Image processing is one of the abilities of Artificial Intelligent that can help defining different types of images. Currently, in the medical field, experts are needed to define the location and size of tumors. Occasionally, these experts might make human errors, and as a result, the images from image processing must be used as Decision Support Systems (DSS) for the experts. Moreover, the DSS that were produced from the image processing in the current study will be used to help the experts to identify tumors with more precision and accuracy. In this study, 20 images from ultrasound that have low contrast were used. There are six different alternatives that can be used in image processing, to define the location of tumors, by using images from ultrasound: the ISODATA method, Li entropy, Minimum cross-entropy, Otsu’s method, the Triangle method, and Yen entropy. Thereby, these six methods were compared to find the best method. In each method, there are different ways, to find the threshold and present the image qualities, that are different from those of the original image. The ISODATA method, Li entropy, and Otsu’s method showed bright colors and have high contrast on the boundaries. The Triangle method and Yen entropy tend to cover overestimated edges. and transformed bright colors into black, which can prevent the locating of tumors. Minimum cross-entropy transformed most colors into white. In summary, it was found that Otsu’s method and the ISODATA method are the best options, for increasing the contrast in images, which make it easier to clarify tumors)