Abstract
A histogram is a type of statistical graph with frequency plotted on the vertical axis and classes on the horizontal axis. Histograms measure the frequency of pixel values in images and provide a visual aggregation of the distribution of continuous numerical variables. The x-axis of the histogram is the ‘bin’, which divides the pixel values into ranges for each fixed interval. A bar is drawn for each ‘bin’ to represent the density of pixel values and the height of the bar represents the number of pixels within that range. Visual recognition of the distribution of pixel values provides important information on image characteristics and stable image quality. Specifically, the maximum, minimum and average pixel values, degree of separation, brightness, and contrast can be determined. In addition, image processing or compression of the original image results in a histogram that reflects the processing. Therefore, clinical image evaluation using histograms has also been attempted. This paper reports on the use of histogram analysis and image processing in mammography.
