基于改进VGG的轻量级香菇品质分类模型

doi:10.13733/j.jcam.issn.2095‑5553.2025.02.018

摘要/Abstract

摘要： 为实现对香菇品质的快速、准确分类，依据大棚种植环境下香菇图像的特性，对VGG16深度卷积网络进行轻量化改进，提出一种香菇品质分类检测模型。首先，对VGG16网络进行轻量化处理，利用均和池化层代替全连接层对特征图进行下采样；然后，在特征提取网络中引入空洞融合分离卷积和通道注意力SE模块提升模型的识别精度；之后，利用数据增强方法将数据集扩充；最后，使用迁移学习训练得到香菇品质分类检测模型。在相同的试验条件下，与VGG16、GoogLeNet、VGG19、ResNet50、MobileNetv1五种深度卷积网络模型相比较。结果表明：该模型的综合性能最好，改进后的VGG16网络的识别准确率为95.5%；模型大小约为原始VGG16模型体量的10.9%；训练时间为原始VGG16模型的55.1%。

关键词: 香菇, 品质分类, VGG16, SE模块, 均和池化, 空洞融合分离卷积

Abstract: In order to achieve rapid and accurate classification of shiitake mushroom quality, according to the characteristics of shiitake mushroom images in greenhouse planting environment, a shiitake quality classification detection model is proposed based on a lightweight improvement of VGG16 deep convolutional network. Firstly, the VGG16 network is lightweighted, and the hybrid pooling layer is used instead of the fully connected layer to downsample the feature map. Then, the hole fusion separation convolution and channel attention SE modules are introduced in the feature extraction network to improve the recognition accuracy of the model. After that, the dataset is enriched with data augmentation methods. Finally, transfer learning is used to train the shiitake mushroom quality classification detection model. Under the same experimental conditions, five deep convolutional network models are compared with VGG16, GoogLeNet, VGG19, ResNet50 and MobileNetv1. The results show that the comprehensive performance of the proposed model is the best, and the recognition accuracy of the improved VGG16 network is 95.5%. The model size is about 10.9% of the original VGG16 model volume. The training time is 55.1% of the original VGG16 model.

Key words: Lentinula edodes, shiitake quality classification, VGG16, SE module, equalization pooling, void fusion separation convolution

中图分类号:

孙岩, 朱凤武, 张宇清, 张伟健, 吴铖轩. 基于改进VGG的轻量级香菇品质分类模型 [J]. 中国农机化学报, 2025, 46(2): 120-125.

Sun Yan, Zhu Fengwu, Zhang Yuqing, Zhang Weijian, Wu Chengxuan. A lightweight shiitake quality classification model based on improved VGG[J]. Journal of Chinese Agricultural Mechanization, 2025, 46(2): 120-125.

