Circuit board diagnosis method based on infrared imaging

Abstract

The invention discloses a circuit board diagnosis method based on infrared imaging. Circuit board fault detection comprises the following steps: S11, initializing connection with hardware; s22, synchronously acquiring multi-modal data; s33, carrying out registration on different source images; s44, detecting a fault area; and S55, carrying out fault positioning and outputting. Through the infrared image technology, whether the circuit board works abnormally or not is distinguished and judged according to the difference of heat radiation when the element works, and the element which abnormally radiates and emits heat in the working state can be detected through a non-contact method.

Description

Technical Field

[0001] This invention belongs to the field of infrared detection technology and relates to a circuit board diagnostic method based on infrared imaging. Background Technology

[0002] The development of circuit board manufacturing technology is advancing rapidly. The integration of electronic components is becoming higher and higher, and the circuits are becoming more and more complex. If traditional contact testing is used after a problem occurs, a lot of time and effort will be required. Therefore, non-contact testing methods are becoming increasingly important.

[0003] Infrared thermal imaging detection technology is a non-contact detection technology that has been successfully applied in many fields, with circuit board fault detection being one of its important uses. When a circuit board is in operation, each component emits different amounts of thermal radiation. After acquiring infrared images, image processing is performed. The processed faulty infrared images are compared and analyzed with those of non-faulty circuit boards, and intelligent algorithms are used to determine the location and component of the fault. Summary of the Invention

[0004] To address the above problems, the technical solution of this invention is: a circuit board diagnostic method based on infrared imaging, wherein the general circuit board testing method includes the following steps: S10, build a system maintenance platform; S20, Set up the test process; S30, execute the test; S40, Data Analysis; S50, results storage and visualization.

[0005] Preferably, S10 includes: User permission management, including role assignment and permission granularity down to the module level; Instrument calibration and regular blackbody radiation calibration are performed.

[0006] Preferably, S20 includes: Test case design includes defining stimulus signals and sampling parameters; Configure diagnostic rules and set fault thresholds.

[0007] Preferably, S30 includes: Automated control, using SCPI commands to control the power supply and oscilloscope; Real-time monitoring and dynamic display of temperature field T(x,y,t) and gradient distribution. .

[0008] Preferably, S40 includes: Feature extraction and calculation of regional average temperature Standard deviation k is the component number; Fault classification uses an SVM classifier to distinguish between short circuits, open circuits, and aging.

[0009] Preferably, S50 includes: Database archiving stores raw data and thermal image sequences; Report generation: Generate a PDF report containing thermal images, fault location markers, and repair recommendations.

[0010] Preferably, circuit board fault detection includes the following steps: S11, Initialize hardware connections; S22, multimodal data synchronous acquisition; S33, heterogeneous image registration; S44, Fault Area Detection; S55, Fault location and output; S11 includes: Hardware self-test checks the connection status of the infrared camera, visible light camera, and environmental control box, and verifies whether the transmission rate of the communication interface reaches the preset value. Parameter configuration: Set the infrared camera sampling frequency and temperature range; Configure the environmental chamber temperature and humidity control parameters. S22 includes: Optical image acquisition triggers a visible light camera to capture a global image of the circuit board, which is then saved as a reference image. Resolution 0.244mm / pixel; Infrared thermal image sequence recording: Power-on excitation circuit board, synchronous acquisition of infrared thermal image sequences. ; Dynamic range 14-bit, noise equivalent temperature difference ≤50mK.

[0011] Preferably, S33 includes: Feature point extraction, for and Extract SIFT feature points separately; The transformation matrix is ​​solved using the RANSAC algorithm to calculate the affine transformation matrix: Reprojection error < 0.5 pixels Where u,v are the pixel coordinates of the infrared image; u',v' are the coordinates of the registered optical image; and H is the affine transformation matrix. Thermal image alignment, aligning infrared thermal image sequences Mapped to the optical image coordinate system.

[0012] Preferably, S44 includes: Thermal anomaly analysis, calculation of temperature gradient field Extract local extreme points, where T(x,y) is the temperature distribution function; Spatiotemporal features are fused to construct a spatiotemporal cube C(x,y,t), and 3D CNN is used to detect abnormal heating patterns.

[0013] Preferably, S55 includes: Coordinate transformation: Transform the thermal anomaly point (u,v) to the robot's coordinate system (X,Y,Z); Generate a report, outputting a JSON format report containing the location of the faulty component, temperature deviation, and confidence level.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention uses infrared thermal imaging to acquire images and data in a non-contact manner. Non-contact temperature measurement does not interfere with the normal production process and has the advantages of fast response time, non-contact operation, safe use and long service life.

