Circuit board defect detection method, system and device

Abstract

The application discloses a circuit board defect detection method, system and device, and belongs to the technical field of computers. The method comprises the following steps: a circuit board defect detection device acquires a target wavelength pair matched with characteristic parameters (at least including a material type) of a to-be-detected circuit board, and sends a first control signal carrying the target wavelength pair (including a first wavelength and a second wavelength) to a wavelength tuning module; under the action of the first control signal, the wavelength tuning module adjusts its working parameters to first filtering parameters and second filtering parameters corresponding to the first wavelength and the second wavelength, respectively; an imaging module sequentially collects fluorescence intensity images of fluorescence signals generated by the to-be-detected circuit board under excitation under different filtering parameters; the fluorescence intensity images are subjected to differential processing to obtain differential images; and visual detection is performed on the differential images to obtain a defect detection result. The filtering wavelength is matched with the characteristic parameters of the to-be-detected circuit board, and the accuracy of the defect detection result of the circuit board is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, system and device for detecting defects in circuit boards. Background Technology

[0002] In the manufacturing process of printed circuit boards (PCBs) and multilayer boards, the defect detection results of the PCBs are crucial to the accuracy of the PCB yield. Related technologies typically employ visible light reflection imaging, which uses a charge-coupled device (CCD) to acquire reflected images of the circuitry and relies on image algorithms to enhance image contrast.

[0003] However, the highly reflective surface of the metals used in circuit fabrication (such as copper) reduces the contrast between the circuitry and the substrate. When there are inconsistencies in metal thickness, metal oxidation, or foreign matter on the circuit board, the reflected image becomes blurry and unclear, resulting in lower accuracy of defect detection results. Summary of the Invention

[0004] This application provides a method, system, and device for detecting defects in printed circuit boards (PCBs), which improves the accuracy of defect detection results. The technical solution is as follows: A first aspect provides a method for detecting defects in a circuit board, the method being applied to a circuit board defect detection device, the method comprising: acquiring a target wavelength pair matching characteristic parameters of the circuit board under test, the characteristic parameters including at least a material type, the target wavelength pair including a first wavelength and a second wavelength; sending a first control signal to a wavelength tuning module, the first control signal carrying the target wavelength pair; causing the wavelength tuning module, under the action of the first control signal, to sequentially adjust its operating parameters to a first filter parameter corresponding to the first wavelength and a second filter parameter corresponding to the second wavelength; receiving a first fluorescence intensity image and a second fluorescence intensity image sequentially acquired by an imaging module; wherein the first fluorescence intensity image is a fluorescence intensity image of the fluorescence signal generated by the circuit board under test under excitation at the first filter parameter, and the second fluorescence intensity image is a fluorescence intensity image of the fluorescence signal at the second filter parameter; performing differential processing on the first fluorescence intensity image and the second fluorescence intensity image to obtain a differential image; and performing visual inspection on the differential image to obtain a defect detection result of the circuit board under test.

[0005] Secondly, a circuit board defect detection system is provided, the system comprising a circuit board defect detection device, a wavelength tuning module, and an imaging module; the circuit board defect detection device is used to acquire a target wavelength pair matching the characteristic parameters of the circuit board under test, the characteristic parameters including at least material type, and the target wavelength pair including a first wavelength and a second wavelength; send a first control signal to the wavelength tuning module, the first control signal carrying the target wavelength pair; the wavelength tuning module is used to, under the action of the first control signal, sequentially adjust the operating parameters to a first filter parameter corresponding to the first wavelength and a second filter parameter corresponding to the second wavelength; the imaging module is used to sequentially acquire a first fluorescence intensity image of the fluorescence signal generated by the circuit board under test under the first filter parameter and a second fluorescence intensity image under the second filter parameter, and transmit them to the circuit board defect detection device; the circuit board defect detection device is further used to perform differential processing on the first fluorescence intensity image and the second fluorescence intensity image to obtain a differential image; and perform visual inspection on the differential image to obtain the defect detection result of the circuit board under test.

[0006] Thirdly, a circuit board defect detection device is provided, the circuit board defect detection device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the method described in the first aspect.

[0007] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0008] Fifthly, a computer program product containing instructions is provided that, when run on a computer, causes the computer to perform the method described in the first aspect.

[0009] This application provides a method, system, and device for detecting defects in printed circuit boards (PCBs). The method is applied to a PCB defect detection device. According to the solution provided in this application, a target wavelength pair matching the characteristic parameters of the PCB under test is obtained. The characteristic parameters include at least the material type, and the target wavelength pair includes a first wavelength and a second wavelength. A first control signal carrying the target wavelength pair is sent to a wavelength tuning module. The wavelength tuning module is a tunable filter, which, under the action of the first control signal, sequentially adjusts its operating parameters to a first filter parameter corresponding to the first wavelength and a second filter parameter corresponding to the second wavelength. An imaging module sequentially acquires a first fluorescence intensity image of the fluorescence signal generated by the PCB under test under the first filter parameter and a second fluorescence intensity image under the second filter parameter. After the fluorescence signal generated by the PCB under test passes through the wavelength tuning module, the imaging module completes fluorescence imaging. The target wavelength pair used for filtering can adapt to changes in material type and, through an adaptive adjustment process, matches the characteristic parameters of the PCB under test, thus obtaining first and second fluorescence intensity images with significant differences. The first and second fluorescence intensity images are differentially processed to obtain a differential image. In the differential image, the fluorescence signal in the substrate region or circuit region is greatly suppressed, enhancing the contrast between the substrate region (RoB) and the circuit region (RoD). The differential image is then subjected to visual inspection to obtain the defect detection results of the circuit board under test. The filter wavelength used in this technical solution is adapted to the characteristic parameters of the circuit board under test, which can improve the accuracy of the defect detection results. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of fluorescence spectral information provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a fluorescence imaging system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 1 ; Figure 4 This is a schematic diagram of the structure of a handwheel filter provided in an embodiment of this application; Figure 5This is a schematic diagram of the structure of a differential image provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 2 ; Figure 7 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 3 ; Figure 8 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 4 ; Figure 9 This is a schematic diagram of the structure of an excitation light source provided in an embodiment of this application; Figure 10 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 5 ; Figure 11 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 6 ; Figure 12 This is a flowchart of a circuit board defect detection method provided in an embodiment of this application; Figure 13 This is a flowchart of another circuit board defect detection method provided in the embodiments of this application; Figure 14 This is a flowchart of another circuit board defect detection method provided in the embodiments of this application; Figure 15 This is a schematic diagram of the structure of a circuit board defect detection device provided in an embodiment of this application. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0013] It should be understood that "multiple" as mentioned in this application refers to two or more. In the description of this application, unless otherwise stated, " / " indicates "or," for example, A / B can mean A or B; "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist, for example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, to facilitate a clear description of the technical solutions of this application, the terms "first," "second," etc., are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first," "second," etc., do not limit the quantity or execution order, and that "first," "second," etc., do not necessarily imply differences.

[0014] Before providing a detailed explanation of the embodiments of this application, the relevant technologies of the embodiments of this application will be described first.

[0015] In related technologies, ultraviolet-excited fluorescence mechanisms are introduced to overcome the physical limitations of optical reflection on metal surfaces. This is achieved by utilizing the principle that the fluorescence effects of metals and their compounds, as well as substrates, differ at specific wavelengths. Figure 1 As shown, Figure 1 This is a schematic diagram of fluorescence spectral information provided in an embodiment of this application. Figure 1 The horizontal axis represents wavelength, measured in nanometers (nm), and the vertical axis represents light intensity, which can be measured in relative fluorescence intensity. Figure 1 The image shows the wavelength curve of the incident light and the fluorescence spectrum curve produced when the material is excited. Figure 1 It can be seen that when a substance is irradiated by incident light of a specific wavelength, it absorbs energy and enters an excited state. Subsequently, it is rapidly de-excited and emits visible light with a longer wavelength. This process is called fluorescence.

