Noise Removal and Contrast Enhancement for X-Ray Images

X-ray image play s a very important role in the medical diagnosis. To help the doctors for diagnosis of the disease, some algorithms for enhancing X-ray images were proposed in the past decades. However, the enhancement of images will also amplify the noise or produce distortion of image, which are unfavorable to the diagnosis. Therefore, appropriate techniques for noise suppression and contrast enhancement are necessary. This paper proposed an algorithm including t wo-stage filtering and contrast enhancement for X-ray images. By using adaptive median filter and bilateral filter, our method is able to suppress the mixed noise which contains Gaussian noise and impulsive noise, while preserving the important structures (e.g., edges) in the images. Afterwards, the contrast of image is enhanced by using gray-level morphology and contrast limited histogram equalization (CLAHE). In the experiments, we evaluate the performance of noise removal and contrast enhancement separately with quantitative indexes and visual results. For the mixed noise case, our method is able to achieve averaged PSNR 39.89 dB and averaged SSIM 0.9449 ; for the contrast enhancement, our method is able to enhance more detail structures (e.g., edges, textures) than CLAHE.