摘要 |
An image processing method and device by which objects (6) supplied successively on a conveyor belt (1) or other means of conveyance are imaged individually and image processing is performed on the images so obtained. Image processing is performed, for example, to compare markings on an object image with a model image to identify variations between the two images. If variations are detected, the model image in memory can be modified or replaced with the input object image without shutting down an entire production line in order to create an entirely new model image. In this way the recorded model data can be updated without interrupting the flow on the belt (1), and appropriate correction data can be recorded efficiently. <IMAGE> |