[0016] Based on the above fluorescence phenomena, a fluorescence imaging system is proposed, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of a fluorescence imaging system provided in an embodiment of this application. The fluorescence imaging system includes a fixed-wavelength excitation light source (or illumination light source), an objective lens, a dichroic mirror, a fixed-wavelength filter, a lens barrel, a photosensitive element, and a circuit board defect detection device. The objective lens, dichroic mirror, fixed-wavelength filter, lens barrel, and photosensitive element are arranged sequentially. The fixed-wavelength excitation light source illuminates the circuit board under test placed on the stage according to its fixed wavelength, exciting the circuit board to generate a fluorescence signal. The fluorescence signal is transmitted sequentially along the path constructed by the objective lens, dichroic mirror, fixed-wavelength filter, and lens barrel, and then the photosensitive element completes the fluorescence imaging. The objective lens magnifies the circuit board under test; the dichroic mirror filters out the light signal that excites the circuit board to generate a fluorescence signal, and transmits the fluorescence signal generated by the excited circuit board; the fixed-wavelength filter selects light of a specific wavelength from the fluorescence signal for output; the lens barrel focuses the fluorescence signal output by the fixed-wavelength filter onto the photosensitive element to achieve fluorescence imaging. The circuit board defect inspection equipment identifies the fluorescence intensity images captured by the photosensitive element to obtain the defect detection results of the circuit board. Utilizing this fluorescence phenomenon detection technology significantly improves the signal-to-noise ratio of circuit board defect detection imaging and enhances the imaging contrast of micron-level circuit board defects, meeting the requirements of online inspection in high-precision circuit board manufacturing.

[0017] In this context, the circuit board mentioned can specifically include PCBs, integrated circuit (IC) packaging substrates, or other circuit boards used to achieve chip interconnection. It should be noted that circuit boards are used to carry electronic components (chips, resistors, capacitors, inductors, diodes, transistors, amplifiers, etc.) and realize electrical connections between different electronic components. Specifically, circuit boards in this context include ordinary boards, multilayer boards, high-density interconnects (HDI), flexible printed circuit boards (FPCs), and IC carrier boards and their semi-finished products made of organic substrates, ceramic substrates, or glass. In some feasible embodiments, the circuit board can be formed by stacking electrically conductive layers and electrically insulating layers in parallel and mechanically pressing and / or hot-pressing them. The electrically insulating layer is used to isolate the electrically conductive layers and provide local interconnections between them. The electrically insulating layer can specifically be made of materials such as glass fiber and resin. The electrically conductive layer is used to transmit current and signals and can specifically be made of materials with good conductivity, such as metals (especially copper) or carbon (especially graphene).

[0018] However, the above Figure 2 The fluorescence-based excitation technology shown in the diagram still has the following technical problems: 1. The fluorescence-based detection scheme adopts a fixed-wavelength excitation and fixed-wavelength filtering detection strategy. It excites the circuit board under test (PCB) with a fixed-wavelength light source and uses the fluorescence spectrum information of the PCB after excitation at a fixed wavelength to determine the defect detection result. However, it does not consider the differences in fluorescence spectra between the PCB substrate, circuits, compounds, and residues. This fixed-wavelength detection strategy is difficult to adapt to these differences, resulting in poor imaging effects. 2. The fluorescence-based detection scheme does not consider the fluorescence spectrum information of PCBs with different material types, different production batches, or different processes. It cannot change the detection strategy in real time according to different material characteristics, resulting in low accuracy in PCB defect detection. It lacks intelligent optimization capabilities when facing production batch fluctuations, process fluctuations, and new material types, leading to poor industrial adaptability. 3. When conducting online detection of multiple PCBs under test, spot plotting or line scanning is required to obtain the fluorescence spectrum information of the PCBs under test, resulting in low online detection efficiency.

[0019] For circuit boards made of new materials, in new production batches, or using new processes, the fluorescence spectral characteristics are unknown. Figure 2 The fluorescence-based excitation technique shown in the figure uses a fixed wavelength for detection, which leads to lower accuracy of the detection results.

[0020] Based on the aforementioned technical problem of low accuracy in the detection results, this application provides a circuit board defect detection system. For example... Figure 3 , Figures 6-8 , Figures 10-11 As shown, Figure 3 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 1 The system includes a circuit board defect detection device 100, a wavelength tuning module 200, and an imaging module 300. The circuit board under test is fixed on a stage and excited by a laser light source (e.g., visible light, ultraviolet light, infrared light, white point light, red-green-blue ring light, etc.) to generate a fluorescence signal. This fluorescence signal passes through the wavelength tuning module 200 and then enters the imaging module 300. The circuit board defect detection device 100 is used to acquire a target wavelength pair that matches the characteristic parameters of the circuit board under test. The characteristic parameters include at least the material type, and the target wavelength pair includes a first wavelength and a second wavelength. A first control signal is sent to the wavelength tuning module 200, carrying the target wavelength pair. The wavelength tuning module 200 is a tunable filter that, under the action of the first control signal, can sequentially adjust its operating parameters to a first filter parameter corresponding to the first wavelength and a second filter parameter corresponding to the second wavelength.

[0021] In this embodiment, the wavelength tuning module 200 can be an acousto-optic tunable filter (AOTF) or a liquid crystal tunable filter (LCTF). Among them, AOTF has high reliability in the field of fluorescence microscopy and spectroscopy imaging, and can be applied to complex industrial environments such as vibration and temperature changes, thus increasing the applicable scenarios of the circuit board defect detection system.

[0022] The wavelength tuning module 200 is located in the spectral imaging optical path and receives the fluorescence signal emitted by the circuit board under test. By receiving the first control signal, it can accurately and rapidly scan the transmission center wavelength from λ_min (e.g., 400 nm) to λ_max (e.g., 750 nm) within milliseconds, achieving rapid capture of the fluorescence spectrum. This eliminates the need for manual pre-setting or frequent filter switching to capture fluorescence, improving the flexibility of the fluorescence imaging method and increasing detection efficiency.

[0023] The wavelength tuning module 200 can also be a handwheel filter, which enables multi-wavelength imaging, such as... Figure 4 As shown, Figure 4 This is a schematic diagram of the structure of a handwheel filter provided in an embodiment of this application. Figure 4 The handwheel filter shown can switch between nine filter wavelengths.

[0024] The wavelength tuning module in this embodiment can be an AOTF, LCTF, or a handwheel filter, which improves the diversity of spectral imaging optical paths.

[0025] The imaging module 300 is located after the wavelength tuning module 200. It is used to sequentially acquire the first fluorescence intensity image under the first filter parameter and the second fluorescence intensity image under the second filter parameter of the fluorescence signal generated by the circuit board under test, and transmit them to the circuit board defect detection device 100.

[0026] The fluorescence signal generated by the circuit board under test after being excited passes through the wavelength tuning module 200 and is then imaged by the imaging module 300. The target wavelength used for filtering can adapt to changes in material type and is matched with the characteristic parameters of the circuit board under test through an adaptive adjustment process. This allows for the acquisition of a first fluorescence intensity image and a second fluorescence intensity image with significant differences.

[0027] The circuit board defect detection equipment 100 is also used to perform differential processing on the first fluorescence intensity image and the second fluorescence intensity image to obtain a differential image; the fluorescence signal in the substrate area or circuit area in the differential image is greatly suppressed, enhancing the contrast between the substrate area and the circuit area. Figure 5 As shown, Figure 5 This is a schematic diagram of the structure of a differential image provided in an embodiment of this application. Due to the significant difference in fluorescence intensity between the circuit area and the substrate area, the final differential image has high grayscale contrast, making the micron-level circuit board exhibit a distinct black and white contrast feature. Furthermore, the circuit board defect detection device 100 is also used to perform visual inspection on the differential image. The visual inspection includes, but is not limited to, detection of open and short circuits, detection of contaminant defects, detection of substrate cracking defects, and other defect types, to obtain the defect detection results of the circuit board under test. The defect detection results include, but are not limited to, the identification and labeling of defect types and locations.

[0028] In this embodiment, the characteristic parameters may include parameters that affect the defect detection results, such as material type, production batch, and process. Thus, the filter wavelength used in this technical solution is adapted to the characteristic parameters of the circuit board under test, which can improve the accuracy of the defect detection results.

[0029] The circuit board defect detection system provided in this application embodiment deeply integrates tunable filtering technology (corresponding to wavelength tuning module 200), high-speed imaging technology (corresponding to imaging module 300) and intelligent image processing algorithm (corresponding to circuit board defect detection device 100) to form a brand-new system-level solution. At the same time, the flexibility of the algorithm enables adaptability that exceeds that of a fixed hardware structure.

