节点文献
基于改进的DOG算子在图像边缘提取中的应用研究
Application of Improved DOG Operator in Image Edge Extraction
【摘要】 边缘检测是图像处理领域中最基本的问题,其目的是标识数字图像中亮度变化明显的点,进而提取出有价值的信息。通过对现有经典图像边缘提取算法研究,提出一种采用小尺寸卷积核的DOG(Different-of-Gaussian)边缘提取算法,仅需要对图像进行一次卷积运算。实验仿真结果表明,该算法具有较快的处理速度并保留原图像的细节信息,同时在图像噪声较少且细节密度较低的情况下也可有效提取图像边缘信息。
【Abstract】 Advantages and disadvantages of classical edge extraction algorithms are analyzed in this paper. An improved DOG(Different-of-Gaussian) edge extraction algorithm is proposed which can effectively preserve the details of the image and has fasterprocessing speed. The image edge can be effectively extracted with low image noise and low detail density.
【关键词】 拉普拉斯算子;
索贝尔算子;
高斯模糊;
高斯差分算子;
【Key words】 Laplacian operator; Canny operator; Gaussian blur; DOG operator;
【Key words】 Laplacian operator; Canny operator; Gaussian blur; DOG operator;
- 【文献出处】 电脑知识与技术 ,Computer Knowledge and Technology , 编辑部邮箱 ,2018年06期
- 【分类号】TP391.41
- 【被引频次】14
- 【下载频次】343