CN-121027305-B - Copper-clad plate copper-clad quality detection method and system

CN121027305BCN 121027305 BCN121027305 BCN 121027305BCN-121027305-B

Abstract

The invention discloses a copper-clad plate copper-clad quality detection method and system, wherein the method comprises the steps of applying ultrasonic excitation signals with preset frequency and power to a copper-clad plate to be detected, and synchronously acquiring dynamic micro-vibration response image sequences of the surface of the copper-clad plate under ultrasonic excitation by using a high-speed camera; the method comprises the steps of carrying out phase locking analysis on a dynamic micro-vibration response image sequence to construct a copper-clad plate dynamic response feature map, inputting the copper-clad plate dynamic response feature map into a trained defect identification model, outputting a preliminary defect region distribution map, carrying out multi-scale morphological analysis and region connectivity verification on the preliminary defect region distribution map to generate an accurate positioning defect region map, and generating a comprehensive evaluation report of copper-clad quality according to a preset quality grading rule based on the accurate positioning defect region map. By utilizing the embodiment of the invention, the high-precision positioning and classification of the defects of the copper-clad plate can be realized, and a high-efficiency and reliable automatic solution is provided for the quality evaluation of the copper-clad plate.

Inventors

YE ZHI
WU XINGYOU
ZOU HUA
WANG MINGDONG
SHENG CONGCONG
WU XINGMAO

Assignees

江西省宏瑞兴科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20250909

Claims (9)

