| Budget Amount *help |
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2013: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2012: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
|
|---|
| Research Abstract |
Automatic color recognition technology that can correctly discriminate a categorical color in various environments like a human is required. If we want to make a computer vision system able to recognize color like a human, we must consider human visual characteristics such as categorical color perception and color constancy in the color recognition system. To create the model, the relationship between chromaticity of color chips under different illuminations and categorical color perception of the color chips under these illuminations by a human has be learned using a structured neural network.
We propose a new model with modified training data for high recognition performance and perception of multiple colors. In addition, this study proposes a method of recognition of object colors under varying illumination condition in video. The recognition function is the categorical color perception model, and it use the depth information from RGB-D camera.
|
