Paper
Novel Adaptive Truncated Average Filter for De-noising Grey and Colour Images
- Authors:
- Geeta Hanji; M. V. Latte
- Abstract
- Digital images are quite often affected by different noise types that mask important details of the image, during the process of acquisition, transmission, storage, and retrieval. Such images, if not processed properly, become useless for subsequent image-processing operations like segmentation, classification, etc. Hence, the improvement of noisy images becomes an important operation in the field of image processing. Different de-noising filters have been proposed in the literature to suppress the effects of each type of image noise. The work presented in this paper primarily aimed at negotiating the impulse noise from digital grey and colour images. A novel truncated average filter proposed through this paper, to address low-to-high range impulse noise resulted in excellent performance in terms of subjective and objective measures compared to state of the art filters. The effectiveness of the proposed method was also evaluated with standard grey and colour images with Gaussian and random noise (or mixed Gaussian-impulse noise (MGIN)) of varying densities. Experimental results were presented with an analysis of the performance of various other standard and competitive noise reduction filters. From the comparative analysis, it was evident that the proposed Adaptive Truncated Average Filter method out-performed the other methods, especially in cases of high-density salt-and-pepper and random-valued impulse noises.
- Keywords
- Truncated Average,Salt and Pepper Noise (SPN); Random Valued Impulse Noise (RVIN); Mixed Gaussian Impulsive Noise (MGIN); Peak-Signal-to-Noise Ratio (PSNR)
- StartPage
- 1
- EndPage
- 11
- Doi
- 10.18005/JCEN0301001