融合注意力与多尺度特征的城市街景实例分割  

Instances Segmentation of Urban Streetscape Incorporating Attention and Multi-scale Feature

作　　者：王军[1,2] 吕佳 程勇 WANG Jun;LYU Jia;CHENG Yong(School of Software,Nanjing University of Information Science&Technology,Nanjing 210044,China;Science and Technology Industries Division,Nanjing University of Information Science&Technology,Nanjing 210044,China)

机构地区：[1]南京信息工程大学软件学院,南京210044 [2]南京信息工程大学科技产业处,南京210044

出　　处：《计算机系统应用》2025年第1期90-99,共10页Computer Systems & Applications

基　　金：国家自然科学基金(41975183)。

摘　　要：城市街道场景实例分割算法可以显著提升城市环境感知和智能交通系统的准确性与效率,针对城市街景行人和车辆之间相互遮挡和背景干扰严重等问题,提出一种基于频率注意力机制和多尺度特征融合的实例分割模型FMInst.首先,构建一种高低频注意力机制进行交互编码从而增加高分辨率细节信息.其次,在Swin Transformer主干网络的Patch Merging层引入软池化操作,减少特征信息损失,有效提高小尺度目标分割结果.最后,结合MLP层构建多尺度的深度卷积,有效增强目标局部信息提取,提升实例分割精度.在Cityscapes公共数据集进行对比实验,结果表明FMInst的mAP提高1.2%,达35.6%,同时AP50提高2.2%,达61.4%,极大地改善实例分割的掩码质量和分割效果.Algorithms for the instance segmentation of urban street scenes can significantly improve the accuracy and efficiency of urban environment perception and intelligent transportation system.To address mutual occlusions between pedestrians and vehicles and significant background interference in urban street scenes,this study proposes an instance segmentation model,FMInst,based on a frequency attention mechanism and multi-scale feature fusion.Firstly,a high and low-frequency attention mechanism is constructed for interactive coding to increase high-resolution detail information.Secondly,a soft pooling operation is introduced into the Patch Merging layer of the Swin Transformer backbone network to reduce the loss of feature information and effectively improve the segmentation of small-scale targets.Finally,an MLP layer is combined to construct multi-scale deep convolution,which effectively enhances the extraction of local information and improves the segmentation accuracy.Comparison experiments conducted on the public dataset Cityscapes show that FMInst reaches an mAP of 35.6%,with an improvement of 1.2%,and an AP50 of 61.4%,with an improvement of 2.2%.The mask quality and the segmentation effect of the instance segmentation are greatly improved.

关 键 词：城市街景 实例分割 频率注意力机制 多尺度特征融合 小目标 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]