摘要 |
This object detection method is executed by an image analysis device configured by a computer having a processor and a memory, the method comprising: a first step in which the processor receives an input of an image and stores that in the memory; a second step in which the processor extracts, from the inputted image, a partial region which is one portion thereof and stores that in the memory; a third step in which the processor evaluates the symmetry of the partial region by using a characteristic amount extracted from the partial region and a mirror-reflection characteristic amount obtained through mirror-reflection conversion of the characteristic amount; and a fourth step in which the processor determines, as an object region in which a detection object is captured, a partial region evaluated as having higher symmetry than a predetermined threshold. |