The swift evolution of deep learning has greatly benefited the field of intensive aquaculture.Specifically,deep learning-based shrimp larvae detection has offered important technical assistance for counting shrimp lar...The swift evolution of deep learning has greatly benefited the field of intensive aquaculture.Specifically,deep learning-based shrimp larvae detection has offered important technical assistance for counting shrimp larvae and recognizing abnormal behaviors.Firstly,the transparent bodies and small sizes of shrimp larvae,combined with complex scenarios due to variations in light intensity and water turbidity,make it challenging for current detection methods to achieve high accuracy.Secondly,deep learning-based object detection demands substantial computing power and storage space,which restricts its application on edge devices.This paper proposes an efficient one-stage shrimp larvae detection method,FAMDet,specifically designed for complex scenarios in intensive aquaculture.Firstly,different from the ordinary detection methods,it exploits an efficient FasterNet backbone,constructed with partial convolution,to extract effective multi-scale shrimp larvae features.Meanwhile,we construct an adaptively bi-directional fusion neck to integrate high-level semantic information and low-level detail information of shrimp larvae in a matter that sufficiently merges features and further mitigates noise interference.Finally,a decoupled detection head equipped with MPDIoU is used for precise bounding box regression of shrimp larvae.We collected images of shrimp larvae from multiple scenarios and labeled 108,365 targets for experiments.Compared with the ordinary detection methods(Faster RCNN,SSD,RetinaNet,CenterNet,FCOS,DETR,and YOLOX_s),FAMDet has obtained considerable advantages in accuracy,speed,and complexity.Compared with the outstanding one-stage method YOLOv8s,it has improved accuracy while reducing 57%parameters,37%FLOPs,22%inference latency per image on CPU,and 56%storage overhead.Furthermore,FAMDet has still outperformed multiple lightweight methods(EfficientDet,RT-DETR,GhostNetV2,EfficientFormerV2,EfficientViT,and MobileNetV4).In addition,we conducted experiments on the public dataset(VOC 07+12)to further verify the effectiveness of FAMDet.Consequently,the proposed method can effectively alleviate the limitations faced by resource-constrained devices and achieve superior shrimp larvae detection results.展开更多
In this paper,we propose a novel approach to the multiple-sample testing problem using a studentized test statistic based on the random lifter technique.The method reformulates the classical Ksample test as an indepen...In this paper,we propose a novel approach to the multiple-sample testing problem using a studentized test statistic based on the random lifter technique.The method reformulates the classical Ksample test as an independence test between two random variables,enabling more efficient handling of complex data types.We solve problems with the non-standard normal limiting distributions of degenerate U-statistics using a random lifter approach.This creates a test statistic that is asymptotically normal under the null hypothesis.Numerous simulations and real-world applications have demonstrated that our method performs well with many data types,including Euclidean,directional,and symmetric positive definite data.It is also very good at controlling Type I errors.Our method also shows significant computational efficiency,outperforming existing K-sample tests,particularly when applied to large datasets.These results suggest that the proposed method is a powerful and practical solution for multiple-sample testing in complex data scenarios.展开更多
Low-angle estimation for very high frequency(VHF)radar is a difficult problem due to the multipath effect in the radar field,especially in complex scenarios where the reflection condition is unknown.To deal with this ...Low-angle estimation for very high frequency(VHF)radar is a difficult problem due to the multipath effect in the radar field,especially in complex scenarios where the reflection condition is unknown.To deal with this problem,we propose an algorithm of target height and multipath attenuation joint estimation.The amplitude of the surface reflection coefficient is estimated by the characteristic of the data itself,and it is assumed that there is no reflected signal when the amplitude is very small.The phase of the surface reflection coefficient and the phase difference between the direct and reflected signals are searched as the same part,and this represents the multipath phase attenuation.The Cramer-Rao bound of the proposed algorithm is also derived.Finally,computer simulations and real data processing results show that the proposed algorithm has good estimation performance under complex scenarios and works well with only one snapshot.展开更多
representation capability of deep learning(DL) and the optimal decision making and control capability of reinforcement learning(RL), is a good approach to address this problem. Traffic environment is built up by combi...representation capability of deep learning(DL) and the optimal decision making and control capability of reinforcement learning(RL), is a good approach to address this problem. Traffic environment is built up by combining intelligent driver model(IDM) and lane-change model as behavioral model for vehicles. To increase the stochastic of the established traffic environment, tricks such as defining a speed distribution with cutoff for traffic cars and using various politeness factors to represent distinguished lane-change style, are taken. For training an artificial agent to achieve successful strategies that lead to the greatest long-term rewards and sophisticated maneuver, deep deterministic policy gradient(DDPG) algorithm is deployed for learning. Reward function is designed to get a trade-off between the vehicle speed, stability and driving safety. Results show that the proposed approach can achieve good autonomous maneuvering in a scenario of complex traffic behavior through interaction with the environment.展开更多
基金supported by the National Shrimp and Crab Industry Technical System Construction Project 2022(No.CARS-48)the National Natural Science Foundation of China(No.62076244)+2 种基金the Chinese Universities Scientific Fund(No.2022TC109)the Double First-class Project of China Agricultural University(2022)and the Double First-class International Cooperation Project of China Agricultural University(No.10020799).
文摘The swift evolution of deep learning has greatly benefited the field of intensive aquaculture.Specifically,deep learning-based shrimp larvae detection has offered important technical assistance for counting shrimp larvae and recognizing abnormal behaviors.Firstly,the transparent bodies and small sizes of shrimp larvae,combined with complex scenarios due to variations in light intensity and water turbidity,make it challenging for current detection methods to achieve high accuracy.Secondly,deep learning-based object detection demands substantial computing power and storage space,which restricts its application on edge devices.This paper proposes an efficient one-stage shrimp larvae detection method,FAMDet,specifically designed for complex scenarios in intensive aquaculture.Firstly,different from the ordinary detection methods,it exploits an efficient FasterNet backbone,constructed with partial convolution,to extract effective multi-scale shrimp larvae features.Meanwhile,we construct an adaptively bi-directional fusion neck to integrate high-level semantic information and low-level detail information of shrimp larvae in a matter that sufficiently merges features and further mitigates noise interference.Finally,a decoupled detection head equipped with MPDIoU is used for precise bounding box regression of shrimp larvae.We collected images of shrimp larvae from multiple scenarios and labeled 108,365 targets for experiments.Compared with the ordinary detection methods(Faster RCNN,SSD,RetinaNet,CenterNet,FCOS,DETR,and YOLOX_s),FAMDet has obtained considerable advantages in accuracy,speed,and complexity.Compared with the outstanding one-stage method YOLOv8s,it has improved accuracy while reducing 57%parameters,37%FLOPs,22%inference latency per image on CPU,and 56%storage overhead.Furthermore,FAMDet has still outperformed multiple lightweight methods(EfficientDet,RT-DETR,GhostNetV2,EfficientFormerV2,EfficientViT,and MobileNetV4).In addition,we conducted experiments on the public dataset(VOC 07+12)to further verify the effectiveness of FAMDet.Consequently,the proposed method can effectively alleviate the limitations faced by resource-constrained devices and achieve superior shrimp larvae detection results.
基金supported by National Natural Science Foundation of China(Grant Nos.12231017,72171216,71921001 and 71991474)the National Key R&D Program of China(Grant No.2022YFA1003803)supported by National Natural Science Foundation of China(Grant Nos.71873128 and 72293573).
文摘In this paper,we propose a novel approach to the multiple-sample testing problem using a studentized test statistic based on the random lifter technique.The method reformulates the classical Ksample test as an independence test between two random variables,enabling more efficient handling of complex data types.We solve problems with the non-standard normal limiting distributions of degenerate U-statistics using a random lifter approach.This creates a test statistic that is asymptotically normal under the null hypothesis.Numerous simulations and real-world applications have demonstrated that our method performs well with many data types,including Euclidean,directional,and symmetric positive definite data.It is also very good at controlling Type I errors.Our method also shows significant computational efficiency,outperforming existing K-sample tests,particularly when applied to large datasets.These results suggest that the proposed method is a powerful and practical solution for multiple-sample testing in complex data scenarios.
基金the Fund for Foreign Scholars in University Research and Teaching Programs(the 111 Project)(No.B18039)。
文摘Low-angle estimation for very high frequency(VHF)radar is a difficult problem due to the multipath effect in the radar field,especially in complex scenarios where the reflection condition is unknown.To deal with this problem,we propose an algorithm of target height and multipath attenuation joint estimation.The amplitude of the surface reflection coefficient is estimated by the characteristic of the data itself,and it is assumed that there is no reflected signal when the amplitude is very small.The phase of the surface reflection coefficient and the phase difference between the direct and reflected signals are searched as the same part,and this represents the multipath phase attenuation.The Cramer-Rao bound of the proposed algorithm is also derived.Finally,computer simulations and real data processing results show that the proposed algorithm has good estimation performance under complex scenarios and works well with only one snapshot.
文摘representation capability of deep learning(DL) and the optimal decision making and control capability of reinforcement learning(RL), is a good approach to address this problem. Traffic environment is built up by combining intelligent driver model(IDM) and lane-change model as behavioral model for vehicles. To increase the stochastic of the established traffic environment, tricks such as defining a speed distribution with cutoff for traffic cars and using various politeness factors to represent distinguished lane-change style, are taken. For training an artificial agent to achieve successful strategies that lead to the greatest long-term rewards and sophisticated maneuver, deep deterministic policy gradient(DDPG) algorithm is deployed for learning. Reward function is designed to get a trade-off between the vehicle speed, stability and driving safety. Results show that the proposed approach can achieve good autonomous maneuvering in a scenario of complex traffic behavior through interaction with the environment.
