YOLO-Banana:An Effective Grading Method for Banana Appearance Quality

作     者：Dianhui Mao Xuesen Wang Yiming Liu Denghui Zhang Jianwei Wu Junhua Chen Dianhui Mao;Xuesen Wang;Yiming Liu;Denghui Zhang;Jianwei Wu;Junhua Chen

作者机构：School of ComputerBeijing Technology and Business UniversityBeijing 100048China Research Center of Information TechnologyBeijing Academy of Agriculture and Forestry SciencesBeijing 100097China Beijing PAIDE Science and Technology Development Co.Ltd.Beijing 100097China Institute of Standardization Theory and StrategyChina National Institute of StandardizationBeijing 100088China 

出 版 物：《Journal of Beijing Institute of Technology》 (北京理工大学学报（英文版）)

年 卷 期：2023年第32卷第3期

页      面：363-373页

核心收录：

学科分类：08[工学] 0901[农学-作物学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the Beijing Science Foundation(No.9232005) the Beijing Municipal Philosophy and Social Science Foundation of China(No.19GLB036) the Beijing Science and Technology Project(No.Z221100005822014)。 

主　　题：YOLOv5 banana appearance grading clustering algorithm weighted non-maximum suppression(weighted NMS) progressive aggregated network(PANet) 

摘      要：The increasing trend towards independent fruit packaging demands a high appearance quality of individually packed fruits.In this paper,we propose an improved YOLOv5-based model,YOLO-Banana,to effectively grade banana appearance quality based on the number of banana defect points.Due to the minor and dense defects on the surface of bananas,existing detection algorithms have poor detection results and high missing rates.To address this,we propose a densitybased spatial clustering of applications with noise(DBSCAN)and K-means fusion clustering method that utilizes refined anchor points to obtain better initial anchor values,thereby enhancing the network’s recognition accuracy.Moreover,the optimized progressive aggregated network(PANet)enables better multi-level feature fusion.Additionally,the non-maximum suppression function is replaced with a weighted non-maximum suppression(weighted NMS)function based on distance intersection over union(DIoU).Experimental results show that the model’s accuracy is improved by 2.3%compared to the original YOLOv5 network model,thereby effectively grading the banana appearance quality.