This chapter presents objective criteria for the evaluation of image quality that are based on the models of visual perception. Two major classes of models for image quality have been considered-general models and models that have been designed specifically for image compression applications. The general models are more elaborate and computationally intensive. The coder-specific models are simpler and computationally efficient. The coder-specific models have been shown to be quite effective when used to predict the performance of the given coder. However, the real challenging task is to predict the performance of coders with different structures-that is, wavelet versus DCT-based coders. The general models are expected to be more effective in this case.
ASJC Scopus subject areas
- Computer Science(all)