Signal converting apparatus

摘要

A color region extracting portion detects a specific color region from an input foreground image signal. A mapping portion categorizes an output signal of the color region extracting portion as a class. A high level resolution signal is generated with a predicting coefficient pre-learnt corresponding to the class. A threshold value Th is supplied from a terminal to a stretching portion. The stretch portion performs a stretching process. The stretching portion supplies a coefficient k as a key signal to a multiplier and a compensation signal generating portion. The compensation signal generating portion supplies a coefficient (1-k) to a multiplier. The multiplier multiplies the coefficient k by the coefficient (1-k) and outputs the result as a composite image of the foreground image and the background image.