Underwater sound and video observations were made at noon, sunset, and midnight in sand, gravel, and boulder habitat in the Stellwagen Bank National Marine Sanctuary, Gulf of Maine, USA in October 2001 using a remotel...Underwater sound and video observations were made at noon, sunset, and midnight in sand, gravel, and boulder habitat in the Stellwagen Bank National Marine Sanctuary, Gulf of Maine, USA in October 2001 using a remotely operated vehicle （ROV）. Seventeen species of fish and squid were observed with clear habitat and time differences. Observations of feeding behavior, disturbance behavior, and both interspecific and intraspecific interactions provided numerous opportunities for potential sound production; however, sounds were recorded only during a single dive. Although high noise levels generated by the ROV and support ship may have masked some sounds, we conclude that fish sound production in the Gulf of Maine during the fall is uncommon. The recorded fish sounds are tentatively attributed to the cusk Brosme brosme. Cusk sounds consisted variously of isolated thumps, widely spaced thump trains, drumrolls, and their combinations. Frequency peaks were observed at 188, 539, and 1195 Hz. Use of a remotely operated vehicle （ROV） as a passive acoustic observation platform was problematic due to high ROV self-noise and the ROV＇s inability to maintain a fixed position on the bottom without thruster power. Some fishes were clearly also disturbed by ROV noise, indicating a potential ROV sampling bias. Based on our observations, we suggest that new instruments incorporating both optic and passive acoustic technologies are needed to provide better tools for in situ behavioral studies ofcusk and other fishes [Current Zoology 56 （1）： 90-99 2010].展开更多
Accurate and rapid detection of fish behaviors is critical to perceive health and welfare by allowing farmers to make informed management deci-sions about recirculating the aquaculture system while decreasing labor.Th...Accurate and rapid detection of fish behaviors is critical to perceive health and welfare by allowing farmers to make informed management deci-sions about recirculating the aquaculture system while decreasing labor.The classic detection approach involves placing sensors on the skin or body of the fish,which may interfere with typical behavior and welfare.The progress of deep learning and computer vision technologies opens up new opportunities to understand the biological basis of this behavior and precisely quantify behaviors that contribute to achieving accurate management in precision farming and higher production efficacy.This study develops an intelligent fish behavior classification using modified invasive weed optimization with an ensemble fusion(IFBC-MIWOEF)model.The presented IFBC-MIWOEF model focuses on identifying the distinct kinds of fish behavior classification.To accomplish this,the IFBC-MIWOEF model designs an ensemble of Deep Learning(DL)based fusion models such as VGG-19,DenseNet,and Effi-cientNet models for fish behavior classification.In addition,the hyperparam-eter tuning of the DL models is carried out using the MIWO algorithm,which is derived from the concepts of oppositional-based learning(OBL)and the IWO algorithm.Finally,the softmax(SM)layer at the end of the DL model categorizes the input into distinct fish behavior classes.The experimental validation of the IFBC-MIWOEF model is tested using fish videos,and the results are examined under distinct aspects.An Extensive comparative study pointed out the improved outcomes of the IFBC-MIWOEF model over recent approaches.展开更多
In this study,fish behavior and fish injury at different operating conditions are investigated via numerical simulation to evaluate the fish-friendliness of an axial pump that comprises an inlet pipe,a rotor with six ...In this study,fish behavior and fish injury at different operating conditions are investigated via numerical simulation to evaluate the fish-friendliness of an axial pump that comprises an inlet pipe,a rotor with six blades,a stator with eight vanes,and an outlet pipe.To precisely obtain the flow field when the fish passes through the axial pump,a hybrid large eddy simulation and immersed boundary method is adopted with the full consideration of the fluid-structure interaction comprehensively.The results indicate that the collision between the fish and the wall of flow components in the axial pump is concentrated near the inlet of the rotor,which results in the complexity of the fish trajectory,especially under the large flow rate condition.It is noted that the fish is likely to move in the reverse direction of the main flow after the impact with the rotor blade if the flow rate coefficient is too large,which increases the possibility of collision between the fish and the rotor blade.It is also indicated that the primary factor affecting the strike injury on the fish when it passes through the axial pump is the strike between the fish and the leading edge of the rotor blade.In addition,the strike injury becomes more significant as the flow rate coefficient increases.Furthermore,the results demonstrate that the fish may simultaneously suffer from strike,pressure,and shear stress injuries,once the collision between the fish and the wall of flow components occurs in the axial pump,thus aggravating the combined damage on the fish.Based on these results,it is recommended that hydraulic machinery should not be operated at large flow rates during fish migration from the view of fish-friendliness.展开更多
