抽象的な

Novel Approach for Image Edge Detection

Pankaj Valand, Mayurdhvajsinh Gohil, Pragnesh Patel

Edges of an image are considered crucial information that can be extracted by applying detectors with different methodology. Finding better algorithms for edge detection is still an active area of research. Here a novel GAACO hybrid algorithm is proposed for finding edges of an image. GA is applied to randomly generated individuals, which contain a set of randomly generated points, and before the crossover step is taken ACO is applied to the selected parents for improving the quality of the points. The developed edge detector is compared objectively to traditional edge detectors Canny, Sobel, Roberts and Perwitt using Ground Truth Estimation process. Subjective comparison is left for the readers to judge. In the proposed algorithm, GA-ACO hybrid algorithm performs in par with the traditional methods such as Canny, Sobel, Roberts or Perwitt, used for image edge detection. It was observed that the time taken by this algorithm for producing output was comparatively higher than other mentioned methods. This algorithm shows potential in dealing with dynamic input, where input image is constantly changing. A multi-processor environment can solve the time issues observed in experimentation.