摘要 |
A digital image in CMOS camera images is processed by first generating an output signal g from a CMOS camera. Then its spatio-temporal gradient (gx, gy, gt) is derived from the output signal. A time constant c and a local object shift (ux, uy) are established from prior knowledge. Finally a target signal value q is calculated from the output signal g as <?in-line-formulae description="In-line Formulae" end="lead"?>q=(gx*ux)+(gy*uy)+(g*-1*c)+gt.<?in-line-formulae description="In-line Formulae" end="tail"?> |