| 摘要 |
A plurality of images captured in different exposure times are aligned in order of exposure time, and pairs each of two images adjacent on the line are set. Common regions in long- and short-time images contained in a pair of interest are specified based on the pixel value distributions of the respective images. A location error correction amount for the long-time image is calculated to make the common regions match each other when the long- and short-time images are composited with each other. The long-time image undergoes image processing based on the location error correction amount. A composite image is generated by compositing a plurality of images obtained by performing image processing for each pair. The generated composite image is output.
|
