Paper Info
Title
OECF characterization of a non-linear HDR Color Camera for Automotive Applications.
Abstract
Video cameras used in automotive applications such as driver assistance systems are confronted with natural scenes where light levels vary widely, and the luminance range within scenes can also be extremely wide. Only sensors that capture the full dynamic range of natural scenes in a single frame with sufficient image quality performance and at low cost are suitable for automotive applications. This is accomplished by CMOS technologies that extend the pixel's dynamic range by non-linear highlight compression, but can then cause scene dependent color balance errors to appear in certain scene elements. In silver-halide color reversal films, color balance is maintained over a wide dynamic range by tightly controlling the tone reproduction curves in each color channel. Color correction in a non-linear HDR camera requires the knowledge of its tone reproduction curves. We adapted the ISO 15739 test chart method to HDR requirements to realize its potential to provide a one-step measurement which would characterize the entire image capture system including optics and image processing as well as realistic photospace conditions. This paper reports initial OECF and color balance data from a direct non-linear HDR color camera for automotive applications.
Year
Venue
Field
2007
FIFTEENTH COLOR IMAGING CONFERENCE: COLOR SCIENCE AND ENGINEERING SYSTEMS, TECHNOLOGIES, AND APPLICATIONS, FINAL PROGRAM AND PROCEEDINGS
Computer vision,Nonlinear system,Computer science,Artificial intelligence,Automotive industry
DocType
Citations 
PageRank 
Conference
0
0.34
References 
Authors
0
2
Name
Order
Citations
PageRank
Dirk W. Hertel100.34
Edward Chang212.71