Dissolved oxygen(DO)content is an important index of river water quality.Water quality sensors have been used in China for urban river water monitoring and DO content prediction.However,water quality sensors are expen...Dissolved oxygen(DO)content is an important index of river water quality.Water quality sensors have been used in China for urban river water monitoring and DO content prediction.However,water quality sensors are expensive and difficult to maintain,and have a short operation period and difficult to maintain.This study developed a scientific and accurate method for prediction of DO content changes using fish school features.The behavioral features of the Carassius auratus fish school were described using two-dimensional fish school images.The degree of DO content decline was graded into five levels,and the corresponding numerical ranges of cluster characteristic parameters were determined by considering the opinions of ichthyologists.Finally,the variation of DO content was predicted using the characteristic parameters of the fish school and the multiple-input single-output Takagi-Sugeno fuzzy neural network.The prediction results were basically consistent with the actual variations of DO content.Therefore,it is feasible to use the behavioral features of the fish school to dynamically predict the level of DO content in water,and this method is especially suitable for prediction of sharp decline of DO content in a relatively short time.展开更多
Fish behavior analysis for recognizing stress is very important for fish welfare and production management in aquaculture.Recent advances have been made in fish behavior analysis based on deep learning.However,most ex...Fish behavior analysis for recognizing stress is very important for fish welfare and production management in aquaculture.Recent advances have been made in fish behavior analysis based on deep learning.However,most existing methods with top performance rely on considerable memory and computational resources,which is impractical in the real-world scenario.In order to overcome the limitations of these methods,a new method based on knowledge distillation is proposed to identify the stress states of fish schools.The knowledge distillation architecture transfers additional inter-class information via a mixed relative loss function,and it forces a lightweight network(GhostNet)to mimic the soft probabilities output of a well-trained fish stress state recognition network(ResNeXt101).The fish school stress state recognition model’s accuracy is improved from 94.17%to 98.12%benefiting from the method.The proposed model has about 5.18 M parameters and requires 0.15 G FLOPs(floating-point operations)to process an image of size 224×224.Furthermore,fish behavior images are collected in a land-based factory,and a dataset is constructed and extended through flip,rotation,and color jitter augmentation techniques.The proposed method is also compared with other state-of-the-art methods.The experimental results show that the proposed model is more suitable for deployment on resource-constrained devices or real-time applications,and it is conducive for real-time monitoring of fish behavior.展开更多
基金funded in part by a grant from the National Oceanic Atmospheric AdministrationA publication of the Grantee pursuant to subgrant NAGL 01-02B+2 种基金National Undersea Research Center for the North Atlantic & Great Lakes, University of Connecticut Award No. NA06RU0140The Woods Hole Sea Grant College ProgramThe Sounds Conservancy,Quebec-Labrador Foundation/Atlantic Center for the Environment provided a stipend for M. Hendry-Brogan
文摘Underwater sound and video observations were made at noon, sunset, and midnight in sand, gravel, and boulder habitat in the Stellwagen Bank National Marine Sanctuary, Gulf of Maine, USA in October 2001 using a remotely operated vehicle （ROV）. Seventeen species of fish and squid were observed with clear habitat and time differences. Observations of feeding behavior, disturbance behavior, and both interspecific and intraspecific interactions provided numerous opportunities for potential sound production; however, sounds were recorded only during a single dive. Although high noise levels generated by the ROV and support ship may have masked some sounds, we conclude that fish sound production in the Gulf of Maine during the fall is uncommon. The recorded fish sounds are tentatively attributed to the cusk Brosme brosme. Cusk sounds consisted variously of isolated thumps, widely spaced thump trains, drumrolls, and their combinations. Frequency peaks were observed at 188, 539, and 1195 Hz. Use of a remotely operated vehicle （ROV） as a passive acoustic observation platform was problematic due to high ROV self-noise and the ROV＇s inability to maintain a fixed position on the bottom without thruster power. Some fishes were clearly also disturbed by ROV noise, indicating a potential ROV sampling bias. Based on our observations, we suggest that new instruments incorporating both optic and passive acoustic technologies are needed to provide better tools for in situ behavioral studies ofcusk and other fishes [Current Zoology 56 （1）： 90-99 2010].
文摘Accurate and rapid detection of fish behaviors is critical to perceive health and welfare by allowing farmers to make informed management deci-sions about recirculating the aquaculture system while decreasing labor.The classic detection approach involves placing sensors on the skin or body of the fish,which may interfere with typical behavior and welfare.The progress of deep learning and computer vision technologies opens up new opportunities to understand the biological basis of this behavior and precisely quantify behaviors that contribute to achieving accurate management in precision farming and higher production efficacy.This study develops an intelligent fish behavior classification using modified invasive weed optimization with an ensemble fusion(IFBC-MIWOEF)model.The presented IFBC-MIWOEF model focuses on identifying the distinct kinds of fish behavior classification.To accomplish this,the IFBC-MIWOEF model designs an ensemble of Deep Learning(DL)based fusion models such as VGG-19,DenseNet,and Effi-cientNet models for fish behavior classification.In addition,the hyperparam-eter tuning of the DL models is carried out using the MIWO algorithm,which is derived from the concepts of oppositional-based learning(OBL)and the IWO algorithm.Finally,the softmax(SM)layer at the end of the DL model categorizes the input into distinct fish behavior classes.The experimental validation of the IFBC-MIWOEF model is tested using fish videos,and the results are examined under distinct aspects.An Extensive comparative study pointed out the improved outcomes of the IFBC-MIWOEF model over recent approaches.
