基于BiSeNet的糖心苹果截面糖心特征提取方法  

Method of extracting characteristics of watercore in cross section of watercore apple based on BiSeNet

作　　者：尹治棚 张文斌[2] 赵春林 YIN Zhipeng;ZHANG Wenbin;ZHAO Chunlin(College of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Kunming 650500,China;College of Engineering,Honghe University/Yunnan Province Key Laboratory of Mechanical Performance Analysis and Optimization of Plateau,Mengzi 661199,China)

机构地区：[1]昆明理工大学机电工程学院,昆明650500 [2]红河学院工学院/云南省高校高原机械性能分析与优化省重点实验室,蒙自661199

出　　处：《华中农业大学学报》2023年第2期209-215,共7页Journal of Huazhong Agricultural University

基　　金：国家自然科学基金项目(51769007);云南省地方本科高校基础研究联合专项重点项目(202001BA070001-002);兴滇英才支持计划项目(YNWR-QNBJ-2018-349);云南省地方高校联合专项面上项目(202001BA070001-015)。

摘　　要：为解决因糖心苹果截面糖心特征分布不规则导致的种植过程中对糖心品质评估方法精度低、复杂等问题,以380个糖心苹果作为试验样本,基于BiSeNet模型对苹果糖心特征进行提取与占比计算,分别评估该方法在理论研究与实践研究两方面的多项性能指标,并与FCN、PPLiteSeg与DeepLabV3这3种网络模型进行对比。结果显示,BiSeNet无论是在训练时间还是训练准确度都优于其他3种网络模型,用时130 s,准确率98.36%,交并比85.1%。在实际占比计算中,平均占比计算误差为4.04%,低于其他3种模型,且无较大偏差值。结果表明,基于BiSeNet的糖心苹果截面糖心特征提取方法在提供具体糖心占比的同时,可为糖心苹果的糖心特征等无损检测提供更精确的评估方法和比对目标。A method of extracting watercore feature and calculating proportion based on BiSeNet was proposed to solve the problems of low precision and complex methods of evaluating watercore quality dur⁃ing the planting process due to the irregular distribution of watercore characteristics in the cross section of watercore apple.Four models including BiSeNet,FCN,PPLiteSeg and DeepLabV3 were used to extract the watercore characteristics in the cross section of 380 watercore apple samples.Each evaluation index in theoretical research and practical research were calculated separately.The comprehensive evaluation and comparison was conducted.The results showed that BiSeNet was superior to the other three network mod⁃els in both training time and training accuracy.It took 130 s,with the accuracy rate of 98.36% and the in⁃tersection and combination ratio of 85.1%.In the actual proportion calculation,the average proportion cal⁃culation error was 4.04%,lower than that of the other three models,and there was no large deviation.It is indicated that the method of extracting and calculating watercore characteristics in the cross section of water⁃core apple based on BiSeNet can provide more accurate methods of evaluation and comparison targets for nondestructive testing of watercore characteristics of watercore apples while providing specific proportion of watercore.

关 键 词：糖心苹果 BiSeNet 糖心特征提取 无损检测 FCN PPLiteSeg DeepLabV3 精准评估 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]