Image processing and analysis of this major discussion. Although some parts such as computer vision feature extraction can also be attributed to the image analysis in the past, but in view of their close links with the computer vision, as well as their source, not put them into image processing and analysis in the past. Similarly, there are also some computer vision can be classified to go. This is not important, just know that there is such a method can for their own use, or an inspiration, that was enough.
8. Edge Detection
Edge detection is a fundamental task in image processing. Traditional edge detection method is a gradient operator, especially Sobel operator, Canny edge detection, and based on the classic. Up to now, Canny edge detector and ideas are still widely used. Specific details on the Canny algorithm can book Sonka and canny their papers to find, you can also search online to. The fastest and most direct way is to look at the OpenCV source code, very easy to understand. In terms of edge detection, Berkeley Daniel J Malik and his students in 2004, PAMI proposed method is very good, of course, more complicated. In the case of the complexity of the less demanding, it is worth a try. MIT's Bill Freeman Steerable Filter early work in terms of the edge detection is also very good and easy to implement. Here are a few good literature, including the latest review. Edge detection is a problem in image processing and computer vision in any direction can not escape, deep research in this area can not be overemphasized.
[1980] Theory of Edge Detection
[1983 Canny Thesis] the Find Edge
[1986 PAMI] A Computational Approach to Edge Detection
[1990 PAMI] Scale-Space and Edge Detection a using Anisotropic Diffusion
[1991 PAMI] at The Design and use of Steerable Filters [1995 PR] Multiresolution edge detection techniques
[1996 TIP] Optimal edge detection in two-dimensional images
[1998 PAMI] Local Scale Control for Edge Detection and Blur Estimation
[2003 PAMI] Statistical edge detection_ learning and evaluating edge cues
[2004 IEEE] Edge Detection Revisited
[2004 PAMI] Design of steerable filters for feature detection using canny-like criteria
[2004 PAMI] Learning to Detect Natural Image Boundaries Using Local Brightness, Color, and Texture Cues
[2011 IVC] Edge and line oriented contour detection State of the art
translation
The best two-dimensional image edge detection --http: //tongtianta.site/upload
Author:
Abstract -