Abstract—Makler counting chamber is commonly used microscope kit for the evaluation of semen cells in terms of number and motility analyzes. Mainly, a physically located micron-size grid structure provides the generalization of observations over entire samples owing to the globally accepted criteria by academicians and experts. Evaluation process is mostly realized manually, which can cause mistakes. At first glance, mistakes can be ignored, but when it is generalized by the Makler criteria, results can mislead the diagnoses. In this sense, an autonomous analysis of Makler images is necessity instead of manual observation. This study specifically focuses on grid detection of Makler images, which is planned as the first phase of mentioned system. In this manner, a new approach, combination Hough Transform and K-Means clustering, is presented and tested over synthetically formed Makler images with different angles. Furthermore, pattern-matching idea is performed on the angle detection and evaluation steps.
 

Index Terms—Makler images, microscopic images, Hough transform, K-means clustering, pattern matching, grid detection, semen analyses

Cite: Hamza O. Ilhan and Fethullah Karabiber, "Computer Based Grid Detection on Makler Images," International Journal of Signal Processing Systems, Vol. 4, No. 4, pp. 323-327, August 2016. doi: 10.18178/ijsps.4.4.323-327