[0030] In some embodiments, the above-described high-speed, high-contrast online defect detection process for the circuit board under test is a detection mode. Before performing defect detection on the circuit board, a target wavelength pair matching the characteristic parameters of the circuit board under test can be obtained in the following manner. This process can be called spectral calibration mode, which will be described below. Under the condition of meeting preset conditions (e.g., receiving a user instruction, or every preset detection cycle, or detecting that the defect detection result does not meet the requirements), the circuit board defect detection device 100 sends a second control signal to the tunable filter; under the action of the second control signal, the tunable filter adjusts its operating parameters sequentially to the filter parameters corresponding to each wavelength within a first preset wavelength range (e.g., λ_min to λ_max) according to a first preset step wavelength (e.g., 5nm, 10nm, etc.); the imaging module 300 sequentially acquires fluorescence intensity image samples of fluorescence signal samples under each filter parameter, and the filter parameters, wavelengths, and fluorescence intensity image samples have a one-to-one correspondence; wherein, the fluorescence signal sample is the fluorescence signal generated by the circuit board under test or a circuit board sample with characteristic parameters consistent with the circuit board under test when excited. In other words, before testing the circuit board under test, a spectral calibration mode can be executed first to obtain the target wavelength pair, and then the testing mode can be executed. It is understood that these two modes can be executed separately and are not forcibly linked. That is, it is not always necessary to perform the spectral calibration mode to obtain the target wavelength pair before each execution of the testing mode; and it is not always necessary to execute the testing mode after obtaining the target wavelength pair through the spectral calibration mode. For example, after obtaining the target wavelength pair under preset conditions, the corresponding data can be stored in the defect detection equipment and then directly retrieved when the testing mode is executed subsequently.

[0031] The circuit board defect detection equipment 100 performs differential processing on the fluorescence intensity image samples corresponding to any two wavelengths to obtain multiple differential image samples; each differential image sample corresponds to two wavelengths, and the image contrast of each differential image sample is determined; the two wavelengths corresponding to the differential image sample with the strongest image contrast are taken as the target wavelength pair.

[0032] In this embodiment, the spectral calibration mode is relatively short-time. After a single spectral calibration obtains a specific target wavelength pair of the circuit board based on specific characteristic parameters, the specific target wavelength pair can be repeatedly called until the preset conditions trigger the calibration mode again. The detection mode is related to the switching speed of the filter wavelength and the scanning speed of the fluorescence spectrum. Therefore, this circuit board defect detection scheme can obtain the accuracy of the scheme related to the spectral curve analysis without affecting the detection speed.

[0033] The circuit board defect detection system provided in this application embodiment implements an adaptive optimization fluorescence imaging detection scheme based on fluorescence spectral feature distribution (i.e., fluorescence spectral information). It employs a high-switching-speed wavelength tuning module to rapidly scan the fluorescence signal generated by the circuit board under test. Through a built-in algorithm, it analyzes image sequences at different wavelengths in real time, performs differential processing on the fluorescence intensity images at every two wavelengths, analyzes the image contrast of the resulting differential images, and automatically calculates the optimal detection wavelength pair (λ_optimal1, λ_optimal2) that maximizes the difference in contrast between the circuit area and the substrate area. This parameter is then applied for high-speed, high-contrast online detection, which greatly suppresses the fluorescence signal in the substrate area or circuit area in the differential image during detection mode, enhancing the contrast between the substrate area and the circuit area, thereby improving the accuracy of the circuit board defect detection results.

[0034] The circuit board defect detection system provided in this application has a "self-optimization" capability, which can cope with changes in material type, production batch and process. By adaptively changing, it can always work in the best state, thereby improving the signal-to-noise ratio and difference contrast of defect detection results, and ensuring the consistency of circuit board detection modes for different batches, different material types and different processes.

[0035] In some embodiments, based on Figure 3 ,like Figure 6 As shown, Figure 6 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 2 The system also includes a fluorescence interception module 400, which includes a dichroic mirror 401 and / or a filter 402. Figure 6 The fluorescence interception module 400, including a dichroic mirror 401 and a filter 402, is illustrated as an example. In practical applications, the fluorescence interception module 400 can also be a dichroic mirror 401 or a filter 402. The filter 402 can be a cutoff filter, a bandpass filter, or a polarizer. Compared to a dichroic mirror, a filter has a higher precision filtering wavelength. Based on this, the filter 402 can be placed after the dichroic mirror 401.

[0036] The dichroic mirror 401 is used to filter out the light signal that excites the circuit board under test to generate a fluorescent signal, and to allow the fluorescent signal generated by the circuit board under test to pass through. In other words, the dichroic mirror 401 can reflect or block (i.e. filter out) the light signal of the excitation source and only allow the fluorescent signal to pass through.

[0037] Filter 402 is used to filter out optical signals in a third preset wavelength range and transmit optical signals in a fourth preset wavelength range. The third preset wavelength range and the fourth preset wavelength range do not overlap. The fourth preset wavelength range is determined based on the wavelength range of the fluorescence signal generated by the circuit board under test when excited. The fourth preset wavelength range is the cutoff wavelength of filter 402, and the third preset wavelength range and the fourth preset wavelength range can be complementary.

[0038] For example, the wavelength range of the fluorescence signal is approximately 400 nanometers (nm) to 750 nm. Based on this, the lower limit of the fourth preset wavelength range can be set to 400 nm, the upper limit can be set to 750 nm, and the third preset wavelength range can be set to below 400 nm and above 750 nm.

[0039] In this embodiment, a fluorescence interception module 400 is placed before the wavelength tuning module 200 to ensure that optical noise outside the detection band (corresponding to the fourth preset wavelength range) is effectively suppressed. The fluorescence signal is transmitted sequentially along the optical path constructed by the fluorescence interception module 400 and the wavelength tuning module 200, and then fluorescence imaging is completed by the imaging module 300. By setting the fluorescence interception module 400, the accuracy of the fluorescence intensity image is improved, thereby improving the accuracy of the circuit board defect detection results.

[0040] In some embodiments, based on Figure 6 ,like Figure 7 As shown, Figure 7 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 3 The system also includes a linear polarization component 500, which polarizes the fluorescence signal after passing through the wavelength tuning module 200 to obtain a polarized fluorescence signal. The linear polarization component 500 can be a linear polarizer, capable of suppressing specular reflection components in a specific polarization direction, such as suppressing the fluorescence signal in the substrate area of ​​the circuit board under test, or suppressing the fluorescence signal in the circuit area of ​​the circuit board under test. Figure 4 The above refers to the result after suppressing the fluorescence signal in the circuit region. If the linear polarization component 500 suppresses the fluorescence signal in the substrate region, then the above... Figure 4 The black and white regions will flip, meaning black areas become white and white areas become black. In other words, the polarized fluorescence signal indicates the suppressed fluorescence signal in the substrate area of ​​the circuit board under test, or the suppressed fluorescence signal in the circuit area of ​​the circuit board under test. The imaging module 300 is also used to sequentially acquire a first fluorescence intensity image corresponding to the polarized fluorescence signal, and a second fluorescence intensity image corresponding to the polarized fluorescence signal.

[0041] In this embodiment, the linear polarization component is placed before the imaging module. Two-stage optical processing is achieved through the fluorescence extraction module and the linear polarization component. This dual-stage fluorescence extraction design ensures that optical noise outside the detection band (i.e., the band of the fluorescence signal generated by the circuit board under test when excited) is effectively suppressed. By setting the linear polarization component, the image contrast of the fluorescence intensity image is improved, thereby enhancing the accuracy of the circuit board defect detection results.

[0042] In this embodiment, the excitation light source used to excite the circuit board under test to generate a fluorescent signal can be visible light, ultraviolet light, infrared light, white spot light, red-green-blue ring light, etc., and this embodiment does not limit the application in this regard. If visible light excitation is used, then no additional excitation light source is required, as described above. Figure 3 , Figure 5 and Figure 6 As shown.

[0043] In some embodiments, based on Figure 7 ,like Figure 8 As shown, Figure 8 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 4 The system also includes at least one excitation light source 600. Figure 8 Three excitation light sources are shown. The angle between the axis of the excitation light source 600 and the horizontal plane where the circuit board under test is located is within a preset range. At least one excitation light source 600 is arranged in a ring around a rotation axis, which is a vertical line perpendicular to the horizontal plane.

[0044] For example, the preset range can be set to 45°±15° (i.e., 30° to 60°), so that the ring-shaped excitation light source, as an illumination module, can improve the uniformity of illumination.

[0045] The excitation light source 600 is used to illuminate the circuit board under test to excite it to generate a fluorescence signal. It can adapt to the light transmission characteristics of different material types, providing stable and reliable illumination conditions for the imaging module and improving the clarity of the fluorescence intensity image.

