采用改进YOLOv3-Tiny模型的轻量化莲蓬质量分级算法  

Lightweight classification of lotus seedpod quality using improved YOLOv3-Tiny model

作　　者：张雷[1,2] 严昊 贾永镒 叶秉良 马锃宏[1,2] ZHANG Lei;YAN Hao;JIA Yongyi;YE Bingliang;MA Zenghong(School of Mechanical Engineering,Zhejiang Sci-Tech University,Hangzhou 310018,China;Key Laboratory of Agricultural Intelligent Sensing and Robotics of Zhejiang Province,Hangzhou 310018,China)

机构地区：[1]浙江理工大学机械工程学院,杭州310018 [2]浙江省农业智能感知与机器人全省重点实验室,杭州310018

出　　处：《农业工程学报》2024年第23期248-257,共10页Transactions of the Chinese Society of Agricultural Engineering

基　　金：国家自然科学基金项目(32372005)。

摘　　要：精准高效的莲蓬质量分级算法是实现莲蓬采后自动化加工的重要一环。针对目前莲蓬果实的采后质量分级研究较少的问题,该研究建立了莲蓬果实质量分级原则,提出了改进YOLOv3-Tiny(you only look once version 3-Tiny)模型的莲蓬质量分级算法。首先在3种光照条件下架设摄像头垂直采集莲蓬图像并建立试验数据集,通过数据增强技术扩充数据集;接着使用K均值聚类算法重新设计先验锚框尺度,提高先验锚框的回归精度。随后以YOLOv3-Tiny原骨干网络为基础,加入空间金字塔池化模块(spatial pyramid pooling,SPP),提升网络提取特征信息的能力;最后利用YOLOv3-Tiny的参数进化模块为该模型进化出一套合适的超参数。试验结果表明,改进的YOLOv3-Tiny模型对莲子识别的平均精度均值(mean average precision,mAP)和召回率(recall)分别为96.80%和94.60%;与原YOLOv3-Tiny模型相比,mAP提高12.49个百分点,召回率提高11.59个百分点,并且每秒传输帧数达到25帧,是Faster R-CNN网络模型的1.24倍。试验数据说明所提改进算法对于莲蓬果实上的莲子具有更好的识别效果,而且满足实时检测的要求,可以为莲蓬质量分级研究提供技术参考。An accurate and efficient classification of quality is one of the most important steps to realize the automatic processing of lotus after harvest.However,it is still lacking in the quality classification of lotus after harvest.In this study,the classification of lotus quality was established,according to the evaluation scheme for white lotus quality published by the Chinese Association of Traditional Chinese Medicine.An improved YOLOv3-Tiny(You Only Look Once version 3-Tiny)model was also proposed for quality classification.Firstly,a vertical camera was deployed on an image acquisition platform for lotus seedpod,according to the principles of image recognition.The images of lotus seedpods were collected under three lighting conditions.A comprehensive test dataset was then constructed to augment using three image augmentation techniques:rotation,color switching,and affine transformation.The K-means clustering was also employed to optimize the scales of the prior anchor boxes,in order to enhance the regression accuracy.Ultimately,six scales were generated for the prior anchor boxes,namely(14,18),(19,21),(20,27),(36,44),(42,49),and(50,56).The accuracy rate was achieved at 85.48%.Subsequently,the SPP(Spatial Pyramid Pooling)module was added to extract feature information,according to the original YOLOv3-Tiny backbone network.Finally,the parameter evolution module of YOLOv3-Tiny was used to evolve a set of appropriate hyperparameters for the model.Ablation experiments were carried out to verify the effectiveness of data augmentation and redesign the prior anchor box scales.The backbone network of feature extraction was then optimized to improve the performance in the related hyperparameter evolution.The ablation experiments showed that the recognition precision increased by 4.95 percentage points after data augmentation.The precision of recognition increased by 0.76 percentage points after the K-Means algorithm to reunite classes.Spatial Pyramid Pooling(SPP)module was added to improve the precision of recognition by 1

关 键 词：深度学习 分级 模型 YOLOv3-Tiny 莲蓬 轻量化 

分 类 号：S126[农业科学—农业基础科学]