@article{oai:kyutech.repo.nii.ac.jp:00003653,
 author = {Hossain, Shahera and Serikawa, Seiichi and 芹川, 聖一},
 journal = {Applied Mechanics and Materials},
 month = {Oct},
 note = {Perceptions of image surface are very challenging work for computer vision.Human can amazingly expert at recognizing the reflective properties of surfaces ofvarious materials which a robot can not do easily so far. Smoothly we can differentiate ashiny metallic sphere from the plastic sphere of similar dimensions and structure. In thispaper, various image surfaces are analyzed according to various image statistics for robotvision systems. Identification of synonymous objects under various real-worldillumination or other environments are very daunting task. However, this is verychallenging and crucial for machine vision systems. Both statistical analyses and humanevaluation by various subjects under rigorous illumination conditions, we find significantimprovement in our analysis and emphasis the importance of statistical evaluation ofsurfaces for computer vision. Our findings clearly demonstrate that skewness has directresemblance with the surface glossiness-level. Intensity histogram also shows crucial cluefor surface analysis.},
 pages = {422--429},
 title = {Statistical Analysis and Psychological Evaluation of Surfaces under Various  Illuminations},
 volume = {36},
 year = {2010},
 yomi = {セリカワ, セイイチ}
}