Water Quality Monitoring Based on Fish Movement Behavior and G-RepVGG
-
摘要: 水质恶化会直接造成水产养殖产量下降，严重时会导致水产动物大量死亡，给养殖企业造成严重经济损失。因此对水产养殖中水质参数进行实时监测具有重要意义。本文以斑石鲷为研究对象，提出了一种基于鱼类行为的水质监测方法。该方法通过摄像机拍摄到的图像数据就可以非侵入地完成水质参数的实时监测，避免了安装复杂设备、对鱼类行为进行量化等繁琐过程。为了增加推理速度和降低模型参数量，通过将RepVGG block与GhostNet相结合构建了G-RepVGG模型，使该模型更适用于移动设备的部署。提出了计算量较少、推理速度快、更适合水质快速监测的Cheap Ghost操作和计算量大、精确率高、更适合水质的精确监测Expensive Ghoost操作。由于多分支网络适合进行训练但是在推理速度上低于单分支网络，因此通过模型重参数化首先将卷积层以及批归一化(Batch normalization, BN)层合并，随后再将3路卷积合并为1路，大大降低模型参数量、提高了模型推理速度，使模型更加适用于移动设备的推理。结果表明：使用Cheap Ghost操作的G-RepVGG在测试集中准确率达到96.21%,图像处理速度达到442.27 f/s,使用Expensive Ghost操作的G-RepVGG模型在测试集中准确率达到97.63%,图像处理速度达到349.42 f/s,从而在保证较高精度的前提下依旧具有较高的推理速度，在多个数据集中测试具有较好的鲁棒性。Abstract: Waterin aquaculture is a necessary place for aquatic animals to survive and live. The deterioration of water quality will directly lead to the decline of aquaculture production, and in severe cases, it will lead to the death of a large number of aquatic organisms and cause serious economic losses to aquaculture enterprises.Therefore, the real-time monitoring of water quality parameters in aquaculture is of great significance.A method for water quality monitoring based on fish behavior was proposed with Oplegnathus punctatus as research object.The method can non-invasively complete the real-time monitoring of water quality parameters through the image data captured by the camera, avoiding the tedious installation of complex equipment and the quantification of fish behavior.To increase the inference speed and reduce the amount of model parameters, this method combined RepVGG block with GhostNet.Aiming at the problems of rapid water quality monitoring and accurate water quality monitoring, the Cheap Ghost operation and the Expensive Ghost operation were proposed.Finally, the three branches were merged through model reparameterization, which greatly reduced the amount of model parameters and improved the model inference speed.The results showed that the G-RepVGG operated by Cheap Ghost achieved an accuracy of 96.21% in the test set and can infer 442.27 images per second. The G-RepVGG model operated with Expensive Ghost achieved 97.63% accuracy in the test set and can infer 349.42 images per second. Therefore, it still had a high inference speed under the premise of ensuring high accuracy, and had better robustness in testing in multiple data sets. The research result can quickly and accurately monitor water quality, detect water quality deterioration in time, and reduce losses caused by water quality deterioration, providing ideas and methods for water quality monitoring.
-
-
[1] OPPEDAL F,DEMPSTER T,STIEN L H.Environmental drivers of Atlantic salmon behaviour in sea-cages:a review[J].Aquaculture,2011,311(1-4):1-18.
[2] SEAGAR J,MILN E,IA N,et al.Effects of short-term oxygen depletion on fish[J].Environmental Toxicology&Chemistry,2000,19(12):2937-2942.
[3] JESÚS C V,FRANCISCO-JAVIER M L,BENJAMÍN G G.Oxygen consumption and ventilatory frequency responses to gradual hypoxia in common dentex (Dentex dentex):basis for suitable oxygen level estimations[J].Aquaculture,2006,256(1-4):542-551.
[4] YUAN F,HUANG Y F,CHEN X,et al.A biological sensor system using computer vision for water quality monitoring[J].IEEE Access,2018.DOI: 10.1109/ACCESS.2018.2876336.
[5] XIAO G,FENG M,CHENG Z B,et al.Water quality monitoring using abnormal tail-beat frequency of crucian carp[J].Ecotoxicology&Environmental Safety,2015,111(1):185-191.
[6] 彭红梅.基于生物水质监测的鱼体运动状态检测系统研究[D].西安:西安邮电大学,2017.PENG Hongmei.Research on fish movement state detection system based on biological water quality monitoring[D].Xi'an:Xi'an University of Posts and Telecommunications,2017.(in Chinese) [7] 王丽娜.基于动态图像理解的水质生物监测系统研究与应用[D].杭州:电子科技大学,2011.WANG Li'na.Research and application of water quality biological monitoring system based on dynamic image understanding[D].Hangzhou:University of Electronic Science and Technology,2011.(in Chinese) [8] SUN M,HASSAN S G,LI D.Models for estimating feed intake in aquaculture:a review[J].Computers&Electronics in Agriculture,2016,127(Complete):425-438.
