Digital radiography (DR) contrast balancing method and system

[0003] The technical problem to be solved by the present invention is to provide a method and system for equalizing the contrast of DR images, which solves the problem that existing DR images have large differences in brightn...

As can be seen from the above description, the beneficial effects of the present invention are: above-mentioned scheme sets suitable local area size according to body position information to image data by input module 1, and by the first selection module 2 and computing module 3 take pixel as After the rectangular area is selected in the center, the corresponding pixel grayscale maximum value and minimum value are calculated to obtain the first grayscale maximum value image and the first grayscale minimum value image; the above-mentioned image is filtered by the image filtering module 4 to obtain the second grayscale value. degree minimum value and the second gray-scale maximum value; finally, each pixel of the original image data is based on the corresponding second gray-scale maximum value and the second gray-scale value by the second selection module 5 and the histogram stretching module 6 The minimum value determines the range of the functional histogram, and the histogram stretching process is performed on each pixel according to the corresponding histogram range, so as to finally obtain the image data with balanced contrast while retaining the information of the natural gray level change of the tissue. Different human tissues can be evenly displayed in the image obtained by the system of the present invention, which is convenient for doctors to observe information of different tissues; and the contrast of the tissues is good, which saves the workload for doctors to manually adjust the window width and level to observe different tissues.

In step S5 and step S6, determine the corresponding local histogram range of each pixel point according to the corresponding second grayscale maximum value image and the second grayscale minimum value image to each pixel point of original image data, According to the histogram linear stretching method, each pixel is linearly stretched, and finally the balanced image is obtained. The local equalization method overcomes the limitation that the previous global stretching can not take into account the different densities of the tissue at the same time, and a...