ON-VEHICLE IMAGE PROCESSING DEVICE AND ITS VIEWPOINT CONVERSION INFORMATION GENERATION METHOD

摘要

It is possible to provide an on-vehicle image processing device capable of easily obtaining a virtual image appropriate for a vehicle type and its viewpoint conversion information generation method. A projection model storage unit (41) stores position information on a plurality of points on a predetermined projection model. An operation unit (30) references the projection model storage unit (41) so as to acquire position information on a virtual image correspondence point as a point corresponding to a pixel on a virtual image, in a domain on the projection model specified as an object of the virtual image by an output range specifying unit (20). The operation unit (30) obtains a pixel on the real image corresponding to the virtual image correspondence point, acquires a mapping table indicating correspondence between the pixel on the virtual image and the pixel on the real image, and records it in a mapping table storage unit (42). An image synthesis unit (50) converts the image inputted from an imaging unit (10) into a virtual image viewed from a predetermined viewpoint by referencing the mapping table storage unit (42).