IC wafer surface defect detection method

Abstract

The invention belongs to the technical field of the wafer detection, and especially relates to an IC wafer surface defect detection method for a product with a repeated pattern array structure. The method comprises the following steps: a, acquiring a standard IC wafer image through a white light interferometer, performing image pre-processing on the standard IC wafer image, and storing the same ina standard image sample base, and then acquiring an original image of a to-be-detected IC wafer surface single structure unit through the white light interferometer, namely, the to-be-detected image,and performing the image pre-processing; b, performing image registration on the standard image of the standard image sample base and the to-be-detected image after the image pre-processing; c, orderly performing gray difference computation, image binarization processing and morphological method processing on the standard image and the registered to-be-detected image to obtain a defect image; andd, performing defect type identification and defect location identification on the defect image. Through the detection method disclosed by the invention, the IC wafer surface defect detection is realized under the condition that the wafer surface has the repeated pattern array, and the method has the features of being simple in algorithm and fast and effective in detection.

Description

Technical field [0001] The invention belongs to the field of wafer detection technology, and in particular relates to a method for detecting surface defects of IC (integrated circuit, integrated circuit) wafers for products with repeating pattern array structures. Background technique [0002] In the semiconductor industry, defect detection on wafer surfaces generally requires high efficiency and accuracy, and the ability to capture defects quickly and effectively. For the detection of wafer surface defects, microscope visual inspection was mainly used at first. With the development of technology, the semiconductor process has developed from 90nm to 14nm. Visual inspection of pattern defects can no longer meet the current demand for accuracy. There are more and more data characterizing wafer surface defects, and inspections rely on manual labor. It has become increasingly beyond its capabilities. [0003] Currently, the most widely used wafer surface defect detection metho...

AI Technical Summary

Problems solved by technology

The amount of image data collected by this method is large, the processing steps are cumbersome, and the hardware requirements of the computer are relatively high.

Images