Recognition of partially occluded object

Recognition of partially occluded objects has long been a difficult task in the area of pattern recognition and computer vision. The presence of occlusions complicates the problem of image registration and hence the subsequent tasks of object recognition. Traditionally, separate modules are used to detect occlusions prior to image registration. These are too intensive in terms of computation. In this paper, we propose a new image registration algorithm based on concurrent cross-correlation of two-dimensional multi-scale images to aid in the recognition of partially occluded objects. This algorithm is capable of detecting the presence of occlusions and localizing their positions concurrently during the image registration process. Experimental results show that the proposed algorithm is very effective in dealing with partial occlusion.<
>