Purpose. With the scientific and technological development as well as the plenty of image information and exchange, the edge detection and automatic identification of the target image have become increasingly intensive. Its practical application problems have raised higher and higher requirements on the image edge detection techniques. The research deals with the problems of detecting the ideal edges and determining every parameter of this algorithm. Methodology. We combined Ant Colony Algorithm (ACA) and Differential Evolution Algorithm (DE) and used it in the image edge detection. By analyzing the convergence time and optimization capacity of these two algorithms, we found the best method to combine and apply them to the image edge detection. Findings. We found a way of image edge detection based on ACA and DE. We firstly made a theoretical analysis of image edge detection and found the best combination point of ACA and DE according to their own features. Mainly, it integrated these two algorithms according to the set conditions and included the operating steps of DE in the early phase of image processing followed by the operations of ACA. Originality. We made a study of image edge detection based on the hybrid ant colony algorithm. We discussed how to detect the ideal edges of an image based on ACA and DE. The research on this aspect has not been found at present. Practical value. We integrate these two algorithms to extract the complete edge contour in order to make the detected edges continuous and the edge localization accurate. The experimental result shows that the hybrid algorithm not only enhances the adaptive ability and capacity of global optimization but also has excellent edge detection effect, greatly reducing the computation workload and time.

  Якщо, ви не знайшли інформацію про автора(ів) публікації, маєте бажання виправити або відобразити більш докладну інформацію про науковців України запрошуємо заповнити "Анкету науковця"