This research is an attempt to design an Algorithm to recognize workpiece in order to guide a manipulator approach the workpiece and pick it to feed into a mould mounted on platform of the press. Back propagation algorithm is applied on a supervised classification of workpiece images. Several images of the workpiece are used as training samples and were converted into binary forms and to train the network. To ensure the performance of the system, network parameters such as momentum value, learning rate and number of hidden neuron would be adjusted to get the appropriate setting. It is found that the the proposed algorithm can identify the type of workpiece presented.