Noise is unwanted information present in image that can harm the quality of image. A novel noise detection and reduction technique is proposed in this paper. Technique is used to detect and reduce the noise in the degraded images. The main approach of this proposed technique is detecting and reducing different types of noise present in degraded images. Paper introduces a two step process where first step is used to detect the type of noise which is present in degraded image and second step handles the errors and compensate for various noise sources common in multimedia application, such as Salt and Pepper, Gaussian and speckle noise.