1. The method for detecting the quality of the copper-clad plate copper-clad is characterized by comprising the following steps: Applying ultrasonic excitation signals with preset frequency and power to the copper-clad plate to be detected, and synchronously acquiring a dynamic micro-vibration response image sequence of the surface of the copper-clad plate under ultrasonic excitation by using a high-speed camera, wherein the ultrasonic frequency is associated with the natural frequencies of the copper-clad plate base material and the copper layer so as to excite micro-vibration of a specific mode; The dynamic micro-vibration response image sequence is subjected to phase locking analysis, amplitude-phase change characteristics of each pixel point in an ultrasonic excitation period are extracted, a copper-clad plate dynamic response characteristic diagram is constructed, wherein the phase locking analysis is based on phase reference of ultrasonic excitation signals, micro-vibration signals and environmental noise caused by excitation are separated, phase reference time stamps are extracted from the ultrasonic excitation signals, phase reference vectors are constructed, time domain alignment operation is carried out on the dynamic micro-vibration response image sequence based on the phase reference vectors, an aligned image sequence is output, each pixel point of the aligned image sequence is analyzed by adopting a wavelet packet transformation algorithm, instantaneous amplitude and phase angle of a micro-vibration dominant frequency band are extracted, amplitude-phase matrixes of pixel levels are output, the phase coherence filtering algorithm is used for taking the phase reference vectors as references, environmental noise components in the amplitude-phase matrixes are restrained, pure micro-vibration signals are separated, amplitude-phase characteristic sets after noise restraint are output, and the amplitude-phase characteristic sets after noise restraint are mapped to two-dimensional space coordinates, and the copper-clad plate dynamic response characteristic diagram is constructed; inputting the copper-clad plate dynamic response feature map into a trained defect recognition model, recognizing and positioning a potential copper-clad bonding poor region, a microcrack region and a local stress abnormal region, and outputting a preliminary defect region distribution map, wherein the defect recognition model is a deep learning model obtained by training the dynamic response feature map based on a known defect sample; Carrying out multi-scale morphological analysis and region connectivity verification on the preliminary defect region distribution map, and screening out a real defect region which accords with a physical rule by combining design line topology information of a copper-clad plate to generate a precisely positioned defect region map, wherein the region connectivity verification is used for eliminating false detection caused by image noise or a non-copper-clad region; Based on the accurate positioning defect area diagram, calculating the area occupation ratio, the distribution density and the distance between adjacent key lines of various defects, and generating a comprehensive evaluation report of copper coverage quality according to a preset quality scoring rule, wherein the comprehensive evaluation report comprises defect type statistics, position labeling and quality grade judgment.
2. The method according to claim 1, wherein the applying an ultrasonic excitation signal with a preset frequency and power to the copper-clad plate to be tested and synchronously using a high-speed camera to collect a dynamic micro-vibration response image sequence of the surface of the copper-clad plate under ultrasonic excitation, wherein the ultrasonic frequency is related to the natural frequencies of the copper-clad plate substrate and the copper layer to excite micro-vibration of a specific mode, comprises: based on the material thickness and density parameters of the copper-clad plate, calculating the natural frequency spectrum of the base material and the copper layer through a finite element modal analysis algorithm, and outputting the natural frequency spectrum as a preset frequency candidate set; Dynamically tuning the output frequency of the ultrasonic generator by using a preset frequency candidate set and adopting a self-adaptive resonance tracking algorithm, selecting a frequency point capable of exciting the maximum micro-vibration amplitude, and outputting an optimized ultrasonic excitation signal; According to the periodic characteristics of the optimized ultrasonic excitation signals, configuring a frame rate synchronization module of the high-speed camera, ensuring that the frame rate is matched with the integer multiple of the excitation frequency, and outputting a synchronization parameter set; and applying an optimized ultrasonic excitation signal to the copper-clad plate, and triggering a high-speed camera to acquire based on the synchronous parameter set to generate a dynamic micro-vibration response image sequence.
3. The method of claim 2, wherein the inputting the copper-clad plate dynamic response feature map into a trained defect recognition model, recognizing and locating potential copper-clad poor-bonding regions, microcrack regions and local stress anomaly regions, and outputting a preliminary defect region distribution map, wherein the defect recognition model is a deep learning model trained based on dynamic response feature maps of known defect samples, and comprises: carrying out multi-channel feature enhancement on the dynamic response feature map of the copper-clad plate, weighting and highlighting an abnormal response region through an attention mechanism, and outputting an enhanced feature map; Inputting the enhanced feature map into a pre-trained defect recognition model, wherein the model adopts a map convolution network architecture, and combines dynamic response features and material stress distribution priori knowledge to output an original defect probability map; processing an original defect probability map by applying a self-adaptive threshold segmentation algorithm, dynamically dividing defect category boundaries according to probability values, and outputting a binarized defect mask map; Marking a connected region on the binarized defect mask map, classifying the connected region into copper-clad poor combination, microcrack or stress abnormality, and outputting a preliminary defect region distribution map.
