerial vehicle vision detection technology of green mango on tree in natural environment[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(11):23-29. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20181103&flag=1&jour nal_id=jcsam. DOI:10. 6041/j.issn. 1000-1298. 2018. 11. 003.(in Chinese)
[8]杨洋,张亚兰,苗伟,等.基于卷积神经网络的玉米根茎精确识别与定位研究[J/OL].农业机械学报,2018,49(10):46-53.YANG Yang,ZHANG Yalan,MIAO Wei,et al. Accurate identification and location of corn rhizome based on faster R-CNN[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(10):46-53. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20181006&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2018. 10. 006.(in Chinese)
[9]郑一力,张露.基于迁移学习的卷积神经网络植物叶片图像识别方法[J/OL].农业机械学报,2018,49(增刊):354-359.ZHENG Yili,ZHANG Lu. Plant leaf image recognition method based on transfer learning with convolutional neural networks[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(Supp.):354-359. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=2018s047&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2018. S0. 047.(in Chinese)
[10]薛金林,闫嘉,范博文.多类农田障碍物卷积神经网络分类识别方法[J/OL].农业机械学报,2018,49(增刊):35-41.XUE Jinlin,YAN Jia,FAN Bowen. Classification and identification method of multiple kinds of farm obstacles based on convolutional neural network[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(Supp.):35-41. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=2018s006&flag=1&journal_id=jcsam.DOI:10. 6041/j. issn. 1000-1298. 2018. S0. 006.(in Chinese)
[11]姜红花,王鹏飞,张昭,等.基于卷积网络和哈希码的玉米田间杂草快速识别方法[J/OL].农业机械学报,2018,49(11):30-38.JIANG Honghua,WANG Pengfei,ZHANG Zhao,et al. Fast identification of field weeds based on deep convolutional network and binary Hash code[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(11):30-38. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20181104&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2018. 11. 004.(in Chinese)
[12]傅隆生,冯亚利,ELKAMIL T,等.基于卷积神经网络的田间多簇猕猴桃图像识别方法[J].农业工程学报,2018,34(2):205-211.FU Longsheng,FENG Yali,ELKAMIL T,et al. Image recognition method of multi-cluster kiwifruit in field based on convolutional neural networks[J]. Transactions of the CSAE,2018,34(2):205-211.(in Chinese)
[13]熊俊涛,刘振,汤林越,等.自然环境下绿色柑橘视觉检测技术研究[J/OL].农业机械学报,2018,49(4):45-52.XIONG Juntao,LIU Zhen,TANG Linyue,et al. Visual detection technology of green citrus under natural environment[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2018,49(4):45-52. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? file_no=20180405&flag=1&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2018. 04. 005.(in Chinese)
[14]薛月菊,黄宁,涂淑琴,等.未成熟芒果的改进YOLOv2识别方法[J].农业工程学报,2018,34(7):173-179.XUE Yueju,HUANG Ning,TU Shuqin,et al. Immature mango detection based on improved YOLOv2[J]. Transactions of the CSAE,2018,34(7):173-179.(in Chinese)
[15] XIAO Limei,LI Ce,WU Zonghe,et al. An enhancement method for X-ray image via fuzzy noise removal and homomorphic filtering[J]. Neurocomputing,2016,195:56-64.
[16] GOMES D L,PAIVA A C,SILVA A C,et al. Augmented visualization using homomorphic filtering and Haar-based natural markers for power systems substations[J]. Computers in Industry,2018,97:67-75.
[17] ZHANG Shu,WANG Ting,DONG Junyu,et al. Underwater image enhancement via extended multi-scale Retinex[J].Neurocomputing,2017,245:1-9.
[18] XIE Kun,PAN Wei,XU Suxia. An underwater image enhancement algorithm for environment recognition and robot navigation[J]. Robotics,2018,7(1):14.
[19] REDMON J,DIVVALA S,GIRSHICK R,et al. You only look once:unified,real-time object detection[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2016:779-788.
[20] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Deep residual learning for image recognition[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2016:770-778.
[21] REN Shaoping,HE Kaiming,GIRSHICK R,et al. Faster R-CNN:towards real-time object detection with region proposal networks[C]∥Advances in Neural Information Processing Systems,2015:91-99.
[22] WANG Yiran,ZHU Xiaolin,WU Bo. Automatic detection of individual oil palm trees from UAV images using HOG features and an SVM classifier[J]. International Journal of Remote Sensing,2018,39:1-15.