Resources

Databases

  • Laparoscopic Video Quality (LVQ) Database – direct link https://drive.google.com/file/d/1SoONeacp9vvihTY7zmWssG_cnVzx16oq/view
    Laparoscopic Video Quality Database (LVQ) contains 10 reference laparoscopic videos each of 10 seconds duration which are distorted with 5 different distortions at 4 different levels. The distortions include some of those often encountered during the laparoscopic surgery namely defocus blur, motion blur, uneven illumination, smoke and noise. The subjective scores in the database were obtained both from non-medical observers (29 in total) as well as medical observers (9 in total).

    Citation : Khan, Z.A., Beghdadi, A., Cheikh, F.A., Kaaniche, M., Pelanis, E., Palomar, R., Fretland, Å.A., Edwin, B. and Elle, O.J., 2020, March. Towards a video quality assessment based framework for enhancement of laparoscopic videos. In Medical Imaging 2020: Image Perception, Observer Performance, and Technology Assessment (Vol. 11316, p. 113160P). International Society for Optics and Photonics.

  • Colourlab Image Database:Image Quality (CID:IQ) – direct link https://folk.ntnu.no/mariupe/CIDIQ.zip 
    The Colourlab Image Database: Image Quality (CID:IQ) contain 23 pictorial images are selected as the reference images with six different distortions over 5 levels. The distortions are JPEG compression, JPEG2000 compression, Poisson noise, blurring, and two gamut mapping algorithms. CID:IQ contains the subjective scores of 17 observers. The database is available for download.

Citation : Xinwei Liu, Marius Pedersen, and Jon Yngve Hardeberg,  “CID:IQ – A New Image Quality Database,” To be presented at the International Conference on Image and Signal Processing 2014 (ICISP 2014), June 30-July 2, 2014, Cherbourg, Normandy, France. [URL]

Lead Principal Investigator (LPI)  : Azeddine Beghdadi
Co-PI : Muhammad Qureshi, Bilel Sdiri, Mohamed Deriche, Faouzi Alaya-Cheikh
Citation : A. Beghdadi, M. A. Qureshi, B. Sdiri, M. Deriche, F. Alaya Cheikh, « CEED – A Database for Image Contrast Enhancement Evaluation. CVCS 2018: 1-6, September 19-20, 2018, Gjøvik, Norway

  • Spectral Image Database for Quality (SIDQ) – direct link http://www.ansatt.hig.no/mariusp/sidq.zip
    The SIDQ contains nine original 160-band hyperspectral images (MATLAB *.mat files) of scenes representing pseudo-flat surfaces of different materials (textile, wood, skin. . . ) with a spectral range between 410 and 1000nm. Five spectral distortions were designed, applied to the spectral images and subsequently compared in a psychometric experiment, in order to provide a basis for applications such as the evaluation of spectral image difference measures. The resulting 45 reproductions and raw subjective scores are also provided in the database.Citation:  Steven Le Moan, Sony George, Marius Pedersen, Jana Blahova, and Jon Yngve Hardeberg. “A database for spectral image quality” in Image Quality and System Performance XII, San Francisco, CA, USA, February 2015, vol. 9396, p. 25, IS&T/SPIE.
  • Colourlab Image Database:Perceptual Projection Sharpness (CID:PPS) – direct link http://www.ansatt.hig.no//mariusp/CIDPPS.zip This Colourlab Image Database:Perceptual Projection Sharpness (CID:PPS) contains 7 original images distorted 6 levels of blur. They were shown using a projection system, and subjective scores have been gathered from 15 human observers.

Citation : Zhao, Ping, Yao Cheng, and Marius Pedersen. “Objective assessment of perceived sharpness of projection displays with a calibrated camera.” Colour and Visual Computing Symposium (CVCS), 2015. IEEE, 2015.

  • Detection thresholds in chrominance channels of natural images: This package (zip file) contains the images, the log-Gabor target, and the collected thresholds for detecting the target inserted in the Cr and Cb channels of natural sRGB images. For more info, please read the readme.txt inside the zip file, or the citation.
    Citation: Kitanovski, V. and Pedersen, M., 2017, September. Masking in chrominance channels of natural images—Data, analysis, and prediction. In Proceedings of the 10th International Symposium on Image and Signal Processing and Analysis(pp. 131-136). IEEE.
  • Colon capsule endoscopy images: The data contains colon capsule endoscopy images with pathologies and normal diagnosis from different parts of the colon. There are 30 images chosen by an expert for image enhancement comparison. There are four image folder that contain implementation of two image decomposition techniques along with original and our proposed method. Download link. Citation: Mohammed, A., Farup, I., Pedersen, M., Hovde, Ø. and Yildirim Yayilgan, S., 2018. Stochastic capsule endoscopy image enhancement. Journal of Imaging, 4(6), p.75.
  • EVA dataset: An explainable Visual Aesthetics dataset. This dataset includes 4070 images with 30-40 human annotations each, including overall aesthetic score and several aesthetic attributes. Please refer to this publication for further details: C. Kang, G. Valenzise, F. Dufaux, “EVA: An Explainable Visual Aesthetics Dataset”, 1st Joint Workshop on Aesthetic and Technical Quality Assessment of Multimedia and Media Analytics for Societal Trends (ATQAM/MAST’20), ACM Multimedia, Oct 2020, Seattle, USA Download link

Software repository

  • Traitim : Some image processing tools (low-level treatments) developed by Azeddine Beghdadi  and Alain Le Négrate (University Paris 13).
  • Smartflow : displacement field estimation and some image processing tools developed by  Jérôme Monteil (former PhD student) under the supervision of Prof. Azeddine Beghdadi.
  • QuickEval: online platform for doing subjective experiments. Direct link https://www.ansatt.hig.no/mariusp/quick