期刊论文

中文

武汉轻工大学学报

2095-7386

2023

42

06

57-62,69

10.3969/j.issn.2095-7386.2023.06.008

2022CFB944:湖北省自然科学基金 Q20211609:湖北省教育厅科研项目 2020J06:武汉轻工大学校***科研项目

本校为第一机构

为解决现有番茄成熟度检测模型存在的计算复杂且精度低的问题,以普罗旺斯番茄为样品,利用 HSV通道数据进行分析,将样本图片划分为绿熟期、变色期和红熟期,建立不同成熟度番茄的数据集.基于迁移学习,对MobilenetV2、InceptionV3 和VGG19 三种卷积神经网络模型的超参数进行了优化,建立了基于番茄成熟度的检测模型.进一步分析了 SGD、Ad-am、Adagrad优化算法和不同训练轮数的检测效果,发现MobilenetV2 的表现最好,在优化器为Adam、训练 100 轮时准确率可达 99.05%,损失值为 0.038 4,满足高精度检测.检测方法能在低损失值下获得更高的检测准确率,可为番茄成熟度检测研究提供参考.

In order to solve the problems of complicated calculation and low accuracy of existing tomato ripe-ner detection models,this paper takes Provence tomatoes as samples and conducts analysis using HSV channel data.The sample images are divided into green ripening,discoloration,and red ripening stages,and a dataset of tomatoes with different maturity levels is established.Based on transfer learning,the hyper pa-rameters of three convolutional neural network models,MobilenetV2,InceptionV3,and VGG19,were opti-mized,and a detection model based on tomato maturity was established.The detection effects ...