Development of Machine Vision System Based on BP Neural Network Self-learning

The precision of machine vision calibration is affected by those factors such as the loss of depth information, distortion of camera lens, and errors caused by image processing. In this paper machine vision system is developed by means of BP neural network with self-learning. There are 4 inputs, which are the image coordinates of a match point in left and right camera, and 3 outputs in the network. The sum square of errors between the outputs of the network and actual coordinates in world system is taken as performance index. The network weights are tuned in the light of gradient descend method and can be achieved stable value while the given sum square of errors is reached. Thus each projection matrix of two cameras of machine vision system can be replaced by the weights and the activation function of the neural network, and calibration of system is finished. Finally, the precision analysis is carried out for the system.