结合三维卷积神经网络和Haar小波滤波器的高光谱猕猴桃品质分类  

Hyperspectral Imaging-Based Quality Classification for Kiwifruit by Incorporating Three-Dimensional Convolution Neural Network and Haar Wavelet Filter

作　　者：金科 郭志强[1] 曾云流[2] 丁港 Jin Ke;Guo Zhiqiang;Zeng Yunliu;Ding Gang(College of Information Engineering,Wuhan University of Technology,Wuhan 430070,Hubei,China;Key Laboratory of Horticultural Plant Biology,Ministry of Education,National R&D Center for Citrus Preservation,College of Horticulture and Forestry Sciences,Huazhong Agricultural University,Wuhan 430070,Hubei,China)

机构地区：[1]武汉理工大学信息工程学院,湖北武汉430070 [2]华中农业大学园艺林学学院园艺植物生物学教育部重点实验室,国家柑橘保鲜技术研发专业中心,湖北武汉430070

出　　处：《激光与光电子学进展》2023年第20期41-50,共10页Laser & Optoelectronics Progress

基　　金：国家重点研发计划(2021YFD1200202-08);国家自然科学基金(32272779);猕猴桃质量安全与加工保鲜岗位项目(CARS26);湖北省重点研发项目(2021BBA090)。

摘　　要：针对猕猴桃硬度品质无损检测分类困难的问题,提出了结合高光谱成像技术和卷积神经网络的分类模型。该模型融合Haar小波核提取的空间特征信息和三维卷积核提取的空谱联合信息,采用分解数据通道连接的方式确保所有特征能够流到模型末尾,提升了网络特征提取的能力。通过自制的猕猴桃硬度品质Kiwi_seed数据集上的实验表明,Haar小波变换模块可以显著提升网络的特征提取能力;通过消融实验表明,在增加Haar小波变换模块后模型的分类准确率提升了7.4%,最优可达97.3%,优于经典的图像分类网络,可以很好地解决猕猴桃硬度品质的无损检测分类问题。To address challenges in the non-destructive inspection and classification of kiwifruit hardness quality,we propose a classification model that incorporates hyperspectral imaging technology and a convolution neural network.This network combines the spatial feature information extracted by the Haar wavelet and the space-spectrum joint information extracted by the three-dimensional(3D)convolution kernel.In this network,the data decomposition of channel connections is executed to ensure that all features can be utilized by the model,which improves the ability of network feature learning.Experiments on the acquired hyperspectral image-based,self-made kiwifruit hardness quality dataset(named Kiwi_seed)demonstrate that the Haar wavelet transform module can significantly improve the feature extraction ability of the network.Ablation experiments reveal that the classification accuracy of the model incorporating the Haar wavelet transform module is increased by 7.4%and reaches the optimum level at 97.3%,which is better than the classical image classification network.The proposed classification model can be effectively used for the non-destructive inspection and classification of kiwifruit quality.

关 键 词：高光谱图像 猕猴桃 图像分类 卷积神经网络 HAAR小波变换 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]