Noises in the image results from the scanned image have a serious impact on the final decision made by the Expert. Globally, huge millions of decision made based on the scanned results. This paper focuses on the noise replacement aspect by an image filtering algorithm to perform noise removal automatically. The application has developed based on the fusion of both median and modified Discrete Wavelet Transform (mDWT) filtering algorithm and it is tested on variegated dataset images. Such a de-noised image performance has been analysed by the various performance measures such as Peak Signal to Noise Ratio (PSNR), Normalized Absolute Error (NAE), Structural Context (SC), Normalized Absolute Error (NAE), Structural Similarity Index Measure (SSIM), Normalized Cross correlation (NK) and Maximum Difference (MD). The fused filtering algorithm plays a key role in image processing, since it acts as a pre-processing step.