4. The method of claim 3, wherein the performing multi-scale morphological analysis and region connectivity verification on the preliminary defect region distribution map, combining design line topology information of the copper-clad plate, screening out a real defect region conforming to a physical rule, and generating a precisely positioned defect region map, wherein the region connectivity verification is used for eliminating false detection caused by image noise or a non-copper-clad region, and comprises: performing multi-scale morphological opening and closing operation on the preliminary defect region distribution map, eliminating isolated noise points by using variable-scale structural elements, and outputting a morphological correction map; marking a connected defect area by adopting an area growth algorithm based on the morphological correction map, and outputting a connected area label map; Loading copper-clad plate design line topology information and constructing a line topology diagram; And (3) carrying out physical constraint verification on the connected region label graph by combining the circuit topology graph, verifying whether the connected region accords with a stress propagation model, eliminating false detection of a non-copper-covered region, and outputting a physical verified accurate positioning defect graph.
5. The method of claim 4, wherein calculating area occupation ratio, distribution density, distance between adjacent critical lines and total evaluation report of copper coverage quality based on the accurate positioning defect area map, and generating total evaluation report of copper coverage quality according to preset quality scoring rule, wherein the total evaluation report comprises defect type statistics, position labeling and quality grade determination, and comprises: analyzing the accurate positioning defect area map, extracting the geometric attribute of each defect area, calculating the area occupation ratio and the distribution density, and outputting a defect measurement data set; based on the defect measurement data set and the designed line topology information, calculating the minimum Euclidean distance from the defect area to the key line by adopting a nearest neighbor algorithm, and outputting a distance risk index set; Inputting the defect measurement data set and the distance risk index set into a quality scoring engine, and generating an original quality scoring vector by weighting and fusing according to a preset rule; and generating a comprehensive evaluation report comprising a defect type statistical table, a position label graph and a quality grade judging label according to the original quality grading vector.
6. The utility model provides a copper-clad plate copper-clad quality detection system which characterized in that, the system includes: The system comprises an acquisition module, a high-speed camera and a display module, wherein the acquisition module is used for applying ultrasonic excitation signals with preset frequency and power to a copper-clad plate to be detected, and synchronously acquiring a dynamic micro-vibration response image sequence of the surface of the copper-clad plate under ultrasonic excitation by using the high-speed camera, wherein the ultrasonic frequency is related to the inherent frequencies of a copper-clad plate base material and a copper layer so as to excite micro-vibration in a specific mode; the analysis module is used for carrying out phase locking analysis on the dynamic micro-vibration response image sequence, extracting amplitude-phase change characteristics of each pixel point in an ultrasonic excitation period, constructing a copper-clad plate dynamic response characteristic diagram, wherein the phase locking analysis is based on phase reference of ultrasonic excitation signals, separates micro-vibration signals and environmental noise caused by excitation, extracts phase reference time stamps from the ultrasonic excitation signals, constructs phase reference vectors, carries out time domain alignment operation on the dynamic micro-vibration response image sequence based on the phase reference vectors, outputs an aligned image sequence, adopts a wavelet packet transformation algorithm to analyze each pixel point of the aligned image sequence, extracts instantaneous amplitude and phase angle of a micro-vibration dominant frequency band, outputs an amplitude-phase matrix of a pixel level, takes the phase reference vectors as references, suppresses environmental noise components in the amplitude-phase matrix, separates pure micro-vibration signals, outputs an amplitude-phase characteristic set after noise suppression, and maps the amplitude-phase characteristic set after noise suppression to two-dimensional space coordinates, and constructs the copper-clad plate dynamic response characteristic diagram; The recognition module is used for inputting the dynamic response characteristic diagram of the copper-clad plate into a trained defect recognition model, recognizing and positioning a potential copper-clad poor combination region, a microcrack region and a local stress abnormal region, and outputting a preliminary defect region distribution diagram, wherein the defect recognition model is a deep learning model obtained by training the dynamic response characteristic diagram based on a known defect sample; The verification module is used for carrying out multi-scale morphological analysis and area connectivity verification on the preliminary defect area distribution diagram, screening out a real defect area which accords with a physical rule by combining design line topology information of the copper-clad plate, and generating an accurate positioning defect area diagram, wherein the area connectivity verification is used for eliminating false detection caused by image noise or a non-copper-clad area; The generation module is used for calculating the area occupation ratio, the distribution density and the distance between adjacent key lines of various defects based on the accurate positioning defect area map, and generating a comprehensive evaluation report of the copper coverage quality according to a preset quality scoring rule, wherein the comprehensive evaluation report comprises defect type statistics, position labeling and quality grade judgment.
7. The system according to claim 6, wherein the acquisition module is specifically configured to: based on the material thickness and density parameters of the copper-clad plate, calculating the natural frequency spectrum of the base material and the copper layer through a finite element modal analysis algorithm, and outputting the natural frequency spectrum as a preset frequency candidate set; Dynamically tuning the output frequency of the ultrasonic generator by using a preset frequency candidate set and adopting a self-adaptive resonance tracking algorithm, selecting a frequency point capable of exciting the maximum micro-vibration amplitude, and outputting an optimized ultrasonic excitation signal; According to the periodic characteristics of the optimized ultrasonic excitation signals, configuring a frame rate synchronization module of the high-speed camera, ensuring that the frame rate is matched with the integer multiple of the excitation frequency, and outputting a synchronization parameter set; and applying an optimized ultrasonic excitation signal to the copper-clad plate, and triggering a high-speed camera to acquire based on the synchronous parameter set to generate a dynamic micro-vibration response image sequence.
8. A storage medium having a computer program stored therein, wherein the computer program is arranged to perform the method of any of claims 1-5 when run.
9. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of any of claims 1-5.

