AUTHOR=Zhu Dongqin , Feng Quan , Zhang Jianhua , Yang Wanxia TITLE=Cotton disease identification method based on pruning JOURNAL=Frontiers in Plant Science VOLUME=Volume 13 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.1038791 DOI=10.3389/fpls.2022.1038791 ISSN=1664-462X ABSTRACT=Deep convolutional neural networks have shown promising performance in plant disease recognition. However, these networks cannot be deployed on resource-limited smart devices due to their vast parameters and computations. To address the issue of deployability when developing cotton disease identification applications for mobile/smart devices, we compress the disease recognition models employing the pruning algorithm. The algorithm uses the γ coefficient in the Batch Normalization layer to prune the channels to realize the compression of DCNN. To further improve the accuracy of the model, we suggest two strategies in combination with transfer learning: compression after transfer learning or transfer learning after compression. In our experiments, the source dataset is famous PlantVillage while the target dataset is cotton disease image set which contains images collected from the Internet and takes from the fields. We select VGG16, ResNet164 and DenseNet40 as compressed models for comparison. The experimental results show that transfer learning after compression overall surpass its counterpart. When compression rate is set to 80%, the accuracies of compressed version of VGG16, ResNet164 and DenseNet40 are 90.77%, 96.31% and 97.23%, respectively, and the parameters are only 0.30M, 0.43M and 0.26M, respectively. Among the compact models, DenseNet40 has the highest accuracy and the smallest parameters. We further use the model to develop a cotton disease recognition APP on the Android platform and the average recognition time for a single picture handled by our mobile is 87ms.