Please use this identifier to cite or link to this item: http://hdl.handle.net/10266/1825
Title: Blocking Artifacts Reduction in JPEG Images
Authors: Malhotra, Puneet
Supervisor: Singh, Singara
Keywords: PSNR;JPEG
Issue Date: 6-Aug-2012
Abstract: With the advent of new technologies, it becomes indispensable to keep the existing technology updated and upgraded with new technologies because to replace the existing technology with new technology takes a lot of time and up gradation in the existing system. In the era of multimedia many researchers in different industries devote their time to improve the quality of the multimedia technologies and present the improved and highly sophisticated methods. Today lot of communication on internet and satellites take place in the form of images and videos. Daily millions of images of our earth, the moon, stars, planets and our ecological system are sent at every moment to the metrological centers and space research centers so that they can study and analysis this cosmos. In order to keep these communications and transmission alacritous the images need to be compressed by a sophisticated method. But to improve something could disrupt the other thing, hence on decompression, the recovered images carry some quality disorders like blocking artifacts, ringing effects and blurring which causes loss of information or we can say loss of quality. The main purpose of this thesis is to reduce the blocking artifacts in JPEG image up to a very extent keeping in mind that less computation efforts would be made. It has been observed that post processing of artifacts removal gives better image quality. The post processing is performed either in spatial domain or frequency domain. The proposed method is hybrid model of both spatial domain and frequency domain. In the proposed method, the post-processing is performed in two phases. In phase one, pre-filtering is implemented in spatial domain while in phase two, the selective modifications on DCT-coefficients are performed in frequency domain. The proposed method is tested on different gray-scaled standard images at different bit rates and qualitative measures of these images are analyzed. The collected numerical results are then plotted on graphs to show the effectiveness of the proposed approach.
URI: http://hdl.handle.net/10266/1825
Appears in Collections:Masters Theses@CSED

Files in This Item:
File Description SizeFormat 
1825.pdf2.18 MBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.