mSkinColorCalib – Computer vision based-system for color calibration of mobile acquired skin lesion images


Images captured with different mobile phones have different colors due to different camera specifications and/or illumination conditions. For medical applications like skin lesion image acquisition, the correct representation of colors is a crucial requirement for the specialist visual analysis. Thus, robust color calibration of mobile-acquired images for skin lesion analysis can greatly improve its quality for risk assessment purposes.

The goal of the master thesis is to develop a low cost computer vision-based system for color calibration of mobile-acquired skin lesion images. The candidate should start by reviewing the existent methodologies for color calibration, choose the algorithms that best suits the requirements of this proposal and develop a mobile-based color calibration solution for skin lesion analysis.


For any additional information regarding this project, please contact us using the inquiries form.