| 摘要 |
<p>A method of identifying pixels of a given color in a field of YUV video manipulates color difference signals (R-Y, B-Y) corresponding to defining axes (U,V) of a color space to maximize the video signal in a region of interest of the space, and minimize the signal in all other regions. Additionally, the color difference signal corresponding to one axis may be rotated toward the other axis, or the entire color space may be rotated to bring the region of interest onto or near one of the axes. The signal gains and rotations may be carried out in the signal processing circuits (52) of a conventional color camera (50) so that only one comparator (90) is needed to detect the target color. After rotation and application of gain, the color difference signal defining the region of interest is compared with a threshold value to determine if a pixel is of the target color. A video camera may be mounted on an agricultural sprayer to detect green weeds on the brown soil, by identifying the green pixels in the image captured while the sprayer travels in the field. A spray material may be released upon the weed, once the weed is detected.</p> |
