Abstract:Convolutional neural network (CNN)-based research has been successfully applied in remote sensing image classification due to its powerful feature representation ability. However, these high-capacity networks bring heavy inference costs and are easily overparameterized, especially for the deep CNNs pretrained on natural image datasets. Network pruning is regarded as a prevalent approach for compressing networks, but most existing research ignores model interpretability while formulating pruning criterion. To address these issues, a network pruning method for remote sensing image classification based on interpretable CNNs is proposed. More specifically, an original interpretable CNN with a predefined pruning ratio is trained at first. The filters, namely channels in the high convolutional layer, are able to learn specific semantic meanings in proportion to the predefined pruning ratio. The filters without interpretability are supposed to be removed. As for other convolutional layers, a sensitivity function is designed to assess the risk of pruning channels for each layer, and furthermore, the pruning ratio for each layer is corrected adaptively. The pruning method based on the proposed sensitivity function is effective and requires little computational costs to search abandoned channels without damaging classification performance. To demonstrate the effectiveness, the proposed method is implemented on different scales of modern CNN models, including VGG-VD and AlexNet. The experimental results, obtained on the UC Merced dataset and NWPU-RESISC45 dataset, prove that our method significantly reduces the inference costs and improves the interpretability of networks.

Pruning Convolutional Neural Networks for Remote Sensing Images

A Network Pruning Method for Remote Sensing Image Scene Classification

Loss Constrains Added Squeeze and Excitation Blocks for Pruning Deep Neural Networks

Structured Deep Neural Network Pruning by Varying Regularization Parameters.

Absorption Pruning of Deep Neural Network for Object Detection in Remote Sensing Imagery

Network Pruning for Remote Sensing Images Classification Based on Interpretable CNNs

Remote Sensing Image Target Recognition Based on Pruned Deep Neural Network Models

A Pruning Method Based on the Dissimilarity of Angle among Channels and Filters

Energy-Based CNN Pruning for Remote Sensing Scene Classification

Energy-based CNNs Pruning for Remote Sensing Scene Classification

Regularized Training Framework for Combining Pruning and Quantization to Compress Neural Networks

Multiobjective ResNet Pruning by Means of EMOAs for Remote Sensing Scene Classification

RFPruning: A retraining-free pruning method for accelerating convolutional neural networks

Multi-Objective Net Architecture Pruning for Remote Sensing Classification.

RemoteTrimmer: Adaptive Structural Pruning for Remote Sensing Image Classification

Data Customization-based Multiobjective Optimization Pruning Framework for Remote Sensing Scene Classification

Multi-grained Pruning Method of Convolutional Neural Network.

Three-Stage Global Channel Pruning for Resources-Limited Platform

An IQE Criterion-Based Method for SAR Images Classification Network Pruning.

Small Network for Lightweight Task in Computer Vision: A Pruning Method Based on Feature Representation

Network Pruning Using Sparse Learning and Genetic Algorithm