Unsupervised Blind Quality Estimation of NSS Images using Efficient Feature Extraction
S Gopika, D. Malathi, J D Dorathi · Journal of Physics Conference Series · 2018
In present-day date, public work more with images than with textual content. Often, the image which is transmitted over the internet or in wireless network demand a quality analysis to be done before accepting the same at the user's end. A great quantity of work has been done in recent past to develop an unbiased image quality objective value that relate well with perceived quality measurement. In-spite of the tremendous effort, the success achieved is less in this direction. Through this paper, we discuss about the method, EBUQA, an objective analysis method by which the perceptual quality of Natural Scene Statistic image can be found. This method measures the amount of distortion and is quantized as a numeric parameter. Performance analysis and testing done on the method conforms well with the human observation.