摘要
近年来,深度学习在计算机视觉领域中的应用成效显著,新的深度学习方法和深度神经网络模型不断涌现,算法性能被不断刷新,基于深度学习的图像实例分割方法取得了跨越性进展,已成为处理图像的有力工具。为更好地促进深度学习实例分割算法的研究发展,对该领域的研究进展做了系统的梳理总结。首先,根据图像实例分割方法的过程和特征,分别从两阶段和单阶段的角度介绍对基于深度学习的图像实例分割研究进展;随之,介绍常用的评价指标;最后,结合实例分析分割技术当下存在的不足,提出可行的解决方案,并展望了实例分割技术的发展未来。
The application of deep learning in the field of computer vision achieves significant results in recent years,with new deep learning methods and deep neural network models constantly emerging,and algorithm performance constantly being refreshed.The image instance segmentation method based on deep learning makes significant progress and becomes a powerful tool for image processing.In order to better promote the research and development of deep learning instance segmentation algorithms,a systematic review and summary of the research progress in this field is conducted.Firstly,based on the process and characteristics of image instance segmentation methods,the research progress of image instance segmentation based on deep learning is introduced from a two-stage and a single-stage perspective.Subsequently,commonly used evaluation indicators are introduced.Finally,based on the current shortcomings of instance segmentation technology,feasible solutions are proposed,and the future development of instance segmentation technology is prospected.
作者
孙鹏
孙传聪
徐要要
邹田甜
吴翠杨
甄珍
Sun Peng;Sun Chuancong;Xu Yaoyao;Zou Tiantian;Wu Cuiyang;Zhen Zhen(Department of Medical Devices,Shandong Drug and Food Vocational College,Weihai,Shandong 264200,China)
出处
《机电工程技术》
2024年第8期1-6,共6页
Mechanical & Electrical Engineering Technology
基金
山东省教育科学研究项目(19SR001)
山东省人才服务行业协会研究项目(2020094)。
关键词
计算机视觉
实例分割
深度学习
图像分割
评价指标
computer vision
instance segmentation
deep learning
image segmentation
evaluation indicators