基金supported by the National Natural Science Foundation of China(Grant Nos.51776102 and 91852103)the Institute for Guo Qiang,Tsinghua University(Grant No.2019GQG1019)+1 种基金the Tsinghua National Laboratory for Information Science and Technologythe China Scholarship Council for sponsoring her visit to the University of Minnesota。
文摘In this study,fish behavior and fish injury at different operating conditions are investigated via numerical simulation to evaluate the fish-friendliness of an axial pump that comprises an inlet pipe,a rotor with six blades,a stator with eight vanes,and an outlet pipe.To precisely obtain the flow field when the fish passes through the axial pump,a hybrid large eddy simulation and immersed boundary method is adopted with the full consideration of the fluid-structure interaction comprehensively.The results indicate that the collision between the fish and the wall of flow components in the axial pump is concentrated near the inlet of the rotor,which results in the complexity of the fish trajectory,especially under the large flow rate condition.It is noted that the fish is likely to move in the reverse direction of the main flow after the impact with the rotor blade if the flow rate coefficient is too large,which increases the possibility of collision between the fish and the rotor blade.It is also indicated that the primary factor affecting the strike injury on the fish when it passes through the axial pump is the strike between the fish and the leading edge of the rotor blade.In addition,the strike injury becomes more significant as the flow rate coefficient increases.Furthermore,the results demonstrate that the fish may simultaneously suffer from strike,pressure,and shear stress injuries,once the collision between the fish and the wall of flow components occurs in the axial pump,thus aggravating the combined damage on the fish.Based on these results,it is recommended that hydraulic machinery should not be operated at large flow rates during fish migration from the view of fish-friendliness.
基金supported by the Natural Science Foundation of Changzhou City,China(Grants No.CE20195026 and CE20205031)the Teaching Steering Committee of Electronics Information Specialty in Colleges and Universities of the Ministry of Education(Grant No.2020-YB-42)the Jiangsu Overseas Visiting Scholar Program for University Prominent Young and Middle Aged Teachers and Presidents.
文摘Dissolved oxygen(DO)content is an important index of river water quality.Water quality sensors have been used in China for urban river water monitoring and DO content prediction.However,water quality sensors are expensive and difficult to maintain,and have a short operation period and difficult to maintain.This study developed a scientific and accurate method for prediction of DO content changes using fish school features.The behavioral features of the Carassius auratus fish school were described using two-dimensional fish school images.The degree of DO content decline was graded into five levels,and the corresponding numerical ranges of cluster characteristic parameters were determined by considering the opinions of ichthyologists.Finally,the variation of DO content was predicted using the characteristic parameters of the fish school and the multiple-input single-output Takagi-Sugeno fuzzy neural network.The prediction results were basically consistent with the actual variations of DO content.Therefore,it is feasible to use the behavioral features of the fish school to dynamically predict the level of DO content in water,and this method is especially suitable for prediction of sharp decline of DO content in a relatively short time.
基金supported by the National Science Foundation of China‘Analysis and feature recognition on feeding behavior of fish school in facility farming based on machine vision’(No.62076244)the National Key R&D Program of China‘Next generation precision aquaculture:R&D on intelligent measurement,control and equipment technologies’(China Grant No.2017YFE0122100).
文摘Fish behavior analysis for recognizing stress is very important for fish welfare and production management in aquaculture.Recent advances have been made in fish behavior analysis based on deep learning.However,most existing methods with top performance rely on considerable memory and computational resources,which is impractical in the real-world scenario.In order to overcome the limitations of these methods,a new method based on knowledge distillation is proposed to identify the stress states of fish schools.The knowledge distillation architecture transfers additional inter-class information via a mixed relative loss function,and it forces a lightweight network(GhostNet)to mimic the soft probabilities output of a well-trained fish stress state recognition network(ResNeXt101).The fish school stress state recognition model’s accuracy is improved from 94.17%to 98.12%benefiting from the method.The proposed model has about 5.18 M parameters and requires 0.15 G FLOPs(floating-point operations)to process an image of size 224×224.Furthermore,fish behavior images are collected in a land-based factory,and a dataset is constructed and extended through flip,rotation,and color jitter augmentation techniques.The proposed method is also compared with other state-of-the-art methods.The experimental results show that the proposed model is more suitable for deployment on resource-constrained devices or real-time applications,and it is conducive for real-time monitoring of fish behavior.
