Integrated Circuit Wafer Detection Based on Intelligent Information Processing

In order to effectively detect the defects of wafer in IC production and improve its utilization rate, a wafer image detection method based on intelligent information processing is proposed. Firstly, the images with defects such as contamination, ink dot, crack, single crystal collapse and scratch are preprocessed by median filter. Secondly, mutual information is used as a measure to complete the registration of reference image and defect image. Thirdly, the generalized Hough transform is used to detect the defects based on the difference operation of the registration image. Finally, Outs algorithm is used to improve image defect extraction and achieve better segmentation effect. The experimental results show that the proposed algorithm has good image preprocessing ability for wafer image, and effective extraction, segmentation and detection effect for defect target image.