作者: Kevin Manbeck , Donald Geman , Stuart Geman , Michael N Orton
DOI:
关键词: Mathematics 、 Color histogram 、 Color normalization 、 Color balance 、 ICC profile 、 Cluster analysis 、 Pixel 、 Color depth 、 Telecine 、 Computer vision 、 Artificial intelligence
摘要: A system and method for automatically adjusting color parameters during a film-to-digital transfer. An operator uses standard telecine equipment to preview scenes in order determine natural clustering of fields or scenes. Each cluster is processed separately. All the are treated as one unit sense that adjustments based on histogram, which computed by aggregating colors from all pixels cluster. Constraints specified Variables constraint may be determined an observing sample frames In addition, can nullify if desired. Within each cluster, adjusts red, green blue intensities digitizing device, gains these channels, other parameters. The first determines settings respect constraints then optimizes solution. solution optimized such digital distribution faithful possible film subject constraints. This achieved maximize entropy empirical, histogram among respecting