DETEKSI LINGKARAN DENGAN MENGGUNAKAN IMPROVED HARMONY SEARCH ALGORITHM
Daftar Isi:
- Automatic circle detection of digital images has received much attention during the last few years in the field of Computer Vision as some of the efforts are aimed at optimal loop detectors. This study presents an algorithm for automatic detection of circular shapes which considers the whole process as an optimization problem. This approach is based on Improved Harmony Search Algorithm (IHSA), an improvised algorithm of meta-heuristic optimization derivatives inspired by musicians. The algorithm uses a three-point encoding as a candidate loop (harmony) above the edge-only image. The objective function serves to evaluate the harmony memory if the candidate circle really exists in the edge image. Guided by the value of an objective function, a set of candidate circles encoded using IHSA so that it can match the original circle in the input image. Experimental results from multiple tests on synthetic and natural images with varying complexity ranges have been included to validate the proposed engineering efficiency regarding accuracy, and speed.