Image Quality Plot (BRISQUE, NIQE and PIQE algorithms)
Features:
Image Quality calculates the no-reference image quality score for the image using several methods. The range of the Quality Score is from 0 to 100. A smaller score indicates better perceptual quality.
Calculates the quality of the images as the function of the wavelength.
Steps: (could be used in random order)
Load the data using File → Open... The Quality score of the image is located in the text box under the image as shown below. This value is calculated using BRISQUE algorithm.
Go to the Image Statistics → Image Quality and select Image Quality Plot to visualize the image quality as the function across the entire z-axis (band/channel/wavelength). The available algorithms are: BRISQUE, NIQE, and PIQE.
Tip: You can save Quality plots spectra in Excel format using a Brush button from a Strip Toolbar. Hover over the spectral window to activate a Strip Toolbar and click the Brush icon. When clicked, the Brush icon will turn blue. Brush the entire or part of the plotted spectrum and right-click. Select Copy Data to Clipboard. Paste the copied data into Excel or a text file.
Additional Information:
The algorithm predicts the BRISQUE score by using a Support Vector Regression (SVR) model trained on an image database with the corresponding differential mean opinion score (DMOS) values. The database used for this algorithm contains images with known distortion such as compression artifacts, blurring, and noise, and it contains pristine versions of the distorted images. The image to be scored must have at least one of the distortions for which the model was trained.
PIQE (Perception-based Image Quality Evaluator) scores the image distorted due to blocking artifacts and Gaussian noise. The algorithm generates spatial quality masks that indicate the high spatially active blocks, noticeable artifacts blocks, and noise blocks in the image.
NIQE (Naturalness Image Quality Evaluator) measures the distance between the natural scene statistics (NSS) based features calculated from the image to the features obtained from an image database used to train the model. The features are modeled as multidimensional Gaussian distributions.
References:
BRISQUE:
Mittal, A., A. K. Moorthy, and A. C. Bovik. "No-Reference Image Quality Assessment in the Spatial Domain." IEEE Transactions on Image Processing. Vol. 21, Number 12, 2012, pp. 4695–4708.
Mittal, A., A. K. Moorthy, and A. C. Bovik. "Referenceless Image Spatial Quality Evaluation Engine."Presentation at the 45th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, 2011.
PIQE:
Venkatanath, N., D. Praneeth, Bh. M. Chandrasekhar, S. S. Channappayya, and S. S. Medasani. "Blind Image Quality Evaluation Using Perception Based Features", In Proceedings of the 21 st National Conference on Communications (NCC). Piscataway, NJ: IEEE, 2015.
Sheikh, H. R., Z. Wang, L. Cormack, and A.C. Bovik, "LIVE Image Quality Assessment Database Release 2 ", https://live.ece.utexas.edu/research/quality/ .
NIQE:
Mittal, A., R. Soundararajan, and A. C. Bovik. "Making a Completely Blind Image Quality Analyzer." IEEE Signal Processing Letters. Vol. 22, Number 3, 2013, pp. 209–212.
