Andrea Pennisi, Phd

In the previous post, we discussed about histograms and how to equilize images, in this post we are going to talk about how to compare histograms. Histogram Comparison is one of the oldest way of comparing images in order to classify them according to a reference image. The main core idea behind this technique is to “reduce” the image to a histogram representation and compare it with other image histograms by using a metric in order to provide a rank of similarity between to images.

An Image histrogram represents the frequency of occurrence of each gray-level value. Since it plots the number of pixels for each tonal value, it can be analyzed for finding peaks and valleys. Image histograms are present on many modern digital cameras. Photographers can use them as an aid to show the distribution of tones captured, and whether image detail has been lost to blown-out highlights or blacked-out shadows. For istance, an 8-bit grayscale image there are 256 different possible intensities, and so the histogram will graphically display 256 numbers showing the distribution of pixels amongst those grayscale values.

Hough transform is commonly used for detecting regular curves such as lines, circles, ellipses, etc. It is a technique that requires that the features we want to find, can be specified in a parametric form (e.g. by using a gemotrical equation). Hough transform is useful for computing a global descriptionm of a feature(s), where the number of classes is known in advance, by taking into consideration the local measurements that can be affected from noise.