Recognition of object data in computer integrated manufacturing

A novel techniques has been developed for recognition of object data in computer integrated manufacturing systems (CIMS). A common database is trained which can be accessed by all users of the system. The recognition technique is based on nonlinear adaptive feedforward neural network trained with a novel algorithm, which adjusts the weights sequentially as well as extending to adjust one weight at each iteration. The simulation results carried out in this paper for object recognition show a greater accuracy and leads to the significant reduction in lead time of a product.