[0046] In some embodiments, at least one excitation light source 600 includes a first type of excitation light source and a second type of excitation light source; the included angle corresponding to the first type of excitation light source is located in a first preset range, and the included angle corresponding to the second type of excitation light source is located in a second preset range; the first type of excitation light source and the second type of excitation light source are arranged alternately around the rotation axis.

[0047] The preset interval includes a first preset interval and a second preset interval, and the first preset interval and the second preset interval do not overlap. For example, the preset interval can be set to 45°±15°, wherein the first preset interval can be set to 45°±5° (i.e., 40° to 50°), and the second preset interval can be set to 30° to 40° or 50° to 60°.

[0048] By selecting the first type of excitation source as the main illumination angle and combining it with the second type of excitation source to achieve a supplementary illumination angle of 30°-60°, and employing a ring-symmetrical or mixed arrangement, the uniformity of illumination and light energy utilization are improved. Optimizing the angle arrangement of the excitation sources ensures that the irradiance received by all points on the surface of the circuit board under test is consistent, with edge attenuation <15%, thereby reducing local under-curing or over-exposure, minimizing shadows and curing defects, improving curing quality, providing stable and reliable illumination conditions for the imaging module, and enhancing the clarity of fluorescence intensity images.

[0049] In some embodiments, the excitation light source 600 includes an illumination light source and a homogenizing lens, based on Figure 8 ,like Figure 9 As shown, Figure 9 This is a schematic diagram of the structure of an excitation light source provided in an embodiment of this application. Figure 9 The excitation light source 600 (circular) shown is fixed to the plate (e.g., Figure 8 (as shown in the square plate).

[0050] Figure 9 Figure A shows a front view of the excitation source 600, in which three ultraviolet lamps serve as the illumination source, and a homogenizing lens surrounds the illumination source. Figure 9 Figure B in the diagram shows a side view of the homogenizing lens. The homogenizing lens is fitted onto the illumination source to uniformly emit the light signal emitted by the illumination source, providing stable and reliable illumination conditions for the imaging module and improving the clarity of the fluorescence intensity image.

[0051] In some embodiments, the imaging module 300 may be a charge-coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS). Figure 8 ,like Figure 10 As shown, Figure 10 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 5The imaging module 300 includes a lens barrel 301 (which can be understood as being composed of multiple lenses) and a photosensitive element 302 (e.g., a photosensitive chip). The lens barrel 301 is located between the wavelength tuning module 200 and the photosensitive element 302. The lens barrel 301 is used to focus the fluorescence signal after passing through the wavelength tuning module 200 onto the photosensitive element 302 so that the imaging module 300 can acquire a fluorescence intensity image.

[0052] In some embodiments, such as Figure 10 As shown, the imaging module 300 also includes an objective lens 303, which is located between the circuit board under test and the fluorescence interception module 400, or between the fluorescence interception module 400 and the lens barrel 301. Figure 10 The example shown uses an objective lens 303 positioned between the circuit board under test and the fluorescence capture module 400. The objective lens 303 is used to magnify the circuit board under test, improving the resolution and clarity of the fluorescence intensity image.

[0053] based on Figure 3 , Figures 6-10 ,like Figure 11 As shown, Figure 11 This is a schematic diagram of the structure of a circuit board defect detection system provided in an embodiment of this application. Figure 6 , Figure 11 The layout diagram of each device in the system is shown, including the fluorescence extraction module ( Figure 11 The optical path for spectral imaging consists of a dichroic mirror 401 (shown in the image), a wavelength tuning module 200, and a linear polarization component 500. It should be noted that due to the positional relationship between the dichroic mirror 401 and the wavelength tuning module 200... Figure 11 A reflector was also added.

[0054] The system's workflow is as follows: The circuit board under test is fixed on the stage, and the excitation light source illuminates the surface of the circuit board at a predetermined angle. The fluorescence signal generated by the excited circuit board is collected by the objective lens 303 and enters the imaging module through the spectral imaging optical path. The system has a compact structure. The dichroic mirror 401 and the linear polarization component 500 sequentially filter out interference signals. In the imaging module, the lens barrel 301 focuses the purified fluorescence signal onto the photosensitive element 302, and the circuit board defect detection equipment completes the visual inspection.

[0055] In this embodiment, the objective lens 303 and the lens barrel 301 in the imaging module are separated, the adjustment gap is increased, and other devices (such as a dichroic mirror 401, a filter, a wavelength tuning module 200, etc.) are filled in between. In this way, an imaging detection system based on fluorescence spectroscopy can be constructed to realize the circuit board defect detection process.

[0056] It should be noted that the wavelength tuning module, linear polarization component, and imaging module constitute a spectral adaptive imaging circuit used to complete the fluorescence imaging process. The circuit board defect detection equipment can be integrated with the spectral adaptive imaging circuit to improve circuit integration and reduce circuit area. The circuit board defect detection equipment can also include an adaptive optimization processor and a controller. The controller can be integrated with the spectral adaptive imaging circuit. The controller is used to control the wavelength tuning module to switch between different filter wavelengths and to control the laser source to switch between different illumination wavelengths. The adaptive optimization processor embeds an optimization algorithm to perform image processing and visual inspection. In this way, separating the control function from the memory- and resource-intensive processing functions can reduce the power consumption of the spectral adaptive imaging circuit.

[0057] In this embodiment, fluorescence capture imaging technology is used instead of reflection imaging technology. The spectral imaging optical path serves as a filtering component. By setting the positions of the fluorescence interception module, wavelength tuning module, and linear polarization component, a cascaded filtering effect can be achieved, improving the accuracy and contrast of the fluorescence intensity image. Area array imaging is performed through the imaging module, and the wavelength tuning module can acquire the fluorescence intensity image of the entire field of view at a specific wavelength with just one switch of the filter parameters. This is faster, and the final output is an image rather than a spectral curve, directly serving visual inspection. The circuit board defect detection equipment controls the wavelength tuning module to achieve time-series measurements through control signals. Multiple measurements and visual inspections are achieved using software algorithms, providing adaptive capabilities. This allows for accurate and efficient identification of circuit board defects, reducing the false negative rate.

[0058] Based on the circuit board defect detection device 100 provided in any of the above embodiments, this application provides a circuit board defect detection method, such as... Figure 12 As shown, Figure 12 This is a flowchart of a circuit board defect detection method provided in an embodiment of this application. The circuit board defect detection method includes: S101. Obtain a target wavelength pair that matches the characteristic parameters of the circuit board under test. The characteristic parameters include at least the material type, and the target wavelength pair includes a first wavelength and a second wavelength.

[0059] In this embodiment of the application, the circuit board defect detection device calls the target wavelength pair (λ_optimal1, λ_optimal2) obtained under the above-mentioned spectral calibration mode.

[0060] S102. Send a first control signal to the wavelength tuning module, the first control signal carrying the target wavelength pair; so that the wavelength tuning module, under the action of the first control signal, sequentially adjusts the operating parameters to the first filter parameter corresponding to the first wavelength and the second filter parameter corresponding to the second wavelength.

[0061] S103, the receiving imaging module sequentially acquires a first fluorescence intensity image and a second fluorescence intensity image; wherein, the first fluorescence intensity image is the fluorescence intensity image of the fluorescence signal generated by the circuit board under test under the first filter parameter, and the second fluorescence intensity image is the fluorescence intensity image of the fluorescence signal under the second filter parameter.

[0062] In this embodiment, the wavelength tuning module is a tunable filter that can quickly switch between λ_optimal1 and λ_optimal2 within milliseconds. The imaging module simultaneously acquires two fluorescence intensity images, namely, the first fluorescence intensity image I_1 and the second fluorescence intensity image I_2, to achieve high-speed dual-wavelength imaging.

[0063] S104. Perform differential processing on the first fluorescence intensity image and the second fluorescence intensity image to obtain a differential image.

[0064] The circuit board defect detection equipment performs real-time differential calculations on two fluorescence intensity images to obtain a differential image, Diff_Image = I_2 - k * I_1, where k is a preset calibration coefficient. The fluorescence signal in the substrate or circuit region of the differential image is greatly suppressed, resulting in a significant enhancement of the contrast between the substrate and circuit regions. Since the target wavelength pair is obtained by adaptively optimizing multiple wavelengths using feature parameters, the differential image is the image with the highest contrast and can be used for subsequent identification, extraction, and judgment processes of the circuit board under test.

[0065] S105. Perform visual inspection on the differential image to obtain the defect detection results of the circuit board under test.

[0066] In this embodiment, visual inspection includes, but is not limited to, detection of defects such as open and short circuits, contaminant defects, and substrate cracking defects, to obtain the defect detection results of the circuit board under test. The defect detection results include, but are not limited to, the identification and labeling of defect types and locations.

