电子科技2024，Vol.37Issue(4)：62-68,7.DOI:10.16180/j.cnki.issn1007-7820.2024.04.009
基于注意力机制和图像轮廓的实例分割算法
Instance Segmentation Based on Attention and Image Contour
摘要
Abstract
Based on image contour,the instance segmentation method uses fewer contour nodes to represent an object,which effectively reduces the number of algorithmic parameters and improves its operation efficiency.Howev-er,with the segmentation result of poor quality,it is no match for traditional pixel-by-pixel processing segmenta-tion algorithm in terms of accuracy.To improve the accuracy of the algorithm,it is of great necessity to introduce a refined model of the instance segmentation(Attend the Contour snake,AC-snake),which is based on image con-tour with a combination of attention mechanism.An improved Largekernel + is added to the backbone network to im-prove the receptive field of the model and extract richer feature information.The network structure at the contour ver-tex deformation stage is improved,and the Dual Channel attention(DC-attentio)module is combined to enhance the effective information of contour vertex,reduce the invalid parameters in the training network,and improve the de-tection accuracy and training speed.The experimental results show that in Cityscapes validation data set,the im-proved model proposed in this study has improved performance when compared with the original model.关键词
实例分割/图像轮廓/轮廓顶点/逐像素/注意力机制/大卷积核/感受野/特征信息Key words
instance segmentation/image contour/contour node/pre-pixel/attention meachanism/large kernel/receptive field/feature information分类
信息技术与安全科学引用本文复制引用
顾登华,顾春华..基于注意力机制和图像轮廓的实例分割算法[J].电子科技,2024,37(4):62-68,7.基金项目
上海市科委科技行动计划(20DZ2308700)Science and Technology Action Plan of Shanghai Municipal Science and Technology Commission(20DZ2308700) （20DZ2308700）
