CN-121982039-A - State detection method, electronic device, storage medium, and program product

CN121982039ACN 121982039 ACN121982039 ACN 121982039ACN-121982039-A

Abstract

The application provides a state detection method, an electronic device, a storage medium and a program product. Relates to the industrial product detection technology. The method comprises the steps of obtaining a surface contour image of a semiconductor packaging substrate in a motion state, carrying out image preprocessing on the surface contour image to obtain a corrected contour image, carrying out image preprocessing on the corrected contour image to correct image distortion caused by motion state acquisition, carrying out feature extraction on the corrected contour image to generate a plurality of feature vectors corresponding to the semiconductor packaging substrate, wherein the plurality of feature vectors comprise edge feature vectors, shape feature vectors and color feature vectors, and carrying out comparison processing on the plurality of feature vectors and preset state templates corresponding to preset process states to obtain a target process state corresponding to the semiconductor packaging substrate. The method of the application can improve the reliability of the chip packaging state classification.

Inventors

Cui Yuantai
LU JUNRUI
WANG YURAN
CHEN CHONG
Wen Jiongdun
XUAN JINGQUAN

Assignees

上海升翕光电科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260409

Claims (10)

1. A method for detecting a state, comprising: acquiring a surface contour image of a semiconductor packaging substrate in a motion state; Performing image preprocessing on the surface profile image to obtain a corrected profile image, wherein the image preprocessing is used for correcting image distortion caused by the motion state acquisition; extracting features of the corrected contour image to generate a plurality of feature vectors corresponding to the semiconductor packaging substrate, wherein the plurality of feature vectors comprise edge feature vectors, shape feature vectors and color feature vectors; And comparing the feature vectors with preset state templates corresponding to the preset process states to obtain the target process state corresponding to the semiconductor packaging substrate.
2. The method of claim 1, wherein performing feature extraction on the corrected profile image to generate a plurality of feature vectors corresponding to the semiconductor package substrate, comprises: Performing edge detection on the corrected contour image to obtain an edge feature vector of the contour of the semiconductor packaging substrate; And determining a color feature vector and a shape feature vector corresponding to the corrected contour image based on the edge feature vector.
3. The method of claim 2, wherein determining the color feature vector and the shape feature vector corresponding to the corrected contour image based on the edge feature vector comprises: Determining a target contour region corresponding to the corrected contour image based on the edge feature vector; Performing color analysis on the target contour area to obtain the color feature vector; And carrying out Hough transformation detection on the target contour region to obtain a shape feature vector corresponding to the semiconductor packaging substrate.
4. The method of claim 1, wherein acquiring a surface profile image of the semiconductor package substrate in a moving state comprises: driving the semiconductor package substrate or the image sensor to rotate along a preset path; And acquiring a surface profile image of the semiconductor packaging substrate in a motion state through the image sensor.
5. The method of claim 4, wherein capturing, by the image sensor, a surface profile image of the semiconductor package substrate in motion comprises: Acquiring the real-time rotation speed of the semiconductor packaging substrate or the image sensor and the real-time environment light intensity of the image acquisition area; Adjusting the exposure time and the frame rate of the image sensor according to the real-time rotation speed and the real-time environment light intensity; And acquiring a surface profile image of the semiconductor packaging substrate based on the adjusted exposure time and frame rate.
6. The method of claim 1, wherein comparing the plurality of feature vectors with a preset state template corresponding to each preset process state to obtain a target process state corresponding to the semiconductor package substrate, comprises: determining state matching degrees of the feature vectors and the corresponding preset state templates; and determining the preset state template with the state matching degree larger than the corresponding matching degree threshold value as a target state template, and determining the process state corresponding to the target state template as a target process state.
7. The method of claim 6, wherein for any one of the preset state templates, the preset state template includes feature threshold vectors corresponding to respective feature vectors, determining a state matching degree of the plurality of feature vectors corresponding to the preset state template includes: Determining the sub-matching degree of each feature vector and the corresponding feature threshold vector; And determining the state matching degree corresponding to the preset state template according to the sub-matching degree of each feature vector and the corresponding feature threshold vector.
8. An electronic device comprising at least one processor and a memory, the memory storing computer-executable instructions, the at least one processor executing the computer-executable instructions stored in the memory to implement the method of any one of claims 1-7.
9. A computer readable storage medium having stored therein computer executable instructions for implementing the method of any of claims 1-7 when the computer executable instructions are executed by a processor.
10. A computer program product comprising a computer program which, when executed by a processor, implements the method of any of claims 1-7.

