Mitsubishi Denki Kabushiki Kaisha

Kazuaki Kojima

Hiroaki Sugiura

Tetsuya Kuno

Takeshi Yamada

A method and apparatus for adjusting a color difference signal of an image includes a video signal processor. The video signal processor receives a luminance signal and color difference signals representing the image, and accesses a color saturation level from a color saturation level table stored in memory based on the color difference signals. The video signal processor then detects flesh tone in areas of the image based on the luminance signal and the color saturation level, and generates a flesh tone detecting signal based on results of the flesh tone detection.

Method and apparatus for detecting flesh tones in an image

The color video camera has a processing circuit which converts electrical signals obtained from an imager into a luminance signal and color-difference signals. The color video camera further includes a memory which stores a table of values designating color saturation levels and corresponding to a specific hue of a background color. The memory outputs a color saturation level based on the color difference signals obtained by imaging an object on a background of the specific hue. This color saturation level is compared with the luminance signal, and based on the comparison results the luminance signal and color difference signals are separated into components corresponding to the background area and components corresponding to the object area.

Superimposing circuit performing superimposing based on a color

A flesh-tone area is detected based on color-difference and luminance signals constituting video signals, and luminance correction, color correction, and aperture correction are performed only on the flesh-tone area or a human face area identified in the flesh-tone area. The setting of a focus area or the setting of a photometric area for iris control, automatic gain control, automatic shutter control, etc., in a video camera, is performed with respect to the flesh-tone area or the human face area. Furthermore, based on the color-difference and luminance signals constituting the video signals, a background area is detected, and the video signals are divided into components representing a background area and components representing an object area. An image signal of a desired hue or a still image is superimposed on the detected background area, or special processing is performed on the video signals representing the object area other than the detected background area.

Growing community of inventors

Nagaokakyo, Japan

Kazuaki Kojima