A New Joint Detection Algorithm of Conveyer Belt X-Ray Imaging Using the BP Neural Networks

A new joint detection algorithm of conveyer belt with steel ropes using the BP(back propagation) neural networks model in X-ray imaging system is proposed. Detecting the state of conveyer belt is important in production process such as coal mine. And the joint detection is a primary task for further processing. Because of the neural networks´ massive parallelism and easy implementation, this paper uses the BP neural networks to implement the joint detection. The result shows that, the correct recognition rate of the joint image applying the trained neural network is up to 100%, and the detecting speed is very rapid. This new algorithm for joint detecting can greatly improve the performance of processing, especial for the on-line nondestructive detection of product line. And the result demonstrates that the algorithm is effective.