Abstract: Multiple distortion assessment is a big challenge in image quality assessment (IQA). Here developing a no reference IQA model for multiply-distorted images. The NSS features, which are sensitive to each distortion, make a difference in quality. Multiple distortions are impaired by varying degree of distortions like blur, uneven illumination, noise and jpeg. The basis of model is to create different IQA metrics that are sensitive to blur, jpeg, noise and uneven illumination. The output of these metrics is combined to predict quality score. Subjective score is expensive and time consuming; due to that reason objective assessment should be taken for assessment of images. Applications in liveness detection and can be used for direct image recapture or correction.
Keywords: No-Reference Quality Assessment, Multiple Distortion, Image Quality Metrics, NSS features.