Abstract
One of the most important problems in automated integrated circuit (IC) board inspection is to recognize the codes on the IC. In this paper, we propose a method that is suitable for the recognition of the stamped characters on the IC. It first uses the loss function to establish a classifier. Then, by means of feature extraction the similarity measure between the recognizing pattern and each of the 36 characters is obtained. Finally, the pattern is determined by choosing the character with the maximum similarity. Furthermore, a character matching method is proposed to increase the efficiency and effectiveness of inspection. This matching method can also reduce the false alarm rate due to poor printing quality.