A printing wood crack detection method based on acoustic emission and digital image correlation method

Abstract

The present application relates to wood processing and quality detection technical field, disclose a kind of printing wood crack detection method based on acoustic emission and digital image correlation method, comprising the following steps: step 1: detection preparation, and the detection equipment is debugged, and printing wood sample is pretreated;Step 2: acoustic emission signal acquisition, using acoustic emission signal acquisition system to collect the acoustic emission signal generated in the detection process of printing wood;Step 3: acoustic emission signal processing, using signal processing algorithm to process the acoustic emission signal collected, extract and crack related characteristic parameters;Step 4: digital image acquisition.By acoustic emission signal acquisition processing, the characteristic parameters generated by the crack in wood are obtained, and the suspected crack area on the surface of wood is identified with the aid of digital image correlation algorithm, and then crack determination is realized by multi-modal data fusion.Compared with single detection method, the limitation that single method can only detect surface or internal crack is effectively made up.

Description

Technical Field

[0001] This invention relates to the field of wood processing and quality inspection technology, specifically a method for detecting cracks in printed wood based on acoustic emission and digital image correlation. Background Technology

[0002] Printed wood, combining the texture of natural wood with the decorative effect of printed patterns, is widely used in furniture manufacturing, interior decoration, and other fields. However, cracks, a common quality defect in the production and use of printed wood, not only reduce its mechanical properties but also affect its appearance and may even pose safety hazards. Therefore, efficient and accurate crack detection of printed wood is crucial. Currently, various detection methods exist for printed wood cracks, with single detection methods being widely used. For example, acoustic emission detection detects internal cracks by collecting acoustic emission signals generated during the propagation of internal cracks. This method has good sensitivity to internal cracks but struggles to accurately identify surface cracks, especially fine surface cracks under the printing layer, easily leading to missed detections. Digital image correlation (DIC) methods, on the other hand, acquire and analyze images of the wood surface to identify surface cracks. This method clearly presents the morphological characteristics of surface cracks but cannot obtain information about internal cracks, resulting in blind spots in internal crack detection. To compensate for the shortcomings of single detection methods, some technologies attempt to combine acoustic emission detection with digital image correlation detection for wood crack detection. However, existing combinations are mostly simple superpositions of detection data, without establishing an effective multimodal data fusion mechanism. This fails to fully leverage the synergistic effect of the two detection methods, and the accuracy and reliability of the detection need to be improved. At the same time, existing detection methods lack a systematic calibration mechanism for detection equipment. The parameters of the detection equipment are easily affected by environmental factors, leading to deviations and reduced reliability of the detection data. Furthermore, when dealing with printed wood of different materials, textures, and printing processes, it is difficult to achieve adaptive detection through parameter adjustments, resulting in poor detection versatility.

[0003] Furthermore, existing detection methods can only output crack detection results and cannot provide guidance for subsequent processing based on the results, leading to a lack of seamless integration between detection and production processes and reducing the practical application value of the detection technology. Therefore, those skilled in the art propose a crack detection method for printed wood based on acoustic emission and digital image correlation to address the above problems. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for detecting cracks in printed wood based on acoustic emission and digital image correlation, thus solving the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting cracks in printed wood based on acoustic emission and digital image correlation, comprising the following steps: Step 1: Test preparation, debugging the test equipment, and pre-processing the printed wood samples; Step 2: Acoustic emission signal acquisition. Acoustic emission signals generated by the printed wood during the inspection process are acquired using an acoustic emission signal acquisition system. Step 3: Acoustic emission signal processing. The acquired acoustic emission signal is processed using signal processing algorithms to extract characteristic parameters related to the crack. Step 4: Digital image acquisition. A digital image acquisition device is used to acquire a digital image of the surface of the wood for printing in a preset manner. Step 5: Digital image analysis. The acquired surface digital image is analyzed using digital image correlation algorithms to identify suspected crack areas on the wood surface. Step 6: Multimodal data fusion, which involves fusing the acoustic emission feature parameters extracted in Step 3 with the information on suspected crack areas identified in Step 5; Step 7: Crack assessment. Based on the fusion treatment results, determine whether there are cracks in the printed wood and the severity of the cracks.

