An automatic nuclei cells counting approach using effective image processing methods

Manual counting of nuclei cells from histological images is considered tedious process, time-consuming and subjected to human errors. Therefore, automated the process of nuclei cells counting is become important and necessary for effective analyzing of histological images. Current systems and approaches of nuclei cells counting are based on color or grayscale images leading to inaccurate results and have several limitations. In this paper, we propose a novel accurate approach for automatic nuclei cells counting using effective image processing methods. The new techniques are designed based on image thresholding method, morphological image processing operations, and connected component algorithm. The new approach was evaluated experimentally on 37 images of a public data set of 100 histological images. The experimental results demonstrated that the approach achieved a high accuracy up to 89.5% compared with previous works. We concluded the effectiveness of the proposed approach for automatic counting of nuclei cells from histological images.