[0067] The circuit board defect detection method provided in this application uses a spectral imaging optical path to perform fluorescence spectral scanning, and uses the acquired fluorescence intensity image for differential processing. This changes the target of spectral analysis from identifying substances to enhancing image contrast (or difference contrast), and can directly serve the circuit board identification and annotation tasks in industrial visual inspection to obtain defect detection results.

[0068] According to the scheme provided in this application, a target wavelength pair matching the characteristic parameters of the circuit board under test is obtained. The characteristic parameters include at least the material type, and the target wavelength pair includes a first wavelength and a second wavelength. A first control signal carrying the target wavelength pair is sent to the wavelength tuning module. The wavelength tuning module is a tunable filter, which, under the action of the first control signal, can sequentially adjust its operating parameters to a first filter parameter corresponding to the first wavelength and a second filter parameter corresponding to the second wavelength. The imaging module sequentially acquires a first fluorescence intensity image of the fluorescence signal generated by the circuit board under test under the first filter parameter and a second fluorescence intensity image under the second filter parameter. After the fluorescence signal generated by the circuit board under test is excited, it is imaged by the imaging module. The target wavelength pair used for filtering can adapt to changes in material type and, through an adaptive adjustment process, matches the characteristic parameters of the circuit board under test, thus obtaining a first fluorescence intensity image and a second fluorescence intensity image with significant differences. The first fluorescence intensity image and the second fluorescence intensity image are differentially processed to obtain a differential image. The fluorescence signal in the substrate region or circuit region in the differential image is greatly suppressed, enhancing the contrast between the substrate region and the circuit region. The differential image is then subjected to visual inspection to obtain the defect detection results of the circuit board under test. The filter wavelength used in this technical solution is adapted to the characteristic parameters of the circuit board under test, which can improve the accuracy of the defect detection results.

[0069] In some embodiments, S104 can be implemented as follows: A difference image is determined based on the difference between the pixel value corresponding to the first fluorescence intensity image and the calibration pixel value corresponding to the second fluorescence intensity image; the calibration pixel value is the product of the pixel value corresponding to the second fluorescence intensity image and a preset calibration coefficient k.

[0070] The imaging module acquires a first fluorescence intensity image I_1 and a second fluorescence intensity image I_2 at two specific wavelengths λ_optimal1 and λ_optimal2. I_1 and I_2 cover the same region, the same group of pixels, or the same location within the entire field of view, thus encompassing a nearly uniform field of view (FOV) and pixel coordinates. Based on this, the pixel values ​​of the pixels in I_1 are subtracted from the calibration pixel values ​​of the pixels in I_2 to obtain the difference pixel values ​​for each pixel, thus yielding the difference image.

[0071] In the difference image Diff_Image = I_1 - k * I_2, the gray value of a pixel depends on the relative strength of the fluorescence response at wavelengths λ_1 and λ_2. The optimization algorithm iterates through all wavelength pairs (λ_1, λ_2), with the core objective of finding the specific wavelength combination that minimizes the gray values ​​of pixels in the line area (RoD) (displayed as dark) and maximizes the gray values ​​of pixels in the substrate area (RoB) (displayed as bright). In other words, the algorithm aims to find a set of parameters that maximizes the inter-class variance between (RoD) and (RoB) in the difference image. That is, it makes the result infinitely close to making the line area (RoD) appear darker and the substrate area (RoB) appear brighter, achieving maximum contrast.

[0072] In this embodiment, the differential image is obtained by calculating I_1(x, y) - k * I_2(x, y) for each corresponding pixel of the full field of view image. The result is a new two-dimensional image covering the entire field of view (FOV). In this image, the contrast between the lines and the background has been significantly enhanced. This global image-oriented differential processing essentially utilizes the differences in the spectral responses of different materials at different wavelengths to reconstruct the image. Its output serves subsequent visual detection algorithms, thereby greatly improving the accuracy of defect detection results.

[0073] For example, if the target wavelength pair, i.e. the optimal detection wavelength pair (λ_optimal1, λ_optimal2), is obtained, the difference value of point P is 200 - 10 = 190, and the difference value of point Q is 15 - 205 = -190.

[0074] In the original image, the grayscale difference between point P and point Q is 200 - 15 = 185, which we assume represents the original contrast. In the difference image, the grayscale difference between the two points expands to 190 - (-190) = 380. Thus, through difference processing, the signal difference between the circuit and the substrate area is significantly amplified. In subsequent algorithm recognition, the difference results containing both positive and negative values ​​are linearly mapped; for example, the minimum value -190 is mapped to grayscale 0 (pure black), and the maximum value 190 is mapped to grayscale 255 (pure white). This mapping significantly improves the accuracy of the algorithm's recognition.

[0075] In this embodiment, at a corresponding pixel point, position, or region, the pixel value of I_1 is subtracted from the pixel value of k * I_2 (i.e., the calibration pixel value) to obtain a difference image. The difference image is a new two-dimensional image covering the entire field of view (FOV) and can be used for visual inspection. Differential processing is a preprocessing operation for the global image (including I_1 and I_2), which reconstructs the image by utilizing the difference in fluorescence intensity at different wavelengths. By utilizing image algorithms, the accuracy of defect detection results is improved.

[0076] In some embodiments, the preset calibration coefficient can be determined by: acquiring first standard spectral information of a fluorescent standard sheet at a first wavelength and second standard spectral information at a second wavelength; and determining the preset calibration coefficient based on the first and second standard spectral information.

[0077] In this embodiment, factors such as the different transmittance of the filter in the wavelength tuning module at different wavelengths and the different responsivity of the imaging module at different wavelengths will bring about inherent differences. Therefore, a compensation coefficient (preset calibration coefficient k) needs to be used in the differential processing to compensate for the inherent differences in fluorescence intensity at different wavelengths.

[0078] The preset calibration coefficient k can be obtained in the following way: the circuit board defect detection equipment obtains the standard spectral information of the known fluorescent standard sheet under λ_optimal1 and λ_optimal2 respectively, and calculates the preset calibration coefficient k based on the standard spectral information under λ_optimal1 and λ_optimal2.

[0079] By using known fluorescent standard sheets to calibrate during the initialization of the circuit board defect detection system, a preset calibration coefficient k is determined to compensate for the difference between I_1 and I_2 during differential processing, ensuring the accuracy of the differential image and enhancing the image contrast of the differential image.

[0080] In some embodiments, the above-described S101 can be implemented in the following ways, such as Figure 13 As shown, Figure 13 This is a flowchart of another circuit board defect detection method provided in the embodiments of this application.

[0081] S1011. Obtain multiple fluorescence intensity image samples related to the feature parameters; one fluorescence intensity image sample corresponds to one wavelength.

[0082] The imaging module is located after the wavelength tuning module and is used to acquire a series of fluorescence intensity image samples I (λ_i) at different transmission center wavelengths λ_i, where i is a positive integer.

[0083] The above S1011 multiple fluorescence intensity image samples can be obtained through the following examples.

[0084] Example 1: Under preset conditions, the circuit board defect detection device sends a second control signal to the wavelength tuning module. Under the action of the second control signal, the wavelength tuning module sequentially adjusts its operating parameters to the filter parameters corresponding to each wavelength within a first preset wavelength range (e.g., λ_min to λ_max) according to a first preset step wavelength (e.g., 5nm, 10nm). The imaging module sequentially acquires multiple fluorescence intensity image samples and sends these samples to the circuit board defect detection device. These multiple fluorescence intensity image samples are fluorescence intensity images of fluorescence signal samples under various filter parameters. The fluorescence signal samples are fluorescence signals generated by the excitation of fluorescence signals or circuit board samples with characteristic parameters.

[0085] Taking λ_min as 400nm, λ_max as 750nm, and the first preset step wavelength as 5nm as an example, under the action of the second control signal, the wavelength tuning module scans its transmission center wavelength (also known as the detection wavelength) from 400nm to 750nm in a step of 5nm, while the imaging module sequentially acquires fluorescence intensity image samples at each transmission center wavelength.

[0086] In this example, there is a one-to-one correspondence between the fluorescence intensity image sample, the filter parameters, and the transmission center wavelength of the wavelength tuning module. By controlling the wavelength tuning module to scan its transmission center wavelength according to the first preset step wavelength, the imaging module synchronously exposes at each transmission center wavelength λ_i to acquire a fluorescence intensity image sample I(λ_i). Within seconds, a three-dimensional fluorescence spectral image cube containing dozens of wavelength channels (i.e., multiple fluorescence intensity image samples) can be obtained. Through rapid fluorescence spectral scanning imaging, the transmission center wavelength is optimized to determine the target wavelength pair.

