Su Zhenzhu, Zhang Chu, Yan Tianying, Zhu Jianan, Zeng Yulan, Lu Xuanjun, Gao Pan, Feng Lei, He Linhai, Fan Lihui
Institute of Biotechnology, Zhejiang University, Hangzhou, China.
School of Information Engineering, Huzhou University, Huzhou, China.
Front Plant Sci. 2021 Sep 10;12:736334. doi: 10.3389/fpls.2021.736334. eCollection 2021.
Maturity degree and quality evaluation are important for strawberry harvest, trade, and consumption. Deep learning has been an efficient artificial intelligence tool for food and agro-products. Hyperspectral imaging coupled with deep learning was applied to determine the maturity degree and soluble solids content (SSC) of strawberries with four maturity degrees. Hyperspectral image of each strawberry was obtained and preprocessed, and the spectra were extracted from the images. One-dimension residual neural network (1D ResNet) and three-dimension (3D) ResNet were built using 1D spectra and 3D hyperspectral image as inputs for maturity degree evaluation. Good performances were obtained for maturity identification, with the classification accuracy over 84% for both 1D ResNet and 3D ResNet. The corresponding saliency maps showed that the pigments related wavelengths and image regions contributed more to the maturity identification. For SSC determination, 1D ResNet model was also built, with the determination of coefficient ( ) over 0.55 of the training, validation, and testing sets. The saliency maps of 1D ResNet for the SSC determination were also explored. The overall results showed that deep learning could be used to identify strawberry maturity degree and determine SSC. More efforts were needed to explore the use of 3D deep learning methods for the SSC determination. The close results of 1D ResNet and 3D ResNet for classification indicated that more samples might be used to improve the performances of 3D ResNet. The results in this study would help to develop 1D and 3D deep learning models for fruit quality inspection and other researches using hyperspectral imaging, providing efficient analysis approaches of fruit quality inspection using hyperspectral imaging.
成熟度和品质评估对于草莓的收获、贸易和消费至关重要。深度学习已成为用于食品和农产品的一种高效人工智能工具。将高光谱成像与深度学习相结合，用于测定具有四个成熟度等级的草莓的成熟度和可溶性固形物含量（SSC）。获取并预处理每个草莓的高光谱图像，然后从图像中提取光谱。使用一维光谱和三维高光谱图像作为输入，构建一维残差神经网络（1D ResNet）和三维（3D）ResNet来评估成熟度。在成熟度识别方面取得了良好的性能，1D ResNet和3D ResNet的分类准确率均超过84%。相应的显著性图表明，与色素相关的波长和图像区域对成熟度识别的贡献更大。对于SSC测定，也构建了1D ResNet模型，训练集、验证集和测试集的决定系数（ ）超过0.55。还探索了1D ResNet用于SSC测定的显著性图。总体结果表明，深度学习可用于识别草莓成熟度并测定SSC。需要进一步努力探索使用三维深度学习方法进行SSC测定。1D ResNet和3D ResNet在分类方面的相近结果表明，可能需要更多样本以提高3D ResNet的性能。本研究结果将有助于开发用于水果品质检测的一维和三维深度学习模型以及其他使用高光谱成像的研究，提供使用高光谱成像进行水果品质检测的有效分析方法。