[0015] Infrared thermal imagers can monitor the temperature distribution of the entire circuit in the early stages of circuit board development, making it easier for R&D engineers to make reasonable layouts.

[0016] Infrared thermal imaging technology is also applicable to the maintenance field. With the help of thermal imagers, maintenance engineers can often quickly locate short circuits and thermal defects on the board without circuit diagrams, so as to carry out further processing.

[0017] This invention can be widely applied to electronic devices prone to thermal defects, such as resistors, capacitors, diodes, and transistors; and to modular circuits, such as power drive circuits and data acquisition circuits. Attached Figure Description

[0018] Figure 1 This is a general detection structure diagram of a circuit board based on infrared imaging, according to a specific embodiment of the present invention. Figure 2 This is a fault detection architecture diagram of a circuit board diagnostic method based on infrared imaging, according to a specific embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] Conversely, this invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of the invention as defined in the claims. Furthermore, to provide a better understanding of the invention, certain specific details are described in detail below. However, those skilled in the art will fully understand the invention even without these detailed descriptions.

[0021] A circuit board diagnostic method based on infrared imaging, see [link to relevant documentation]. Figure 1 The general testing method for circuit boards includes the following steps: S10, build a system maintenance platform; S20, Set up the test process (TP); S30, execute the test; S40, Data Analysis; S50, results storage and visualization.

[0022] S10 includes: user access control, including role assignment (administrator / operator / visitor) and access control granularity down to the module level; Instrument calibration, and regular blackbody radiation calibration (reference temperature error ±0.5℃).

[0023] S20 includes: test case design, including defining excitation signals (such as 5V / 1kHz square wave) and sampling parameters (such as infrared frame rate 60Hz). Configure diagnostic rules and set fault thresholds (e.g., a resistance temperature rise > 10℃ is considered abnormal).

[0024] S30 includes: automated control, which controls the power supply and oscilloscope via SCPI commands; Real-time monitoring and dynamic display of temperature field T(x,y,t) and gradient distribution. .

[0025] S40 includes: feature extraction, calculation of the region's average temperature. Standard deviation k is the component number; for fault classification, an SVM classifier (kernel function: RBF, γ=0.1) is used to distinguish between short circuit, open circuit and aging.

[0026] S50 includes: database archive, storage of raw data (SQLite), and thermal image sequences (HDF5 format). Report generation: Generate a PDF report containing thermal images, fault location markers, and repair recommendations.

[0027] The system maintenance platform primarily handles system maintenance tasks such as user management, instrument resource management, network settings, database configuration, printer settings, operation logs, and database maintenance, while also storing test information in the data server. The TP development platform is mainly responsible for the development and debugging of the TP and its operating interface, and also saves the TP and operating interface to the data server for easy user management. The TP execution platform automatically retrieves the latest version of the TP from the data server, then uses VISA to drive instrument resources to complete the test execution process, saves the test data to the database, and finally generates and prints reports. The self-testing and calibration platform mainly implements the platform's self-testing and calibration functions. The data management platform provides users with a comprehensive data management platform that includes functions such as integrated query, information sharing, log management, and report management.

[0028] See Figure 2 Circuit board fault detection includes the following steps: S11, Initialize hardware connections; S22, multimodal data synchronous acquisition; S33, heterogeneous image registration; S44, Fault Area Detection; S55, Fault location and output; S11 includes: Hardware self-test checks the connection status of the infrared camera, visible light camera, and environmental control box, and verifies whether the transmission rate of the communication interface (such as GigE, USB3.0) reaches the preset value. Configure parameters, set the infrared camera sampling frequency (≥30Hz), temperature range (-20℃~150℃), and configure the environmental chamber temperature and humidity control parameters (e.g., 25℃±1℃, RH50%±5%). S22 includes: Optical image acquisition triggers a visible light camera (2448×2048 pixels) to capture a global image of the circuit board, which is then saved as a reference image. Resolution 0.244mm / pixel; Infrared thermal image sequence recording: Power-on excitation circuit board, synchronous acquisition of infrared thermal image sequences. ; Dynamic range 14-bit, noise equivalent temperature difference ≤50mK.

