Increasing the accuracy of diagnosing biological tissues using digital image processing
Abbas F. Nori, Dhamya K. Sarhan · Al-Kufa University Journal for Biology · 2017
In diagnosing diseases, researchers and specialists usually rely on direct viewing with a microscope at a tissue sample to identify good tissue from the patient. Therefore, this paper was concerned with improving the accuracy and illustrate of biological tissue images by relying on a computer program, which is MATLAB, to perform image processing operations for images obtained from the microscope camera to illustrate it in a good way, so that specialists can make the best diagnosis using these results. Since there are many images of tissues that differ in terms of lighting, this paper relied on several illustration processes in two programs. The first focused on using histograms, it was based on a comparison between the original image and the image that was processed by spreading the histogram lighting once, and again with an image that was processed several times to get the best result, in addition to comparison with binary images and grayscale images. The second focused on detecting the edges of the image and using the complementary image. Several types of edge detection were used, the first using the detection function, the second using the Sobel filter, and the third using the Canny filter, through the results of the histogram and visual observation, it was found that the method of distributing histogram lighting is the best in obtaining distinctive results, after that complement of the image.