分割
计算机科学
班级(哲学)
比例(比率)
背景(考古学)
像素
人工智能
差异(会计)
计算
图像分割
编码(集合论)
模式识别(心理学)
数据挖掘
算法
地图学
地理
会计
业务
考古
集合(抽象数据类型)
程序设计语言
作者
Xiaowen Ma,Mengting Ma,Chenlu Hu,Zhiyuan Song,Ziyan Zhao,Tian Feng,Wei Zhang
标识
DOI:10.1109/icassp49357.2023.10095835
摘要
Remote sensing images are known of having complex backgrounds, high intra-class variance and large variation of scales, which bring challenge to semantic segmentation. We present LoG-CAN, a multi-scale semantic segmentation network with a global class-aware (GCA) module and local class-aware (LCA) modules to remote sensing images. Specifically, the GCA module captures the global representations of class-wise context modeling to circumvent back-ground interference; the LCA modules generate local class representations as intermediate aware elements, indirectly associating pixels with global class representations to reduce variance within a class; and a multi-scale architecture with GCA and LCA modules yields effective segmentation of objects at different scales via cascaded refinement and fusion of features. Through the evaluation on the ISPRS Vaihingen dataset and the ISPRS Potsdam dataset, experimental results indicate that LoG-CAN outperforms the state-of-the-art methods for general semantic segmentation, while significantly reducing network parameters and computation. Code is available at https://github.com/xwmaxwma/rssegmentation.
科研通智能强力驱动
Strongly Powered by AbleSci AI