[0087] Example 2: The number of laser sources can be one or more, as described above. Figure 8 As shown. Under preset conditions, the circuit board defect detection equipment sends a third control signal to each tunable wavelength laser source. Under the action of the third control signal, each tunable wavelength laser source sequentially irradiates the circuit board under test or a circuit board sample with characteristic parameters within a second preset wavelength range (e.g., 300nm to 500nm) according to a second preset step wavelength (e.g., 5nm, 10nm, etc.), thereby exciting the circuit board under test or the circuit board sample to generate fluorescence signals at each irradiation wavelength (also called excitation wavelength). The imaging module sequentially acquires multiple fluorescence intensity image samples and sends these multiple fluorescence intensity image samples to the circuit board defect detection equipment. These multiple fluorescence intensity image samples are fluorescence intensity images of fluorescence signal samples at each irradiation wavelength, and the fluorescence signal samples are fluorescence signals generated by the excitation of the circuit board sample.

[0088] Taking a second preset wavelength range of 300nm to 500nm and a second preset step wavelength of 10nm as an example, under the action of the third control signal, the laser light sources of each adjustable wavelength are adjusted synchronously, and their irradiation wavelengths are adjusted from 300nm to 500nm in a step of 10nm. At the same time, the imaging module sequentially acquires fluorescence intensity image samples under each irradiation wavelength.

[0089] In this example, there is a one-to-one correspondence between the fluorescence intensity image samples and the illumination wavelength of the excitation source. By adjusting the illumination wavelength of the laser source output, the imaging module (which can be equipped with a long-pass filter) is simultaneously exposed at each illumination wavelength to acquire fluorescence intensity image samples of the circuit board under test or the circuit board sample. By analyzing the intensity changes of the fluorescence intensity images under different illumination wavelengths, the excitation spectrum of the circuit board under test or the circuit board sample can be obtained, thereby achieving wavelength optimization and determining the target wavelength pair.

[0090] It should be noted that the PCB defect detection equipment communicates with the imaging module to receive fluorescence intensity images acquired by the imaging module; it also communicates with the wavelength tuning module to control the switching of different filter wavelengths; and it communicates with the tunable wavelength laser source to control the switching of different irradiation wavelengths. The PCB defect detection equipment is also used to perform differential processing and visual inspection. Since the PCB defect detection equipment has both control and image processing functions, it can also be called an image processing and control device.

[0091] Example 3: Combining Examples 1 and 2 above, under preset conditions, the circuit board defect detection equipment sends a second control signal to the wavelength tuning module and a third control signal to each tunable wavelength laser source. Under the action of the second control signal, the wavelength tuning module sequentially adjusts its operating parameters to the filter parameters corresponding to each wavelength within a first preset wavelength range according to a first preset step wavelength. Under the action of the third control signal, each tunable wavelength laser source sequentially irradiates the circuit board under test or a circuit board sample with characteristic parameters within a second preset wavelength range according to a second preset step wavelength, thereby exciting the circuit board under test or the circuit board sample to generate fluorescence signals at each irradiation wavelength. The imaging module sequentially acquires multiple fluorescence intensity image samples and sends these multiple fluorescence intensity image samples to the circuit board defect detection equipment.

[0092] In this example, the scanning of the transmission center wavelength and the adjustment of the illumination wavelength are achieved by controlling a single variable. For example, the filter parameters of the fixed wavelength tuning module are fixed and the laser light sources of each adjustable wavelength are adjusted to multiple illumination wavelengths in sequence. Alternatively, the illumination wavelengths of each adjustable wavelength laser light source are fixed and the wavelength tuning module is adjusted to the filter parameters in sequence (i.e., the transmission center wavelength is scanned, thereby achieving bidirectional optimization of the laser light source and filter parameters).

[0093] In this example, the fluorescence intensity image sample, filter parameters, transmission center wavelength of the wavelength tuning module, and illumination wavelength of the excitation source have a one-to-one correspondence.

[0094] In this embodiment, the irradiation wavelength can be fixed, and the transmission center wavelength can be scanned to obtain fluorescence spectral information (e.g., multiple fluorescence intensity image samples obtained in Example 1); alternatively, the transmission center wavelength can be fixed, and the irradiation wavelength can be scanned to obtain excitation spectral information (e.g., multiple fluorescence intensity image samples obtained in Example 2); or a two-dimensional scan can be performed to obtain an excitation-emission spectrum, thus obtaining the most complete fluorescence spectral information (e.g., multiple fluorescence intensity image samples obtained in Example 3). Then, differential processing is performed on the multiple fluorescence intensity image samples to obtain target wavelength pairs that match the characteristic parameters of the circuit board under test, thereby improving the accuracy of the defect detection results of the circuit board.

[0095] The preset conditions in the above three examples are any of the following: receiving a user instruction indicating that a circuit board defect detection should be performed on a circuit board with new feature parameters; the interval between the previous detection cycle is greater than a preset interval; and the contrast difference between the circuit area and the substrate area in the differential image is less than a preset threshold.

[0096] In this embodiment of the application, when a user instruction is received, it indicates that circuit board defect detection is required for circuit boards with new characteristic parameters. This triggers the spectral calibration mode, and the circuit defect detection device performs the step of acquiring target wavelength pairs to improve the accuracy of defect detection results for circuit boards with new characteristic parameters.

[0097] In this embodiment, a periodic automatic triggering method can be adopted. When the interval between the current time and the previous detection cycle is detected to be greater than a preset interval, that is, the spectral calibration mode is triggered every preset detection cycle, and the line defect detection device performs the step of acquiring the target wavelength pair. The periodic automatic triggering mode improves the accuracy of defect detection results. The preset interval or preset detection cycle can be appropriately set by those skilled in the art according to the actual situation, and this embodiment does not limit this.

[0098] In this embodiment of the application, in the detection mode, when the difference contrast (e.g., defect signal-to-noise ratio) between the line area and the substrate area in the differential image is less than a preset threshold, the spectral calibration mode is triggered. The line defect detection device performs the step of acquiring the target wavelength pair, realizing adaptive adjustment of the target wavelength pair throughout the entire life cycle, so as to improve the accuracy of the defect detection results.

[0099] S1012. Perform differential processing on the fluorescence intensity image samples corresponding to any two wavelengths to obtain multiple differential image samples.

[0100] The system automatically iterates through all possible wavelength pairs (λ_m, λ_n) and calculates the differential image sample I(λ_m) - k * I(λ_n) for each wavelength pair. Taking any two fluorescence intensity image samples I(λ_m) and I(λ_n) as examples, I(λ_m) and I(λ_n) are complete fluorescence images acquired by the imaging module at two specific wavelengths λ_m and λ_n. I(λ_m) and I(λ_n) cover the same region, group of pixels, or location within the entire field of view, encompassing a nearly uniform field of view (FOV) and pixel coordinates. Based on this, the pixel value of I(λ_m) at the corresponding pixel, location, or region is subtracted from the pixel value of k * I(λ_n) (i.e., the calibration pixel value) to obtain the differential image sample. This differential image sample is a new two-dimensional image covering the entire field of view (FOV) and can be used for visual inspection.

[0101] S1013. Take the two wavelengths corresponding to the difference image sample with the strongest image contrast as the target wavelength pair.

[0102] In this embodiment of the application, after obtaining multiple differential image samples, the difference contrast between the line area sample and the substrate area sample in each differential image sample is determined, the difference contrast is used as the image contrast, the differential image sample with the strongest image contrast is selected, and its corresponding two wavelengths are used as the target wavelength pair.

[0103] In this embodiment, under spectral calibration mode, within an extremely short exposure time, based on rapid spectral scanning technology, the true fluorescence spectral information (i.e., multiple fluorescence intensity samples) of the circuit board under test or a circuit board sample with the same characteristic parameters is acquired through pre-experimentation. Based on the optimization method of signal-to-noise ratio or inter-class variance, intelligent image processing optimization calculation oriented towards image contrast (or difference contrast) is used to obtain the difference image with the largest signal-to-noise ratio or inter-class variance, thereby determining the optimal detection wavelength pair (i.e., target wavelength pair) that can significantly improve image contrast.

