Hole site defect detection method based on Hough circle transformation principle

Through the hole position defect detection method based on the Hough circular transformation principle, the hole position of the semiconductor spray head is identified and supplemented and improved with high accuracy, solving the problems of low efficiency and large errors of traditional detection methods, achieving high-precision and fast hole position detection, ensuring the reliability of product quality.

CN120027701APending Publication Date: 2025-05-23HONGHU SEMICONDUCTOR EQUIPMENT (FOSHAN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510233259.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-28
Publication Date
2025-05-23

Smart Images

  • Figure CN120027701A_ABST
    Figure CN120027701A_ABST
Patent Text Reader

Abstract

The invention discloses a hole site defect detection method based on a Hough circle transformation principle, which is applied to the technical field of control of image data processing, and specifically comprises the key steps of preprocessing, identifying and marking hole sites through a Hough circle transformation algorithm, supplementing labels, repeatedly supplementing the labels until all the hole sites are marked, and the like. The method comprises the following steps: taking a hole site as a center, inputting the number of hole sites adjacent to the periphery of the hole site as a reference number, then traversing image information by utilizing a Hough circle conversion algorithm, identifying and marking a plurality of hole sites and circle center position coordinates thereof, and supplementing marks by calculating an average distance between a center point and the marked hole site and a reference deviation angle. In addition, the method further comprises a repeated mark supplementing step, the newly-marked hole site serves as a center point to execute mark supplementing again so as to ensure that all the hole sites are accurately marked, and the method has the advantages of being efficient and accurate and is suitable for defect detection of the hole sites of the semiconductor spray header.
Need to check novelty before this filing date? Find Prior Art

Claims

1. A hole defect detection method based on the Hough circle transform principle, characterized in that: The following steps are involved: A1. Preprocessing the input semiconductor shower head hole map; A2. Initialize the parameter space through the Hough circle transformation algorithm, and traverse the image information to vote and detect the peak value of the parameter space, so as to identify, obtain and mark several hole positions, and obtain the center position coordinates of each hole position; A3. Take each marked hole as the center point and perform the following supplementary marking steps: B3.

1. Calculate and obtain the average distance between any center point and each adjacent hole position marked around it; B3.

2. According to the number of adjacent holes around the center point, the relative offset angle between the center point and the adjacent holes is determined as the reference offset angle; B3.

3. Take one of the marked adjacent holes around the center point as the starting point, and perform circumferential offsets successively around the center point with the average distance value as the radius and the reference offset angle; B3.

4. Once a week, check whether there is a marked hole at the offset position. If not, draw a marked hole at the offset position. A4. Repeat steps B3.1-B3.4 until all wells have completed the above-mentioned re-marking steps, and an image of all marked wells is obtained.

2. The hole defect detection method based on the Hough circle transform principle according to claim 1 is characterized in that: In step A2, based on the equation (x−a) 2 +(y−b) 2 =r 2 , x, y are the coordinates of the edge points, a, b are the potential center coordinates, r is the radius value of the current traversal, and the edge point coordinates are obtained by presetting the parameter radius or distance. After calculating all possible radii, the corresponding values ​​stored in the parameter space are updated. The peak value in the parameter space corresponds to the center coordinates.

3. The hole defect detection method based on the Hough circle transform principle according to claim 2 is characterized in that: In step B3.1, the average distance value is determined by calculating the distance between the center coordinates of the center point and the center coordinates of each adjacent hole position around it.

4. The hole defect detection method based on the Hough circle transform principle according to claim 2 is characterized in that: In step B3.3, first, the distribution angles of each marked adjacent hole position around the center point relative to the center point are calculated, then they are summarized and sorted to establish the minimum distribution angle, and finally, the adjacent hole position corresponding to the minimum distribution angle is defined as the starting point of the circumferential offset.

5. The hole defect detection method based on the Hough circle transform principle according to claim 1 is characterized in that: The following steps are also included: defining each marked hole position drawn in step B3.4 as a new center point, and executing the supplementary marking steps B3.1-B3.4 again.

6. The hole defect detection method based on the Hough circle transform principle according to claim 1 is characterized in that: The preprocessing is to input the number of holes distributed adjacently around a hole.