Image Enhancement Using Multilevel Contrast Stretching and Noise Smoothening Technique for CT Images

Image enhancement is the process of improving the visual quality of an image. This paper proposes a new algorithm called Multilevel Contrast Stretching and a Noise smoothening technique for enhancement of medical images. The scanned images or X-ray images are possibly affected by illumination, sharpness, contrast and noises, due to this degradation, and as result the diagnosis of disease becomes difficult. The proposed algorithm works effectively for medical applications. The initial step is transforming the medical image using Discrete Wavelet Transformation, the second step consists of enhancing the image by using multilevel contrast stretching algorithm. The Multilevel Contrast Stretching algorithm identifies the regions featuring homogeneous intensity and bearing contrast to their adjacent neighbors. The image split into two sub images, namely object-approximation image selected by using inter object level and object-error image chosen by intra object level. Then the final step involves eliminate the noise by using the median filter. The Proposed enhancement method is implemented in MATLAB and more medical images of different parts of the body are used for evaluating the performance. The resultant image is enhanced, sharpened and edges preserved, increasing the quality without distorting the images.