换一批摘要Crop diseases seriously affect the quantity and quality of crop yields, causing huge economic losses and posing a serious threat to global food security. However, overuse of chemicals in traditional agriculture may be harmful to humans and livestock. Therefore, early diagnosis of crop diseases, which helps to avoid the heavy use of chemicals and provides feasible solutions, is much desired. We proposed a system to segment images with small and scattered regions. Experimental results show that our method not only outperforms state-of-the-art segmentation methods for the lesion segmentation task, but also shows its capability to perform well on more general tasks.
更多相关知识
关键词
weakly supervised learninglesion segmentationdisease detectionsemantic segmentationagricultureweakly supervised learninglesion segmentationdisease detectionsemantic segmentationagriculture
栏目名称
DOI
10.15302/J-QB-021-0272
发布时间
2024-07-26（万方平台首次上网日期，不代表论文的发表时间）
- 浏览16
- 被引0
- 下载0
相似文献
- 中文期刊
- 外文期刊
- 学位论文
- 会议论文
