ABSTRACT

Quality of video processing equipment is directly related to the quality of images it produces. Unfortunately reliable objective yard sticks of image quality assessment are not available. Measures most widely used for assessing the quality of video images are the Root Mean Square (RMS) error measures and correlation between the original and the reconstructed images. Both measures do provide a general figure of image quality but under certain conditions provide results which are extremely unreliable.

The need for reliable image quality measures has led to the development of two algorithms to be used to assess image quality in terms of similarity between respective image blocks in the reference and the processed image frames. The first algorithm, known as the Minimum Distance Algorithm for Image Similarity, is based on the Minimum Distance Classification of pattern recognition. The second algorithm is logical in nature. It is known as the Significant Difference Density Algorithm. However, in order to achieve optimum performance in its application, this algorithm may require further modifications.

Development process of an Automatic Test Equipment to test JPEG Encoder Decoder card pairs is explained in detail. These card pairs are checked for functionality by comparing a reference image frame with its reconstruction after compression and decompression. Tests employed check for jitters or shifts induced within the image frames and calculation of signal to noise ratio between the reference and the reconstructed image blocks. Individual blocks are also compared using the Minimum Distance Algorithm for Image Similarity.

This system has been developed as a prototype of a stand alone Automatic Test Equipment.

Return to Omar’s library

Return to Omar’s interests

Return to Omar’s web page