[0029] S33 includes: feature point extraction, for and Extract SIFT feature points separately; The transformation matrix is ​​solved using the RANSAC algorithm to calculate the affine transformation matrix: Reprojection error < 0.5 pixels Where u,v are the pixel coordinates of the infrared image, which are two-dimensional coordinates in the original infrared thermal image; u',v' are the registered optical image coordinates, which are aligned to the visible light camera coordinate system through affine transformation; H is the affine transformation matrix, including rotation, translation, and scaling parameters; Thermal image alignment, aligning infrared thermal image sequences Mapped to the optical image coordinate system.

[0030] S44 includes: thermal anomaly analysis, calculation of temperature gradient field. The Sobel operator is used to locate regions of sudden temperature changes (such as short-circuit points) and extract local extreme points. T(x,y) is the temperature distribution function. Spatiotemporal features are fused to construct a spatiotemporal cube C(x,y,t), and 3D CNN is used to detect abnormal heating patterns.

[0031] S55 includes: coordinate transformation, which transforms the thermal anomaly point (u,v) to the robot's coordinate system (X,Y,Z); Generate a report, outputting a JSON format report containing the location of the faulty component, temperature deviation, and confidence level.

[0032] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An infrared imaging-based circuit board diagnosis method, characterized by, The general detection method of the circuit board comprises the following steps: S10, constructing a system maintenance platform; S20, setting a test flow; S30, executing a test; S40, data analysis; S50, result storage and visualization.

2. The infrared imaging-based circuit board diagnosis method according to claim 1, characterized by, The S10 comprises: User permission management, including role assignment and permission granularity to module level; Instrument calibration, periodically performing blackbody radiation calibration.

3. The infrared imaging-based circuit board diagnosis method according to claim 1, characterized by, The S20 comprises: Test case design, including defining excitation signals and sampling parameters; Diagnostic rule configuration, setting fault threshold values.

4. The infrared imaging-based circuit board diagnosis method according to claim 1, characterized by, The S30 comprises: Automatic control, controlling power supply and oscilloscope through SCPI instructions; Real-time monitoring, dynamic display of temperature field T(x, y, t) and gradient distribution .

5. The infrared imaging-based circuit board diagnosis method according to claim 1, characterized by, The S40 comprises: Feature extraction, calculating area average temperature , standard deviation , k is the element number; Fault classification, using an SVM classifier to distinguish short circuit, open circuit and aging.

6. The infrared imaging-based circuit board diagnostic method of claim 1, wherein, The S50 comprises: Database archiving, storing original data and thermal image sequences; Report generation, generating a PDF report containing thermal image, fault location marking and repair suggestions.

7. The infrared imaging-based circuit board diagnostic method of claim 1, wherein, The circuit board fault detection comprises the following steps: S11, initialization and hardware connection; S22, multi-modal data synchronous acquisition; S33, heterogeneous image registration; S44, fault area detection; S55, fault positioning and output; The S11 comprises: Hardware self-test, detecting the connection state of the infrared camera, visible light camera and environmental control box, and verifying whether the transmission rate of the communication interface reaches the preset value; Parameter configuration, setting the sampling frequency and temperature range of the infrared camera, and configuring the temperature and humidity control parameters of the environmental box; The S22 comprises: Optical image acquisition, trigger the visible light camera to take a global image of the circuit board, save as a reference image , resolution 0.244 mm / pixel; Infrared thermal image sequence recording, power-on excitation circuit board, synchronous acquisition of infrared thermal image sequence ; Dynamic range 14bit, noise equivalent temperature difference ≤50mK.

8. The infrared imaging-based circuit board diagnostic method of claim 1, wherein, The S33 comprises: Feature point extraction, to and SIFT feature points are extracted respectively; Transformation matrix solution, calculating the affine transformation matrix through the RANSAC algorithm: reprojection error < 0.5 pixels, Wherein, u, v are infrared image pixel coordinates; u', v' are the coordinates of the registered optical image; H is the affine transformation matrix; Thermal image alignment, mapping of infrared thermal image sequences to optical image coordinate systems.

9. The infrared imaging-based circuit board diagnostic method of claim 1, wherein, The S44 comprises: thermal anomaly analysis, calculating temperature gradient field , extracting local extreme points, T(x,y) is a temperature distribution function; Spacetime feature fusion, constructing a spacetime cube C(x, y, t), and detecting abnormal heating patterns using a 3D CNN.

10. The infrared imaging-based circuit board diagnostic method of claim 1, wherein, The S55 comprises: Coordinate conversion, converting the thermal anomaly point (u, v) to the manipulator coordinate system (X, Y, Z); Report generation, outputting a JSON format report containing the fault element location, temperature deviation and confidence.

Images