A New Micro Genetic Algorithm Based Image Stitching Approach for Camera Arrays at Production Lines

Abstract

Nowadays, using image processing techniques on the image taken by camera at the top of production line for detection of faulty products is widespread. The quality of images can decline due to the distance between camera and production line while trying to get whole image of the bigger products. Therefore, in this study we employed camera array instead of only one camera and placed it at a nearer position to production line. As the images taken from cameras in camera array contain specific parts of the whole product, it is required to stitch images and obtain single image of the product before analyze it. Production line speed or the product position can be vary so it is required to find the optimal stitching points in order to get single image. As an optimization technique micro genetic algorithm (µGA) which refers to Genetic Algorithm (GA) with small population size and re-initialization process is popular for its proper and fast solutions. In this paper, we have developed a method using µGA known by its high convergence rate and low termination possibility at a local optimum to be used in image stitching process. Using µGA known by its high convergence rate, less need of computational source and rare premature solution provided us optimal and fast solutions. Experimental results show that µGA, which outperforms conventional GA, gives good results in a reasonable time and it can be used in image stitching process as an alternative.

Optimizasyon Tabanlı Adaptif Görüntü Mozaikleme Algoritması

“…Görüntü mozaikleme, aynı sahneye ait birden fazla görüntünün içerdikleri ortak alanlara göre çakıĢtırılarak daha çok detay içeren tek bir görüntü elde edilmesi iĢlemidir [14]. Görüntü mozaikleme sayesinde görüntü alacağımız alanın resimlerini birbiriyle çakıĢacak Ģekilde parça parça çeker ve birleĢtirirsek daha kaliteli ve yüksek pikselli resimler elde edebiliriz.…”

Adaptive Vision Based Condition Monitoring and Fault Detection Method for Multi Robots at Production Lines in Industrial Systems

“…In this study, we used Scale Invariant Feature Transform method [17]. After extracted SIFT features, it is searched corresponding interest points from second image to interest points from first image according to similarity [18]. In this stage second nearest neighbour distance rate is used to prevent mismatches [19].…”

Kuantum Uyarlamalı Genetik Algoritmalar için Çözüm Kalitesini Artıracak Yeni Bir Yaklaşım

“…Popüler bir optimizasyon yöntemi olan genetik algoritma (GA), günümüze kadar bir çok problemin çözümünde kullanılmıştır [14][15]. Evrimsel süreci örnek alan GA, birey adı verilen çözüm adaylarından oluşmaktadır.…”