[9] MARTINS C I M,GALHARDO L,NOBLE C,et al.Behavioural indicators of welfare in farmed fish[J].Fish Physiology and Biochemistry,2012,38(1):17-41.
[10] 颜波,石平.基于物联网的水产养殖智能化监控系统[J].农业机械学报,2014,45(1):259-265.YAN Bo,SHI Ping.Intelligent monitoring system for aquaculture based on Internet of things[J].Transactions of the Chinese Society for Agricultural Machinery,2014,45(1):259-265.(in Chinese) [11] YE Z,ZHAO J,HAN Z,et al.Behavioral characteristics and statistics-based imaging techniques in the assessment and optimization of Tilapia feeding in a recirculating aquaculture system[J].Transactions of the ASABE,2016,59(1):345-355.
[12] 孙龙清,刘婷,陈帅华,等.多目标鱼体对象提议检测算法研究[J].农业机械学报,2019,50(12):260-267.SUN Longqing,LIU Ting,CHEN Shuaihua,et al.Research on multi-target fish object proposal detection algorithm[J].Transactions of the Chinese Society for Agricultural Machinery,2019,50(12):260-267.(in Chinese) [13] LI Y,LEE J M,CHON T S,et al.Analysis of movement behavior of Zebrafish (Danio rerio) under chemical stress using hidden markov model[J].Modern Physics Letters B,2013,27(2):1350014.
[14] LIU Y,LEES H,CHON T S,et al.Analysis of behavioral changes of zebrafish (Danio rerio) in response to formaldehyde using self-organizing map and a hidden Markov model[J].Ecological Modelling,2011,222(14):2191-2201.
[15] LIAO Y,XU J Y,WANG W J.A method of water quality assessment based on biomonitoring and multiclass support vector machine[J].Procedia Environmental Sciences,2011,10(part-PA):451-457.
[16] AVIDAN S.Support vector tracking[J].IEEE Trans.Pattern Anal.Mach.Intell.,2004,26(8):1064-1072.
[17] JOHNSON A E,HEBERT M.Using spin-image for efficient multiply model recognition in cluttered scenes[J].IEEETransactions on Pattern Analysis and Machine Intelligence,1999,21:433-449.
[18] PAPADAKIS V M,PAPADAKIS I E,LAMPRIANIDOU F,et al.A computer-vision system and methodology for the analysis of fish behavior[J].Aquacultural Engineering,2012,46:53-59.
[19] HASLER C T,SUSKI C D,HANSON K C,et al.Effect of water temperature on laboratory swimming performance and natural activity levels of adult largemouth bass[J].Canadian Journal of Zoology,2009,87(7):589-596.
[20] LEE C G,FARRELL A P,LOTTO A G,et al.The effect of temperature on swimming performance and oxygen consumption in adult sockeye (Oncorhynchus nerka) and coho (O.kisutch) salmon stocks[J].Journal of Experimental Biology,2003,206(Pt 18):3239-3251.
[21] ZHOU C,XU D,CHEN L,et al.Evaluation of fish feeding intensity in aquaculture using a convolutional neural network and machine vision[J].Aquaculture,2019.DOI: 10.1016/j.aquaculture.2019.04.056.
[22] LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(7553):436-444.
[23] YANG L,YU H,CHENG Y,et al.A dual attention network based on efficientNet B2 for short-term fish school feeding behavior analysis in aquaculture[J].Computers and Electronics in Agriculture,2021,187:106316.
[24] WANG G,MUHAMMAD A,LIU C,et al.Automatic recognition of fish behavior with a fusion of RGB and optical flow data based on deep learning[J].Animals,2021,11(10):2774.
[25] DING X,ZHANG X,MA N,et al.RepVGG:making VGG-style convNets great again[C]∥CVPR,2021.
[26] HAN K,WANG Y,XU C,et al.GhostNets on heterogeneous devices via cheap operations[J].International Journal of Computer Vision,2022,130(4):1050-1069.
[27] HUANG J,LIU S,HASSAN S G,et al.A hybrid model for short-term dissolved oxygen content prediction[J].Computers and Electronics in Agriculture,2021,186:106216.
[28] LIU Z,LIN Y,CAO Y,et al.Swin transformer:hierarchical vision transformer using shifted windows[J].arXiv:2103.14030.
[29] HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]∥IEEE Conference on Computer Vision and Pattern Recognition,2016.
[30] HOWARD A,SANDLER M,CHU G,et al.Searching for MobileNetV3[C]∥Proceedings of the IEEE/CVF International Conference on Vision,2019:1314-1324.
[31] SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[J].arXiv:1409,1556,2014.
计量
- 文章访问数: 0
- HTML全文浏览量: 0
- PDF下载量: 0
下载:
