Aiming at the problem of low detection accuracy due to the different scale sizes of apple leaf disease spots and their similarity to the background,this paper proposes a multi-scale lightweight network(MSL-Net).Firstl...Aiming at the problem of low detection accuracy due to the different scale sizes of apple leaf disease spots and their similarity to the background,this paper proposes a multi-scale lightweight network(MSL-Net).Firstly,a multiplexed aggregated feature extraction network is proposed using residual bottleneck block(RES-Bottleneck)and middle partial-convolution(MP-Conv)to capture multi-scale spatial features and enhance focus on disease features for better differentiation between disease targets and background information.Secondly,a lightweight feature fusion network is designed using scale-fuse concatenation(SF-Cat)and triple-scale sequence feature fusion(TSSF)module to merge multi-scale feature maps comprehensively.Depthwise convolution(DWConv)and GhostNet lighten the network,while the cross stage partial bottleneck with 3 convolutions ghost-normalization attention module(C3-GN)reduces missed detections by suppressing irrelevant background information.Finally,soft non-maximum suppression(Soft-NMS)is used in the post-processing stage to improve the problem of misdetection of dense disease sites.The results show that the MSL-Net improves mean average precision at intersection over union of 0.5(mAP@0.5)by 2.0%over the baseline you only look once version 5s(YOLOv5s)and reduces parameters by 44%,reducing computation by 27%,outperforming other state-of-the-art(SOTA)models overall.This method also shows excellent performance compared to the latest research.展开更多
This paper proposes an improved You Only Look Once(YOLOv3)algorithm for automatically detecting damaged apples to promote the automation of the fruit processing industry.In the proposed method,a clustering method base...This paper proposes an improved You Only Look Once(YOLOv3)algorithm for automatically detecting damaged apples to promote the automation of the fruit processing industry.In the proposed method,a clustering method based on Rao-1 algorithm is introduced to optimize anchor box sizes.The clustering method uses the intersection over the union to form the objective function and the most representative anchor boxes are generated for normal and damaged apple detection.To verify the feasibility and effectiveness of the proposed method,real apple images collected from the Internet are employed.Compared with the generic YOLOv3 and Fast Region-based Convolutional Neural Network(Fast R-CNN)algorithms,the proposed method yields the highest mean average precision value for the test dataset.Therefore,it is practical to apply the proposed method for intelligent apple detection and classification tasks.展开更多
Using machine vision to accurately identify apple number on the tree is becoming the key supporting technology for orchard precision production management.For adapting to the complexity of the field environment in var...Using machine vision to accurately identify apple number on the tree is becoming the key supporting technology for orchard precision production management.For adapting to the complexity of the field environment in various detection situations,such as illumination changes,color variation,fruit overlap,and branches and leaves shading,a robust algorithm for detecting and counting apples based on their color and shape modes was proposed.Firstly,BP(back propagation)neural network was used to train apple color identification model.Accordingly the irrelevant background was removed by using the trained neural network model and the image only containing the apple color pixels was acquired.Then apple edge detection was carried out after morphological operations on the obtained image.Finally,the image was processed by using circle Hough transform algorithm,and apples were located with the help of calculating the center coordinates of each apple edge circle.The validation experimental results showed that the correlation coefficient of R2 between the proposed approaches based counting and manually counting reached 0.985.It illustrated that the proposed algorithm could be used to detect and count apples from apple trees’images taken in field environment with a high precision and strong anti-jamming feature.展开更多
文摘Aiming at the problem of low detection accuracy due to the different scale sizes of apple leaf disease spots and their similarity to the background,this paper proposes a multi-scale lightweight network(MSL-Net).Firstly,a multiplexed aggregated feature extraction network is proposed using residual bottleneck block(RES-Bottleneck)and middle partial-convolution(MP-Conv)to capture multi-scale spatial features and enhance focus on disease features for better differentiation between disease targets and background information.Secondly,a lightweight feature fusion network is designed using scale-fuse concatenation(SF-Cat)and triple-scale sequence feature fusion(TSSF)module to merge multi-scale feature maps comprehensively.Depthwise convolution(DWConv)and GhostNet lighten the network,while the cross stage partial bottleneck with 3 convolutions ghost-normalization attention module(C3-GN)reduces missed detections by suppressing irrelevant background information.Finally,soft non-maximum suppression(Soft-NMS)is used in the post-processing stage to improve the problem of misdetection of dense disease sites.The results show that the MSL-Net improves mean average precision at intersection over union of 0.5(mAP@0.5)by 2.0%over the baseline you only look once version 5s(YOLOv5s)and reduces parameters by 44%,reducing computation by 27%,outperforming other state-of-the-art(SOTA)models overall.This method also shows excellent performance compared to the latest research.
基金National Nature Science and Foundation of China under Grants 62202044 and 62002016the Guangdong Basic and Applied Basic Research Foundation under Grant 2020A1515110431+4 种基金Scientific and Technological Innovation Foundation of Foshan under Grant BK22BF009the NSFC Youth Scientist Fund under Grant 52007160the Beijing Natural Science Foundation under Grant L211020the Interdisciplinary Research Project for Young Teachers of USTB(Fundamental Research Funds for the Central Universities)under Grant FRF-IDRY-21-003the Fundamental Research Funds for the Central Universities and the Youth Teacher International Exchange&Growth Program(No.QNXM20220040).
文摘This paper proposes an improved You Only Look Once(YOLOv3)algorithm for automatically detecting damaged apples to promote the automation of the fruit processing industry.In the proposed method,a clustering method based on Rao-1 algorithm is introduced to optimize anchor box sizes.The clustering method uses the intersection over the union to form the objective function and the most representative anchor boxes are generated for normal and damaged apple detection.To verify the feasibility and effectiveness of the proposed method,real apple images collected from the Internet are employed.Compared with the generic YOLOv3 and Fast Region-based Convolutional Neural Network(Fast R-CNN)algorithms,the proposed method yields the highest mean average precision value for the test dataset.Therefore,it is practical to apply the proposed method for intelligent apple detection and classification tasks.
基金The authors acknowledge that this research was supported by Chinese National Science and Technology Support Program(2012BAH29B04)863 Project(2012AA101900).
文摘Using machine vision to accurately identify apple number on the tree is becoming the key supporting technology for orchard precision production management.For adapting to the complexity of the field environment in various detection situations,such as illumination changes,color variation,fruit overlap,and branches and leaves shading,a robust algorithm for detecting and counting apples based on their color and shape modes was proposed.Firstly,BP(back propagation)neural network was used to train apple color identification model.Accordingly the irrelevant background was removed by using the trained neural network model and the image only containing the apple color pixels was acquired.Then apple edge detection was carried out after morphological operations on the obtained image.Finally,the image was processed by using circle Hough transform algorithm,and apples were located with the help of calculating the center coordinates of each apple edge circle.The validation experimental results showed that the correlation coefficient of R2 between the proposed approaches based counting and manually counting reached 0.985.It illustrated that the proposed algorithm could be used to detect and count apples from apple trees’images taken in field environment with a high precision and strong anti-jamming feature.
