摘要
为实现复杂背景图像中高精度边缘的准确提取,提出一种改进的单像素边缘提取算法。在改进的全卷积神经网络中,通过添加辅助输出层与采取多尺度输入的方式初步提取图像多像素边缘,并利用分水岭算法对多像素边缘进行细化重定位,从而获取图像单像素边缘。磁瓦图像上的应用结果表明,该算法具有较强的鲁棒性,能提取高精度且完整连续的单像素边缘。
To accurately extract high precision edges in complex background images,this paper proposes an improved single-pixel edge extraction algorithm.In the improved fully convolutional neural network,this method adds an auxiliary output layer and adopts a multi-scale input method to coarsely extract multi-pixel edges of an image.Then the watershed algorithm is used to refine and relocate the multi-pixel edges to obtain a high precision single-pixel edge of an image.Application results on magnetic tile images show that the algorithm has strong robustness and can extract complete continuous high precision single-pixel edges.
作者
刘畅
张剑
林建平
LIU Chang;ZHANG Jian;LIN Jianping(School of Mechanical Engineering,Tongji University,Shanghai 201804,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2020年第1期262-270,共9页
Computer Engineering
基金
2017年智能制造综合标准化与新模式应用项目“磁性材料智能制造新模式应用”
关键词
单像素边缘检测
全卷积神经网络
分水岭算法
距离误差
磁瓦图像
single-pixel edge detection
fully convolutional neural network
watershed algorithm
distance error
magnetic tile image
