摘要
边缘检测是计算机视觉中非常重要且实用的图像处理方法,被应用在各个领域;然而在图像采集或传输过程中,由于外界环境的干扰,容易出现结果边缘检测率较低或者伪边缘现象,学者们为此提出了很多改进方法;但是通用的边缘检测方法很少,现有的算法都是以处理特定场景或特定情况下的问题为目的;Kirsch联合高低双阈值的RGB图像边缘检测算法正是针对上述问题提出的;首先,提取原图RGB色彩空间下的不同分量图,对每个分量图利用改进的Kirsch算子求取边缘强度;然后利用高低双阈值划分图像的边缘点和背景点,得到不同色彩空间的边缘结果;最后对不同分量的边缘检测结果进行融合,得到最终的边缘结果;利用基准数据集BSDS500数据集中的200张测试图像对算法进行验证评估;实验结果表明,文章算法相比于其他算法检测到的边缘更加清晰,细节更加完整,边缘连贯性更好,检测率更高,适用范围更广。
Edge detection is a very important and practical image processing method in computer vision, it is widely used in various fields. However, due to the interference of external environment, image acquisition or transmission process easily has the phenomenon of low detection rate in the edge or pseudo-edge, existing detection methods cannot be used for general edge detection, and they are mainly applied in specific scenarios or situations. Aimed at this problem, a red, green and blue(RGB) image edge detection based on Kirsch combined with high and low double thresholds is proposed. Firstly, the different component maps under the RGB color space of the original image are extracted, and the improved Kirsch operator for each component map is used to obtain the edge intensity. Then, the high and low double thresholds are used to divide into the edge and background points of the image, and obtain the edge results of different color spaces. Finally, the edge detection results of different components are fused to obtain the final edge results. The algorithm is verified by 200 test images in BSDS500 dataset, the experimental results show that compared with other algorithms, the proposed algorithm has clearer edges, more complete details, better edge coherence, higher detection rate and wider application range.
作者
魏雨
黄玉蕾
WEI Yu;HUANG Yulei(School of Intelligent Science and Information Engineering,Xi'an Peihua University,Xi'an 710125,China)
出处
《计算机测量与控制》
2023年第3期95-101,共7页
Computer Measurement &Control
基金
陕西省教育厅科研计划项目(21JK0822)。
