Method for recognizing objects in a set of images recorded by one or more cameras

摘要

Method for improving the visibly of objects and recognizing objects in a set of images recorded by one or more cameras, the images of said set of images being made from mutual different geometric positions, the method comprising the steps or recording a set or subset of images by means of one camera which is moved rather freely and which makes said images during its movement, thus providing an array of subsequent images, estimating the camera movement between subsequent image recordings, also called ego-motion hereinafter, based on features of those recorded images, registering the camera images using a synthetic aperture method, recognizing said objects.

主权项

1. A method for recognizing objects in recorded images, the images being obtained from mutually different geometric positions, the method comprising the following steps:
recording a set of images by means of a camera which is freely moved and which optically makes said images during its movement, thus providing an array of successive images; estimating the camera movement between successive image recordings based on positions of features detected in those recorded images; using a synthetic aperture method based on the camera movements to compute synthetic aperture images for focal surfaces at different depths, the synthetic aperture images being computed from the array of successive images, wherein said computing comprises: computing respective image transformations from each image of the array of successive images to each of a plurality of respective 3D surfaces, each of the 3D surfaces corresponding to a respective focal surface at a respective depth; warping each of the images to each of the 3D surfaces according to the respective image transformations; and computing each of the synthetic aperture images by fusing the images warped to a respective one of the 3D surfaces.