Spermatozoon Segmentation Towards an Objective Analysis of Human Sperm Morphology

The Computer-Assisted Sperm Analysis (CASA) Technique for Sperm Morphology Evaluation

“…For the same semen sample, different microscope and camera will get different images. This may influcence the results of CASA [5] [6] [7] [8]. Images with high quality are required to get good resluts.…”

“…To get the actual parameters, such as distance, area, etc, we need to obtain the conversion factor between pixel and micrometers. One of simple and effective way is finding a relation between a numaber of pixels and the measurement of a reference , usually a known object, in an image taken with the image acquisition set up with the standard working paramerers [5]. As sperms have a three-dimensional structure, the image might not reveal the accurate specifications of the sperm [9].…”

“…For different image acquisiton, the characteristics of image obtained might be various, the method used for enhanceing may also be different. A power law transformation over each compoent of RGB cube was used in [5], which was used to enhance the constituent parts (nucleus, acrosome, midpiece and tail) of sperms. Also a Closing and Opening algorithm to enhance image [9].…”

“…If the image is a RGB image, we need to transform it to a gray scale image at first. A thresholding method which was reported by Ostu in [10] is used in [5] [6]. The Ostu method was used to choose the threshold for image binarization in [6].…”

New hybrid approach for identification of spermatozoa in human semen sample using microscope image processing techniques

“…Henry Corrilo et.al. [6] was focused on finding the features of spermatozoa. The method called nth-fusion was introduced in his work, the framework of the segmentation algorithm for region of interest head, Acrosome, Nucleus and Mid-piece was segmented.…”