3-D Object Recognition Based On Structured Illumination

The well-known VanderLugt matched filter, appeared in 1964, indicated Optical Pattern Recognition (OPR) technology came into being. With 40 years of research and development, OPR has become an important branch of information optics. With the rapid development of computer technology and spatial light modulation technology, image recognition technology, realized by optical methods, has gradually changed from the traditional system composed of pure optical components to the system that combined optics with computer hardware and software. It has been widely applied. Traditional optical pattern recognition is based on 2D image correlation, which has difficulties in 3D object recognition. 3D object recognition based on structured light projection is proposed in this paper. The characteristic of the method is to construct a new recognizable plural function by structured illumination, in which the height distributing information of the object is encoded as the phase of the plural function. So the new method has the characteristic of intrinsic 3-D recognition. Two methods of 3D object recognition are proposed in this paper. One is based on structured light projection and JTC, another is based on structured light projection and MF. Principles, calculation expressions, recognition system frames and experiment results are proposed. Original work in this paper include:1. In method based on structured light projection and JTC: Sine fringes are projected to the surface of reference object and target object respectively. The fringes are deformed due to the height of the objects. Auto correlation and cross correlation is obtained after Joint Transform Correlation (JTC) of two deformed image. The characteristic of the method is to construct a new recognizable pluralfunction by structured illumination, in which the height distributing information of the object is encoded as the phase of the plural function. So the new method has the characteristic of intrinsic 3-D recognition. The method is proved to have the possibility used in 3-D object recognition through computer simulation.2. In method based on structured light projection and MF: Firstly, by structured light illumination, the input images of reference object in all directions are acquired by CCD camera, which contained the 3-D information by means of the distorted fringe encoding; Secondly, the frequency expression database of target object was come into being after fast Fourier transforming and filtering the images. Finally, by structured light illumination, the input image of target object is acquired by CCD camera, the frequency expression of the target image is computed and matched with that in the database, by which different objects can be distinguished from magnitude of the correlation peaks. This method, proved to be effective by the results of both simulated and on-the-spot experiments, can be used not only for 3-D object recognition but also for gaining the approximate value of rotating angle.