KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Author: search.filters.author.Chivapreecha, SorawatSubject: search.filters.subject.Image compression

Search Results

Now showing 1 - 1 of 1
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Medical image compression and quality assessment
    (2013-01-01)
    Suapang, Piyamas
    ;
    Thongyoun, Methinee
    ;
    Chivapreecha, Sorawat
    In this research proposed to compressed medical images in single frame file format with two different techniques - JPEG and JPEG2000. The significant advantage of JPEG2000 over normal JPEG is that the error from JPEG2000 compression is smaller than the error from JPEG. Nevertheless, both methods share a similar mishap; when the compression ratio increases, they both generate more error than the processes on lower compression ratio do. What's more, the research proposed a neural network approach to image quality assessment. In particular, the neural network measures the quality of an image by predicting the mean opinion score (MOS) of human observers and using a set of key features extracted from the original and test images. Experimental results, using JPEG and JPEG2000 compressed images, show that the neural network outputs correlate highly with the MOS scores, and therefore, the neural network can easily serve as a correlate to subjective image quality assessment. The predicted MOS values have a linear correlation coefficient of0.9543, a Spearman ranked correlation of 0.9591.