With this application, you can analyze the quality of your photos in real-time before you take a picture. Or if you already have a photo library and want to browse through your pictures, this application will analyze your photos and give them a score, based on quality.
The quality of an image is defined by its ‘naturalness’, i.e., whether the image has been distorted. Examples of distortion include overexposure, underexposure, and motion blur. A natural image, also referred to as a ‘reference’ or ‘pristine’ image, is free of distortions that change the statistics of an image. Noticing this relationship, the Laboratory for Image and Video Engineering (LIVE Lab) at The University of Texas at Austin developed algorithms that use only these image statistics to determine image quality. This application uses the Natural Image Quality Evaluator (NIQE), which is the LIVE Lab’s very latest development and represents the state-of-the-art in image quality assessment.
The NIQE algorithm determines quality by calculating the distance of a given image’s statistical features to a statistical profile of reference images. This distance is then mapped to a numeric quality score using a curve computed from a database of images and corresponding human opinion scores.