[0104] In some embodiments, the differential image sample with the strongest image contrast in S1013 above refers to the differential image sample corresponding to the maximum value of the difference contrast (e.g., inter-class variance or signal-to-noise ratio), which can be calculated in the following way: Perform region identification on each differential image sample to obtain the line region sample and substrate region sample of each differential image sample; determine the difference contrast between the line region sample and the substrate region sample in each differential image sample.

[0105] A pre-trained neural network model (or image algorithm) is used to perform region identification (also known as bounding or calibration) on the differential image samples. The pixel value sets of all pixels belonging to the line region sample (RoD) and the pixel value sets of all pixels belonging to the substrate region sample (RoB) are extracted. The inter-class variance or signal-to-noise ratio between the line region sample (RoD) and the substrate region sample (RoB) is calculated, and the inter-class variance or signal-to-noise ratio is used as the difference contrast of the differential image sample.

[0106] Taking difference contrast as an example of inter-class variance, when the internal pixel values ​​of the two classes are as consistent as possible (i.e., small intra-class variance) and the difference in average pixel values ​​between the two classes is as large as possible (i.e., large inter-class variance), it indicates that the two classes of objects have the highest distinguishability. Based on this, inter-class variance = (average pixel value of RoD - average pixel value of RoB)^2 * (number of pixels in RoD * number of pixels in RoB) / total number of pixels^2. The wavelength pair that produces the maximum inter-class variance (i.e., the best difference contrast) is selected as the target wavelength pair (λ_optimal1, λ_optimal2).

[0107] The inter-class variance or signal-to-noise ratio (SNR) of the circuit area (RoD) and substrate area (RoB) samples is calculated to evaluate the contrast difference at each wavelength pair. The wavelength pair with the largest inter-class variance or SNR is selected as the target wavelength pair (λ_optimal1, λ_optimal2). Thus, using this target wavelength pair to acquire differential images in detection mode enhances the contrast between the substrate area (RoB) and the circuit area (RoD) in the differential image, thereby improving the accuracy of circuit board defect detection results.

[0108] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario. For example... Figure 14 As shown, Figure 14 This is a flowchart of another circuit board defect detection method provided in the embodiments of this application.

[0109] S11, Working Mode.

[0110] When the working mode is spectral calibration mode, execute S12; when the working mode is detection mode, execute S18.

[0111] S12, Spectral scanning imaging.

[0112] Under preset conditions, the circuit board defect detection equipment sends a second control signal to the wavelength tuning module. Under the influence of the second control signal, the wavelength tuning module adjusts its operating parameters within a first preset wavelength range to the filter parameters corresponding to the transmission center wavelength, according to a first preset step wavelength. The imaging module acquires fluorescence intensity image samples corresponding to the transmission center wavelength.

[0113] Alternatively, under preset conditions, the circuit board defect detection equipment sends a third control signal to each tunable wavelength laser source. Under the influence of the third control signal, each tunable wavelength laser source irradiates the circuit board under test or a circuit board sample with characteristic parameters within a second preset wavelength range according to a second preset step wavelength, thereby exciting the circuit board under test or the circuit board sample to generate fluorescence signals at the irradiation wavelength. The imaging module acquires fluorescence intensity image samples corresponding to the irradiation wavelength.

[0114] S13, Determine if wavelength scanning is complete.

[0115] Determine whether the wavelength tuning module has completed scanning within the first preset wavelength range according to the first preset step wavelength, or whether the tunable wavelength laser source has completed scanning within the second preset wavelength range according to the second preset step wavelength. If yes, obtain multiple fluorescence intensity image samples, with each fluorescence intensity image sample corresponding to one wavelength (transmission center wavelength or illumination wavelength), and then execute S14; otherwise, continue executing S12 until the wavelength scanning is completed.

[0116] S14. Perform differential processing on any two fluorescence intensity image samples.

[0117] Differential processing is performed on the fluorescence intensity image samples corresponding to any two wavelengths to obtain multiple differential image samples.

[0118] S15, Region Identification.

[0119] Region identification is performed on each differential image sample to obtain the line region sample and the substrate region sample of each differential image sample.

[0120] S16, Wavelength pair optimization calculation.

[0121] Determine the contrast difference between the line region samples and the substrate region samples in each differential image sample. Select the two wavelengths corresponding to the differential image sample with the strongest contrast difference as the target wavelength pair.

[0122] S17. Save the optimal parameters (i.e., the target wavelength pair).

[0123] After S17, execute S23.

[0124] S18. Obtain the optimal parameters (i.e., the target wavelength pair).

[0125] S19, high-speed dual-wavelength imaging.

[0126] A first control signal is sent to the wavelength tuning module, carrying the target wavelength pair (i.e., λ_optimal1 and λ_optimal2). Under the action of the first control signal, the wavelength tuning module rapidly switches to the first filter parameter corresponding to λ_optimal1 and the second filter parameter corresponding to λ_optimal2 within milliseconds. The imaging module simultaneously acquires the fluorescence signal generated by the circuit board under test under the first filter parameter as well as the second fluorescence intensity image under the second filter parameter.

[0127] S20, Real-time Differential Processing.

[0128] The first fluorescence intensity image and the second fluorescence intensity image are differentially processed to obtain a differential image.

[0129] S21, Visual Inspection.

[0130] Visual inspection is performed on the differential image to obtain the defect detection results of the circuit board under test.

[0131] S22, Output the defect detection results.

[0132] S23, Waiting for detection instructions.

[0133] When a test instruction is received, the circuit board defect detection steps for the circuit board to be tested are executed.

[0134] In related technologies, detection strategies using fluorescence technology rely on a fixed-wavelength excitation source and a fixed-wavelength filter (see above). Figure 2 (As described above), this leads to technical problems such as poor adaptability, insufficient flexibility, and difficulty in achieving efficient online detection. When there are fluctuations in material type, production batch, or process, fixed-wavelength detection schemes struggle to maintain the optimal signal-to-noise ratio.

[0135] The circuit board defect detection method provided in this application can quickly acquire the corresponding fluorescence spectral feature distribution (i.e., fluorescence spectral information) in complex and variable circuit board material systems. Based on the acquired fluorescence spectral feature distribution, the optimal detection wavelength pair (λ_optimal1, λ_optimal2) with the maximum difference in contrast between the circuit area and the substrate area is quickly determined, i.e., the target wavelength pair used for differential processing. This circuit board defect detection system has autonomous learning capabilities, enabling "self-learning and adaptive" intelligent detection, improving adaptability to production batch differences, different material types, and different processes. It can break through the traditional fixed wavelength filter architecture, construct a tunable and adaptive optical response mechanism, and intelligently switch the optimal detection wavelength pair for different material types, different production batches, and different processes, thereby improving the accuracy of circuit board defect detection results.

[0136] The circuit board defect detection method and the circuit board defect detection system provided in the above embodiments belong to the same concept. The specific implementation process of the above method steps and the resulting technical effects can be found in the system embodiments section, and will not be repeated here.

[0137] The circuit board defect detection method provided in the above embodiments. Figure 15 This is a schematic diagram of the structure of a circuit board defect detection device provided in an embodiment of this application, as shown below. Figure 15 As shown, the circuit board defect detection device 100 includes: a processor 1001, a memory 1002, and a computer program 1003 stored in the memory 1002 and executable on the processor 1001. When the processor 1001 executes the computer program 1003, it implements the steps in the circuit board defect detection method in the above embodiments.

[0138] The circuit board defect detection device 100 can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the circuit board defect detection device 100 can be a desktop computer, a portable computer, a network server, a handheld computer, a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application embodiment does not limit the type of circuit board defect detection device 100. Those skilled in the art will understand that... Figure 15 This is merely an example of a circuit board defect detection device 100 and does not constitute a limitation on the circuit board defect detection device 100. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0139] Processor 1001 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0140] In some embodiments, memory 1002 can be an internal storage unit of the circuit board defect detection device 100, such as a hard disk or memory of the circuit board defect detection device 100. In other embodiments, memory 1002 can also be an external storage device of the circuit board defect detection device 100, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the circuit board defect detection device 100. Furthermore, memory 1002 can include both internal storage units and external storage devices of the circuit board defect detection device 100. Memory 1002 is used to store operating systems, applications, bootloaders, data, and other programs. Memory 1002 can also be used to temporarily store data that has been output or will be output.

[0141] This application also provides a circuit board defect detection device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. When the processor executes the computer program, it implements the steps in any of the above method embodiments.

[0142] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the various method embodiments described above.

[0143] This application provides a computer program product that, when run on a computer, causes the computer to perform the steps described in the various method embodiments above.