参考文献

[ 1 ] 程洋洋, 惠靖茹, 郝竞霄, 等. 香菇柄发酵酱营养品质及风味成分研究[J]. 中国酿造, 2022, 41(7): 162-167.
[ 2 ] 2021年度全国食用菌统计调查结果分析[J]. 中国食用菌, 2023, 42(1): 118-127.
[ 3 ] 陈红, 夏青, 左婷, 等. 基于纹理分析的香菇品质分选方法[J]. 农业工程学报, 2014, 30(3): 285-292.
Chen Hong, Xia Qing, Zuo Ting, et al. Quality grading method of shiitake based on texture analysis [J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(3): 285-292.
[ 4 ] Liu Q, Fang M, Li Y S, et al. Deep learning based research on quality classification of shiitake mushrooms [J]. LWT, 2022, 168, 113902.
[ 5 ] 张晓蕾. 卷积神经网络光谱分析方法及其在农产品品质检测中的应用[D]. 杭州: 浙江大学, 2021.
[ 6 ] 王相友, 李晏兴, 杨振宇, 等. 基于改进YOLOv4模型的马铃薯中土块石块检测方法[J]. 农业机械学报, 2021, 52(8): 241-247, 262.
Wang Xiangyou, Li Yanxing, Yang Zhenyu, et al. Detection method of clods and stones from impurified potatoes based on improved YOLOv4 algorithm [J]. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(8): 241-247, 262.
[ 7 ] Medeiros A, Bernardes R C, Silva L, et al. Deep learning‑based approach using X-ray images for classifying crambe abyssinica seed quality [J]. Industrial Crops and Products, 2021, 164, 113378.
[ 8 ] Zhang X, Wang G H. Study pose detection based on photometric stereo and lightweight YOLOv4[J]. CAAI Transactions on Intelligence Technology, 2022, 2(1): 32-37.
[ 9 ] Zou Q J, Xiong K, Fang Q, et al. Deep imitation reinforcement learning for self‑driving by vision [J]. CAAI Transactions on Intelligence Technology, 2021, 6(4): 493-503.
[10] Lecun Y, Bengio Y, Hinton G. Deep learning [J]. Nature, 2015, 521(7553): 436-444.
[11] 徐艳蕾, 王琦, 翟钰婷, 等. 基于MICS-CoTNet的黑木耳品质分类方法[J]. 农业工程学报, 2023, 39(5): 146-155.
Xu Yanlei, Wang Qi, Zhai Yuting, et al. Method for the classification of black fungus quality using MICS-CoTNet [J]. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(5): 146-155.
[12] Hu J, Shen L, Sun G. Squeeze‑and‑excitation networks [C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[13] 孟庆宽, 张漫, 杨晓霞, 等. 基于轻量卷积结合特征信息均和的玉米幼苗与杂草识别[J]. 农业机械学报, 2020, 51(12): 238-245.
[14] Roy S K, Dubey S R, Chatterjee S, et al. FuSENet: Fused squeeze‑and‑excitation network for spectral‑spatial hyperspectral image classification [J]. IET Image Processing, 2020, 14(8): 1653-1661.
[15] 刘万军, 李嘉欣, 曲海成. 基于多尺度卷积神经网络的交通标示识别研究[J]. 计算机应用研究, 2022, 39(5): 1557-1562．
[16] He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition [J]. Computer Vision and Pattern Recognition, 2015, 34(10): 47-50.
[17] Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications ［EB/OL］. https://arxiv．org /abs /1704. 04861, 2017-04-17.
[18] Lin T Y，Goyal P，Girshick R，et al. Focal loss for dense object detection [C]. Proceedings of the IEEE International Conference on Computer Vision, 2017: 2980-2988.
[19] Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift [C]. International Conference on Machine Learning. PMLR, 2015: 448-456.
[20] Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: A simple way to prevent neural networks from overfitting [J]. Journal of Machine Learning Research, 2014, 15(1): 1929-1958.
[21] Pan S J, Yang Q. A survey on transfer learning [J]. IEEE Transactions on Knowledge and Data Engineering，2010, 22（10）: 1345-1359.
[22] 张林朋, 杨红云, 钱政, 等. 基于改进的VGG16网络和迁移学习的水稻氮素营养诊断[J]. 中国农业大学学报, 2023, 28(6): 219-229.

[1]	吴秋兰, 陈雪飞, 陈超, 张峰, 王姝妹, 赵恒. 基于改进YOLOv5s的香菇菌棒污染识别方法[J]. 中国农机化学报, 2025, 46(2): 217-223.
[2]	李震, 岳强, 张冬会, 李志强, 于跃, 李嘉楠. 基于离散元法的香菇菌渣颗粒仿真物理参数标定[J]. 中国农机化学报, 2024, 45(10): 170-176.
[3]	马士鑫, 宋卫东, 丁天航, 王明友, 吴耀东, 周德欢. 脱袋转色期香菇菌棒物理力学特性研究[J]. 中国农机化学报, 2023, 44(9): 79-84.
[4]	陈维美, 刘馨蔚, 王铁伟, 徐文凯, 李娟. 基于两级融合深度学习的松材线虫病识别[J]. 中国农机化学报, 2023, 44(7): 214-219.
[5]	付健, 薛新宇, 孙竹, 徐阳. 油菜地块边界提取研究[J]. 中国农机化学报, 2023, 44(4): 137-144.
[6]	马士鑫, 宋卫东, 王明友, 丁天航, 王教领, 周德欢. 我国香菇菌棒脱袋装备研究现状与展望[J]. 中国农机化学报, 2022, 43(11): 50-55.
[7]	鲁梦瑶, 周强, 姜舒文, 王聪, 陈栋, 陈天恩, . 基于深度学习与多尺度特征融合的烤烟烟叶分级方法[J]. 中国农机化学报, 2022, 43(1): 158-166.
[8]	王教领;宋卫东;任彩红;吴今姬;王明友;丁天航;. 我国香菇干燥技术研究进展[J]. 中国农机化学报, 2021, 42(7): 76-83.