Description

State detection method, electronic device, storage medium, and program product Technical Field The present application relates to the field of industrial product detection technologies, and in particular, to a state detection method, an electronic device, a storage medium, and a program product. Background In the semiconductor manufacturing process, electromagnetic interference (Electromagnetic Interference, EMI) may deposit a metal thin film on a substrate surface of a chip Package (PKG) by a sputtering (dispenser) process to block propagation of electromagnetic waves. After the chip is loaded on the annular frame film, the EMI shielding film is finally formed through plasma treatment (PLASMA TREATMENT) and sputtering process in sequence. In the prior art, the semiconductor manufacturing process can be monitored by a photoelectric sensor mounted on a path from a clamp (Cassette) to a ring frame film, and whether the chip completes the sputtering process can be detected by adjusting the sensitivity of the photoelectric sensor. However, the photoelectric sensor can only judge the normal state and the bad state, and the bad products with the broken process and the finished products with the process cannot be distinguished, so that the risk of misjudgment is high, and the reliability of the classification of the chip packaging state is lower. Disclosure of Invention The application provides a state detection method, electronic equipment, a storage medium and a program product, which are used for solving the problem of low reliability of chip packaging state classification in the prior art. In a first aspect, the present application provides a state detection method, including: acquiring a surface contour image of a semiconductor packaging substrate in a motion state; Performing image preprocessing on the surface profile image to obtain a corrected profile image, wherein the image preprocessing is used for correcting image distortion caused by the motion state acquisition; extracting features of the corrected contour image to generate a plurality of feature vectors corresponding to the semiconductor packaging substrate, wherein the plurality of feature vectors comprise edge feature vectors, shape feature vectors and color feature vectors; And comparing the feature vectors with preset state templates corresponding to the preset process states to obtain the target process state corresponding to the semiconductor packaging substrate. Optionally, extracting features of the corrected contour image to generate a plurality of feature vectors corresponding to the semiconductor package substrate, including: Performing edge detection on the corrected contour image to obtain an edge feature vector of the contour of the semiconductor packaging substrate; And determining a color feature vector and a shape feature vector corresponding to the corrected contour image based on the edge feature vector. Optionally, determining, based on the edge feature vector, a color feature vector and a shape feature vector corresponding to the corrected contour image includes: Determining a target contour region corresponding to the corrected contour image based on the edge feature vector; Performing color analysis on the target contour area to obtain the color feature vector; And carrying out Hough transformation detection on the target contour region to obtain a shape feature vector corresponding to the semiconductor packaging substrate. Optionally, acquiring a surface profile image of the semiconductor package substrate in a moving state includes: driving the semiconductor package substrate or the image sensor to rotate along a preset path; And acquiring a surface profile image of the semiconductor packaging substrate in a motion state through the image sensor. Optionally, acquiring, by the image sensor, a surface profile image of the semiconductor package substrate in a motion state, including: Acquiring the real-time rotation speed of the semiconductor packaging substrate or the image sensor and the real-time environment light intensity of the image acquisition area; Adjusting the exposure time and the frame rate of the image sensor according to the real-time rotation speed and the real-time environment light intensity; and acquiring a surface profile image of the semiconductor packaging substrate based on the adjusted exposure time and frame rate. Optionally, comparing the plurality of feature vectors with a preset state template corresponding to each preset process state to obtain a target process state corresponding to the semiconductor package substrate, including: determining state matching degrees of the feature vectors and the corresponding preset state templates; and determining the preset state template with the state matching degree larger than the corresponding matching degree threshold value as a target state template, and determining the process state corresponding to the target state template as a target process state. Opti