Study on casting ultrasonic signal extraction algorithm based on improved MP

Sparse signal decomposition from over-complete dictionaries are the most recent technique in the signal processing community. In this paper, this technique is utilized to cope with ultrasonic flaw detection problem. But its calculation is huge. A new improved algorithm is proposed which combine genetic algorithm with matching pursuit together. The improved algorithm is employed to search the best matching atoms. It can reduce complexity of matching pursuit. According to the characteristic of ultrasonic echo, the best Gabor atom is selected to self-adaptive match atomic ultrasonic flaw signals. Experimental results shows that four parameters of ultrasonic signal blurred by noise can be estimated according to the proposed algorithm, and the signal quality and performance parameters are improved obviously.