[0144] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above method embodiments of this application can be implemented by a computer program instructing related hardware. This computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate form. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, ROM (Read-Only Memory), RAM (Random Access Memory), CD-ROM (Compact Disc Read-Only Memory), magnetic tape, floppy disk, and optical data storage devices. The computer-readable storage medium mentioned in this application can be a non-volatile storage medium; in other words, it can be a non-transient storage medium.

[0145] It should be understood that all or part of the steps of the above embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented in whole or in part as a computer program product. The computer program product includes one or more computer instructions. The computer instructions can be stored in the above-described computer-readable storage medium.

[0146] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0147] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0148] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting defects in circuit boards, characterized in that, The method, applied to circuit board defect detection equipment, includes: Obtain a target wavelength pair that matches the characteristic parameters of the circuit board under test, wherein the characteristic parameters include at least the material type, and the target wavelength pair includes a first wavelength and a second wavelength. A first control signal is sent to the wavelength tuning module, the first control signal carrying the target wavelength pair; so that the wavelength tuning module, under the action of the first control signal, sequentially adjusts its operating parameters to a first filter parameter corresponding to the first wavelength and a second filter parameter corresponding to the second wavelength; The receiving imaging module sequentially acquires the first fluorescence intensity image and the second fluorescence intensity image; Wherein, the first fluorescence intensity image is the fluorescence intensity image of the fluorescence signal generated by the circuit board under test when excited, under the first filter parameters, and the second fluorescence intensity image is the fluorescence intensity image of the fluorescence signal under the second filter parameters; The first fluorescence intensity image and the second fluorescence intensity image are subjected to differential processing to obtain a differential image; Visual inspection is performed on the differential image to obtain the defect detection results of the circuit board under test.

2. The method as described in claim 1, characterized in that, The step of performing differential processing on the first fluorescence intensity image and the second fluorescence intensity image to obtain a differential image includes: The difference image is determined based on the difference between the pixel value corresponding to the first fluorescence intensity image and the calibration pixel value corresponding to the second fluorescence intensity image; the calibration pixel value is the product of the pixel value corresponding to the second fluorescence intensity image and a preset calibration coefficient.

3. The method as described in claim 2, characterized in that, The method further includes: Acquire the first standard spectral information of the fluorescent standard sheet at the first wavelength, and the second standard spectral information at the second wavelength; The preset calibration coefficient is determined based on the first standard spectral information and the second standard spectral information.

4. The method according to any one of claims 1-3, characterized in that, The acquisition of target wavelength pairs that match the characteristic parameters of the circuit board under test includes: Acquire multiple fluorescence intensity image samples related to the feature parameters; each fluorescence intensity image sample corresponds to one wavelength; Differential processing is performed on the fluorescence intensity image samples corresponding to any two wavelengths to obtain multiple differential image samples; The two wavelengths corresponding to the difference image sample with the strongest image contrast are taken as the target wavelength pair.

5. The method as described in claim 4, characterized in that, The difference image sample with the strongest image contrast refers to the difference image sample corresponding to the maximum difference contrast value, and the method further includes: Region identification is performed on each differential image sample to obtain the line region sample and the substrate region sample of each differential image sample; Determine the difference contrast between the line region samples and the substrate region samples in each differential image sample.

6. The method as described in claim 4, characterized in that, The acquisition of multiple fluorescence intensity image samples related to the feature parameters includes: Under the condition that the preset conditions are met, a second control signal is sent to the wavelength tuning module so that the wavelength tuning module, under the action of the second control signal, sequentially adjusts the working parameters to the filter parameters corresponding to each wavelength within the first preset wavelength range according to the first preset step wavelength. The imaging module sequentially acquires multiple fluorescence intensity image samples; wherein the multiple fluorescence intensity image samples are fluorescence intensity images of fluorescence signal samples under various filter parameters, and the fluorescence signal samples are fluorescence signals generated by the fluorescence signal or circuit board samples with the characteristic parameters when excited.

7. The method as described in claim 4, characterized in that, The acquisition of multiple fluorescence intensity image samples related to the feature parameters includes: Under the condition of meeting the preset conditions, a third control signal is sent to each tunable wavelength laser source so that each tunable wavelength laser source, under the action of the third control signal, sequentially irradiates the circuit board under test or the circuit board sample with the characteristic parameters within the second preset wavelength range according to the second preset step wavelength, so as to excite the circuit board under test or the circuit board sample to generate fluorescence signals at each irradiation wavelength. The imaging module sequentially acquires multiple fluorescence intensity image samples; wherein, the multiple fluorescence intensity image samples are fluorescence intensity images of fluorescence signal samples at various irradiation wavelengths, and the fluorescence signal samples are fluorescence signals generated by the fluorescence signal or the circuit board sample upon excitation.

8. The method as described in claim 6 or 7, characterized in that, The preset condition is any one of the following: A user instruction is received, which indicates that a circuit board defect detection should be performed on a circuit board with new characteristic parameters; The interval between the previous detection cycle is greater than the preset interval; Furthermore, the difference in contrast between the line region and the substrate region in the differential image is less than a preset threshold.

9. A circuit board defect detection system, characterized in that, This includes circuit board defect detection equipment, wavelength tuning modules, and imaging modules; The circuit board defect detection equipment is used to perform the steps of the circuit board defect detection method as described in any one of claims 1-8; The wavelength tuning module is used to sequentially adjust the operating parameters to the first filter parameter corresponding to the first wavelength and the second filter parameter corresponding to the second wavelength under the action of the first control signal. The imaging module is used to sequentially acquire a first fluorescence intensity image of the fluorescence signal generated by the circuit board under test under the first filter parameter and a second fluorescence intensity image under the second filter parameter, and transmit them to the circuit board defect detection equipment.

10. The system as described in claim 9, characterized in that, The system also includes a fluorescence interception module, which includes a dichroic mirror and / or a filter; The dichroic mirror is used to filter out the light signal that excites the fluorescent signal generated by the circuit board under test, and to transmit the fluorescent signal generated by the excitation of the circuit board under test. The filter is used to filter out light signals in a third preset wavelength range and transmit light signals in a fourth preset wavelength range; the third preset wavelength range and the fourth preset wavelength range do not overlap, and the fourth preset wavelength range is determined based on the wavelength range of the fluorescence signal generated by the excitation of the circuit board under test.

11. The system as described in claim 9, characterized in that, The system also includes a linear polarization component; The linear polarization component is used to polarize the fluorescence signal after passing through the wavelength tuning module to obtain a polarized fluorescence signal; the polarized fluorescence signal indicates the fluorescence signal of the substrate region in the circuit board under test after suppression or the fluorescence signal of the line region in the circuit board under test after suppression. The imaging module is also used to sequentially acquire a first fluorescence intensity image corresponding to the polarized fluorescence signal and a second fluorescence intensity image corresponding to the polarized fluorescence signal.

12. The system according to any one of claims 9-11, characterized in that, The system also includes at least one excitation light source, the angle between the axis of the excitation light source and the horizontal plane where the circuit board under test is located is within a preset range, the at least one excitation light source is arranged in a ring around a rotation axis, and the rotation axis indicates a vertical line perpendicular to the horizontal plane; The excitation light source is used to irradiate the circuit board under test to excite the circuit board under test to generate a fluorescence signal.

13. The system as described in claim 12, characterized in that, The at least one excitation source includes a first type of excitation source and a second type of excitation source; the included angle corresponding to the first type of excitation source is located in a first preset interval, and the included angle corresponding to the second type of excitation source is located in a second preset interval; the first type of excitation source and the second type of excitation source are arranged alternately around the rotation axis, wherein the preset interval includes the first preset interval and the second preset interval, and the first preset interval and the second preset interval do not overlap.

14. The system as described in claim 12, characterized in that, The excitation source includes an illumination source and a beam-diffusing lens; The uniform light lens is fitted onto the lighting source to uniformly emit the light signal emitted by the lighting source.

15. The system as described in any one of claims 9-11, characterized in that, The imaging module includes a lens barrel and a photosensitive element, with the lens barrel located between the wavelength tuning module and the photosensitive element; The lens barrel is used to focus the fluorescence signal after passing through the wavelength tuning module onto the photosensitive element.

16. The system as described in claim 15, characterized in that, The imaging module further includes an objective lens; the objective lens is located between the circuit board under test and the fluorescence interception module, or between the fluorescence interception module and the lens barrel; The objective lens is used to magnify the circuit board under test.

17. A circuit board defect detection device, characterized in that, The circuit board defect detection device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the method as described in any one of claims 1-8.

Images