Effects of Other Color Spaces on Shape from Focus

DOĞAN H. , Baykal E. , ERCİN M. E. , ERSÖZ Ş. , EKİNCİ M.

4th International Conference on Computer Science and Engineering, UBMK 2019, Samsun, Turkey, 11 - 15 September 2019, pp.42-46 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/ubmk.2019.8907184
  • City: Samsun
  • Country: Turkey
  • Page Numbers: pp.42-46
  • Keywords: 3D shape reconstruction, cytopathological analy-sis, shape from focus, depth of field, light microscopic system, DEPTH, MICROSCOPY

Abstract

Shape From Focus (SFF) that creates the 3D shape of the object using a 2D image sequence with different focus field has been actively studied in the literature for many years. Classical SFF strategies consist of three basic stages: (1) 2D image sequence creation, (2) focus measure extraction and (3) maximum focus measure estimation. The classical SFF strategies utilize only gray levels of pixels in order to compute the pixel focus measures in the image sequence (the second stage of SFF). Our study provides that the other color spaces (HSV, YIQ and YCbCr) are effective on the SFF. In order to evaluate the performance of color spaces more realistic, the dataset from specimen used for cytopathological analysis is created. 3D shape reconstructions obtained using real color microscope images are tested with two different non-reference assessment procedures, and the effects of other color spaces are shown by both objective and visual results.