.png){kind=icon}
返回
EMSCNet: Efficient Multisample Contrastive Network for Remote Sensing Image Scene Classification
.png){kind=icon}
PRE
AI
DOI：10.1109/TGRS.2023.3262840.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Significant progress has been achieved in remote sensing image scene classification (RSISC) with the development of convolutional neural networks (CNNs) and vision transformers (ViTs). However, high intraclass diversity and interclass similarity are still enormous challenges for RSISC. Metric learning can effectively improve the discriminative ability of deep representations by constraining the distance between features. Previous metric learning methods only optimize the feature space representation through metric function, ignoring the information interaction between samples. For complex scene images, similarity and discriminative knowledge need to be summarized from the multiple positive and negative pairs. We propose a novel efficient multisample contrastive network (EMSCNet) to integrate knowledge from multiple samples. Specifically, we construct a dynamic dictionary with momentum updates to mine positive and negative pairs from the entire dataset. Then, the similarity and discriminative knowledge between samples are summarized by introducing a contrastive module. Finally, the knowledge of the contrastive module is transferred to the backbone classifier through knowledge distillation. The proposed contrastive module can be easily embedded into the training process of CNNs or ViT and removed during inference. Experimental results conducted on three datasets demonstrate the effectiveness of the proposed approach.
Keyword:
Measurement
Feature extraction
Dictionaries
Remote sensing
Knowledge engineering
Transformers
Semantics
Convolutional neural networks (CNNs)
knowledge distillation
metric learning
remote sensing image scene classification (RSISC)
vision transformer (ViT)
期刊
IF:
8.6
论文数:
2.1W
被引数:
10.7W
机构
引用论文
分享收藏MGML: Multigranularity Multilevel Feature Ensemble Network for Remote Sensing Scene Classification
OAAI
OAAI分享收藏Crystalline‐State Reaction with Allosteric Effect in Spin‐Crossover, Interpenetrated Networks with Magnetic and Optical Bistability具有磁和光学双稳态的自旋交叉，互穿网络中具有变构效应的晶态反应
PREAI
分享收藏High-Resolution Remote Sensing Image Scene Classification via Key Filter Bank Based on Convolutional Neural Network基于关键滤波器组的卷积神经网络高分辨率遥感影像场景分类
PREAI
分享收藏Training Convolutional Neural Networks with Multi-Size Images and Triplet Loss for Remote Sensing Scene Classification多尺寸图像和三元组损失训练卷积神经网络用于遥感场景分类.png)
SENSORS
OAAI
SENSORSIF3.5
OAAI分享收藏
查看更多内容