RTU Research Information System
Latviešu English

Publikācija: Image Quality Enhancement for Skin Cancer Optical Diagnostics

Publication Type Scientific article indexed in SCOPUS or WOS database
Funding for basic activity Research project
Defending: ,
Publication language English (en)
Title in original language Image Quality Enhancement for Skin Cancer Optical Diagnostics
Field of research 2. Engineering and technology
Sub-field of research 2.2 Electrical engineering, Electronic engineering, Information and communication engineering
Research platform None
Authors Dmitrijs Bļizņuks
Ilona Kuzmina
Katrina Boločko
Alexey Lihachev
Keywords biophotonics, skin cancer, non-contact diagnostics, image quality
Abstract The research presents image quality analysis and enhancement proposals in biophotonic area. The sources of image problems are reviewed and analyzed. The problems with most impact in biophotonic area are analyzed in terms of specific biophotonic task – skin cancer diagnostics. The results point out that main problem for skin cancer analysis is the skin illumination problems, when imaged object surface is not uniform. Since it is often not possible to prevent illumination problems, the paper proposes image post processing algorithm – low frequency filtering. Practical results show diagnostic results improvement after using proposed filter. That allows using diagnostic system in fully automatic mode, without the need of additional dermatologist’s analysis. Along that, filter do not reduces diagnostic results’ quality for images without illumination defects. Current filtering algorithm requires empirical tuning of filter parameters. Further work needed to test the algorithm in other biophotonic applications and propose automatic filter parameter selection.
DOI: 10.1117/12.2297579
Hyperlink: http://spie.org/Publications/Proceedings/Paper/10.1117/12.2297579 
Reference Bļizņuks, D., Kuzmina, I., Boločko, K., Lihachev, A. Image Quality Enhancement for Skin Cancer Optical Diagnostics. Proceedings of SPIE, 2017, Vol. 10592, pp.1-8. ISSN 0277-786X. e-ISSN 1996-756X. Available from: doi:10.1117/12.2297579
Additional information Citation count:
ID 26037