Machine Learning-Based Implementation of Image Corner Detection Using SVM Algorithm for Biomedical Applications

摘要

Support vector machine approach in machine vision with the help of OpenCV simulation tool is used for corner detection.Path of the maximum gray-level changes meant for every edge-pixel is calculated in the picture,this edge-pixel is represented by four-dimensional feature vectors.It is made up of count of the edge pixels in window center and has four directions since their maximum gray-level direction change.This feature vector and support vector are used for designing of support vector machine.For corner detection,it represents critical points in a classification.This algorithm is straight forward with less computational complexity.It has machine learning capability which gives good results.