Design of a vision sensor using fuzzy associative database

In this paper a design method is proposed for a potential new camera-based intelligent vision sensor. This sensor can be used for fast multiple planar object recognition. The mechanism behind the design is a fuzzy associative database (FAD) which consists of a fuzzy database (FD) and a fuzzy search engine (FSE). The FSE uses table one to conduct search over table two, both in FD, through a bank of fuzzy associative memory matrix (BFAMM). In fact, the FSE establishes a correspondence between an object and one of the trained classes in table two. Therefore, the FAD could actually 'remember' the trained objects and the FSE could 'recognize' the incoming object by comparing it with trained information in the database. The experimental results show that this approach is robust and fast.