计算机工程与科学2024，Vol.46Issue(8)：1503-1512,10.DOI:10.3969/j.issn.1007-130X.2024.08.019
一种基于注意力机制的轻量级语义分割
A lightweight semantic segmentation based on attention mechanism
摘要
Abstract
Semantic segmentation is a computer vision technique that requires extracting focused in-formation from a large number of images and then transforming this information into a clearer and easier-to-understand representation by means of a mask.Researchers are trying to find a balance in order to minimize the size of the model while ensuring its accuracy,which is currently a hot topic in designing lightweight network models.Currently,there are many challenges in image semantic segmentation techniques,such as segmentation discontinuity,incorrect segmentation,and high model complexity.To solve these problems,a lightweight semantic segmentation model based on attention mechanism is pro-posed.It uses freeze-thaw training,and the feature extraction network is MobileNetV2.To recover clearer target boundaries,a lightweight convolutional attention(CBAM)module is introduced in the output part of the atrous spatial pyramid pooling(ASPP)or channel attention(ECA-Net)in the decod-ing part.To solve the sample imbalance problem,the focal_loss loss function is introduced.Mixed ac-curacy is used,and the standard convolution in the output section is replaced with DO-Conv convolu-tion.Experiments and validations are conducted on the PASCAL VOC2012 and Cityscapes datasets.The model size is 23.6 MB,with mean intersection over union(mIoU)scores of 73.91％and 74.89％,and class-wise pixel accuracy scores of 82.88％and 84.87％respectively.This successfully achieves a balance between accurate segmentation and computational efficiency.关键词
语义分割/DeepLabV3+/MobileNetV2/CBAM/通道注意力Key words
semantic segmentation/DeepLabV3+/MobileNetV2/CBAM/channel attention分类
信息技术与安全科学引用本文复制引用
马冬梅,王鹏宇,郭智浩..一种基于注意力机制的轻量级语义分割[J].计算机工程与科学,2024,46(8):1503-1512,10.基金项目
国家自然科学基金(61961037) （61961037）
