Title :
Automatic detection and correction for glossy reflections in digital photograph
Author :
Chang, Rong-Chi ; Tseng, Fei-Chun
Author_Institution :
Dept. of Digital Media Design, Asia Univ., Taichung, Taiwan
Abstract :
The popularization of digital technology has made shooting digital photos and using related applications a part of daily life. However, the use of flash, to compensate for low atmospheric lighting, often leads to overexposure or glossy reflections. This study proposes an auto-detection and inpainting technique to correct overexposed faces in digital photography. This algorithm segments the skin color in the photo as well as uses face detection and capturing to determine candidate bright spots on the face. Based on the statistical analysis of color brightness and filtering, the bright spots are identified. Finally, bright spots are corrected through inpainting technology. From the experimental results, this study demonstrates the high accuracy and efficiency of the method.
Keywords :
digital photography; face recognition; image colour analysis; image segmentation; statistical analysis; automatic detection; digital photograph; face detection; flash; glossy reflections correction; inpainting technique; skin color segmentation; statistical analysis; Digital cameras; Digital photography; Face detection; Filtering; Image analysis; Image color analysis; Lighting; Lubricating oils; Optical reflection; Skin; Digital inpainting; Digital photography; Face detection; Glossy reflection; Oily face; Skin detection;
Conference_Titel :
Ubi-media Computing (U-Media), 2010 3rd IEEE International Conference on
Conference_Location :
Jinhua
Print_ISBN :
978-1-4244-6708-2
DOI :
10.1109/UMEDIA.2010.5543933