用改进的知识蒸馏让轻量模型更准地识别土地利用图像
Improved Knowledge Distillation for Land-Use Image Classification
- 用VGG16教MobileNetV2，结合标签硬监督与KL散度+余弦相似度软监督
- 在三个数据集上达到99.04%准确率，优于单损失蒸馏和直接训练学生模型
- 适合需要高精度且低算力的土地分类应用，如遥感图像分析
本文提出一种改进的知识蒸馏框架，用于压缩深度卷积神经网络以实现土地利用图像分类。为在保持高分类精度的同时降低计算复杂度，采用教师-学生学习范式，由VGG16向轻量级MobileNetV2模型传递知识。该框架融合真实标签的硬监督与基于Kullback-Leibler散度和余弦相似度的软监督策略。在三个土地利用数据集上的实验表明，所提方法性能显著提升，准确率达到99.04%，优于基线学生模型训练及单一损失蒸馏方法，同时实现了显著的模型压缩。
原文摘要 · Abstract (English)
In the present article, an improved Knowledge Distillation (KD) framework has been proposed for efficient compression of deep convolutional neural networks for land-use image classification task. Motivated by the need to achieve competitive classification accuracy while reducing computational complexity, a teacher-student learning paradigm is adopted in which a VGG16 network transfers knowledge to a lightweight MobileNetV2 model. The proposed framework integrates hard supervision from ground truth labels with a soft supervision strategy that combines Kullback-Leibler divergence and Cosine Similarity losses. Experiments conducted on three land-use datasets show that the proposed KD-based method yields improved performance, and achieves an accuracy of 99.04%, outperforming both baseline student training and single-loss distillation approaches, while retaining substantial model compression.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。