Description

Copper-clad plate copper-clad quality detection method and system Technical Field The invention belongs to the technical field of quality detection, and particularly relates to a copper-clad plate copper-clad quality detection method and system. Background The copper-clad plate is used as a core substrate of a Printed Circuit Board (PCB), and the combination quality of the copper-clad layer and the substrate directly influences the circuit performance and the reliability. Conventional inspection methods such as visual inspection, electrical performance testing, or X-ray scanning have limitations in that they are inefficient, costly, or difficult to detect micro-scale defects. Especially for hidden defects such as poor copper cladding combination, microcracks and the like, the prior art is difficult to realize high-precision and nondestructive rapid positioning. In recent years, nondestructive detection technology based on vibration response is developed, but is limited by environmental noise interference and signal analysis capability, and micro vibration characteristics under complex working conditions cannot be effectively separated. In addition, the conventional image processing algorithm has insufficient feature extraction capability on the dynamic response of the copper-clad plate, so that the false detection rate is high. Disclosure of Invention The invention aims to provide a copper-clad plate copper-clad quality detection method and system, which are used for solving the defects in the prior art, realizing high-precision positioning and classification of the defects of the copper-clad plate and providing a high-efficiency and reliable automatic solution for the quality evaluation of the copper-clad plate. The embodiment of the application provides a copper-clad plate copper-clad quality detection method, which comprises the following steps: Applying ultrasonic excitation signals with preset frequency and power to the copper-clad plate to be detected, and synchronously acquiring a dynamic micro-vibration response image sequence of the surface of the copper-clad plate under ultrasonic excitation by using a high-speed camera, wherein the ultrasonic frequency is associated with the natural frequencies of the copper-clad plate base material and the copper layer so as to excite micro-vibration of a specific mode; performing phase locking analysis on the dynamic micro-vibration response image sequence, extracting amplitude-phase change characteristics of each pixel point in an ultrasonic excitation period, and constructing a copper-clad plate dynamic response characteristic diagram, wherein the phase locking analysis is based on phase reference of ultrasonic excitation signals, and separating micro-vibration signals and environmental noise caused by excitation; inputting the copper-clad plate dynamic response feature map into a trained defect recognition model, recognizing and positioning a potential copper-clad bonding poor region, a microcrack region and a local stress abnormal region, and outputting a preliminary defect region distribution map, wherein the defect recognition model is a deep learning model obtained by training the dynamic response feature map based on a known defect sample; Carrying out multi-scale morphological analysis and region connectivity verification on the preliminary defect region distribution map, and screening out a real defect region which accords with a physical rule by combining design line topology information of a copper-clad plate to generate a precisely positioned defect region map, wherein the region connectivity verification is used for eliminating false detection caused by image noise or a non-copper-clad region; Based on the accurate positioning defect area diagram, calculating the area occupation ratio, the distribution density and the distance between adjacent key lines of various defects, and generating a comprehensive evaluation report of copper coverage quality according to a preset quality scoring rule, wherein the comprehensive evaluation report comprises defect type statistics, position labeling and quality grade judgment. Still another embodiment of the present application provides a copper-clad plate copper-clad quality detection system, the system comprising: The system comprises an acquisition module, a high-speed camera and a display module, wherein the acquisition module is used for applying ultrasonic excitation signals with preset frequency and power to a copper-clad plate to be detected, and synchronously acquiring a dynamic micro-vibration response image sequence of the surface of the copper-clad plate under ultrasonic excitation by using the high-speed camera, wherein the ultrasonic frequency is related to the inherent frequencies of a copper-clad plate base material and a copper layer so as to excite micro-vibration in a specific mode; The analysis module is used for carrying out phase locking analysis on the dynamic micro-vibration response imag