| 注册
首页|期刊导航|华中农业大学学报|基于XMem-SimAM的半监督猪只视频分割方法

基于XMem-SimAM的半监督猪只视频分割方法

陈萌放 徐迪红 李国亮 刘小磊 周明彦 黎煊

华中农业大学学报2025,Vol.44Issue(2):17-28,12.
华中农业大学学报2025,Vol.44Issue(2):17-28,12.DOI:10.13300/j.cnki.hnlkxb.2025.02.003

基于XMem-SimAM的半监督猪只视频分割方法

XMem-SimAM based semi-supervised video segmentation of pigs

陈萌放 1徐迪红 2李国亮 1刘小磊 3周明彦 2黎煊2

作者信息

  • 1. 华中农业大学信息学院,武汉 430070||农业农村部智慧养殖技术重点实验室,武汉 430070
  • 2. 农业农村部智慧养殖技术重点实验室,武汉 430070||华中农业大学工学院,武汉 430070
  • 3. 湖北洪山实验室,武汉 430070
  • 折叠

摘要

Abstract

The dynamic feeding and growth process of breeding pigs during the performance testing was used to solve the problem of accurate segmentation of pigs caused by complex environments in pig farms,dynamic growth of pigs,and changes in body size.A pig video dataset consisting of 234 video se-quences was constructed.A XMem-SimAM based semi-supervised video segmentation of pigs was pro-posed.The ability of model to extract temporal information at different scales was improved and the tempo-ral features of pigs'dynamic movements were captured by introducing SimAM attention for multi-scale fea-ture fusion.The spatial-channel attention module was used to enhance the model's extraction of temporal se-mantic feature weights.The strategy for multi-scale feature fusion and upsampling module were optimized.The temporal correlation information in video sequences was fully utilized to improve the segmentation ac-curacy of pigs in videos at a fine-grained level.The results of testing and comparison showed that the Jacca-rd index,contour accuracy F-score,average metric J&F,and the Dice coefficient of of Xmem-SimAM model on the pig video dataset was 96.9,95.8,98.0,and 98.0,superior to that of video object segmenta-tion methods including MiVOS,STCN,DEVA,and Xmem++,demonstrating its outstanding perfor-mance of segmentation.The processing speed reached 58.5 frames per second,with a memory consump-tion of 795 MB at the stage of reasoning,achieving a good balance between the efficiency of processing and the utilization of resource.The proposed method can be applied to video segmentation of dynamically grow-ing pigs in the complex environments of a pig farm.

关键词

半监督/视频分割/猪只/SimAM注意力

Key words

semi supervised/video segmentation/pigs/SimAM Attention

分类

信息技术与安全科学

引用本文复制引用

陈萌放,徐迪红,李国亮,刘小磊,周明彦,黎煊..基于XMem-SimAM的半监督猪只视频分割方法[J].华中农业大学学报,2025,44(2):17-28,12.

基金项目

生猪现代育种技术研发及新品种选育(HBZY2023B006-03) (HBZY2023B006-03)

武汉市生物育种重大专项(2022021302024853) (2022021302024853)

华中农业大学-中国农业科学院深圳农业基因组研究所合作基金项目(SZYJY2022031) (SZYJY2022031)

华中农业大学学报

OA北大核心

1000-2421

访问量0
|
下载量0
段落导航相关论文