[0006] Through the above technical solution, step 1 lays a stable and reliable detection foundation for detection preparation. Step 2 collects acoustic emission signals related to internal cracks in the wood, and step 4 acquires surface images. Steps 3 and 5 process and analyze the two types of raw data to extract effective crack information. Step 6 then uses multimodal data fusion to achieve synergistic correlation between internal and external crack information. Finally, step 7 completes the crack determination, forming a complete detection link of "preparation-acquisition-processing-fusion-determination". This link, by combining acoustic emission and digital image correlation methods, overcomes the limitation of single methods that can only detect internal or surface cracks. Multimodal data fusion further improves the comprehensiveness and reliability of crack information, effectively improving the accuracy and comprehensiveness of crack detection in printed wood, and solving the problems of missed detection and false detection by traditional single detection methods.

[0007] Preferably, the acoustic emission signal acquisition system in step 2 includes an acoustic emission sensor and a signal acquisition module, wherein the acoustic emission sensor is attached to the surface of the printed wood sample in a preset layout.

[0008] Through the above technical solution, the acoustic emission signal acquisition system in step 2 adopts a combination structure of acoustic emission sensor and signal acquisition module. With the acoustic emission sensor designed to fit the surface of printed wood in a preset layout, it can efficiently capture acoustic emission signals by utilizing the sensor's sensitivity to elastic waves generated by the energy released from crack propagation. It can also reduce signal acquisition blind spots and ensure full coverage of the detection area by using the preset layout. The signal acquisition module can also convert analog signals into processable digital signals, providing high-quality raw data for subsequent signal processing and feature extraction, effectively improving the accuracy and comprehensiveness of internal crack-related signal acquisition.

[0009] Preferably, the signal processing algorithm in step 3 includes a noise reduction algorithm and a feature extraction algorithm. The noise reduction algorithm is used to remove interference signals from the acoustic emission signal, and the feature extraction algorithm is used to extract peak value, energy, and rise time characteristic parameters from the acoustic emission signal.

[0010] Through the above technical solution, the signal processing algorithm in step 3 adopts a combination design of noise reduction algorithm and feature extraction algorithm. The noise reduction algorithm can effectively filter out useless information such as environmental noise and equipment interference in the acoustic emission signal, significantly improve the signal-to-noise ratio, and provide a clean data foundation for subsequent feature extraction. The feature extraction algorithm specifically extracts feature parameters directly related to cracks, such as peak value, energy, and rise time, to achieve accurate conversion from the original signal to crack feature information. The synergistic effect of the two provides high-quality internal crack feature basis for subsequent multimodal data fusion and crack judgment, effectively improving the accuracy of crack detection.

[0011] Preferably, the digital image acquisition device in step 4 is an industrial camera, and the preset method includes a preset shooting angle and preset lighting conditions, wherein the preset lighting conditions are uniform white light illumination.

[0012] Through the above technical solution, step 4 uses an industrial camera as a digital image acquisition device, combined with a preset shooting angle and a preset method of uniform white light illumination. The high resolution and stability of the industrial camera can ensure image clarity, the preset shooting angle can achieve full coverage of the detection area to reduce missed shots, and uniform white light illumination can weaken the interference of printing texture and improve the contrast between cracks and background. The three work together to provide high-quality surface image data for subsequent digital image analysis to accurately identify suspected crack areas, effectively improving the reliability of surface crack detection.

[0013] Preferably, the digital image correlation algorithm in step 5 includes an image preprocessing sub-algorithm, a sub-pixel matching sub-algorithm, and a displacement field calculation sub-algorithm. The displacement field calculation sub-algorithm is used to obtain the displacement distribution on the wood surface, thereby identifying suspected crack areas.

[0014] Through the above technical solution, the digital image correlation algorithm in step 5 adopts a collaborative design of three sub-algorithms: image preprocessing, sub-pixel matching, and displacement field calculation. The image preprocessing sub-algorithm can optimize image quality and weaken interference, the sub-pixel matching sub-algorithm can improve displacement capture accuracy, and the displacement field calculation sub-algorithm can locate suspected crack areas by obtaining the surface displacement distribution and identifying the displacement abrupt change features at the crack. The three work together to effectively overcome the interference of the printing layer texture, improve the accuracy of identifying fine surface cracks, and provide reliable surface crack information support for multimodal data fusion.

[0015] Preferably, the crack determination in step 7 includes: determining the existence of a crack when the acoustic emission characteristic parameters exceed a preset threshold and a continuous suspected crack area is identified in the digital image; and quantifying the severity of the crack based on the magnitude of the acoustic emission characteristic parameters and the length and width of the suspected crack area.

[0016] Through the above technical solution, step 7 uses the dual verification of acoustic emission characteristic parameter thresholds and continuous suspected crack areas to determine the existence of cracks. This can avoid misjudgment from a single dimension by mutually verifying internal and external information. Combining the acoustic emission parameter values ​​and the geometric dimensions of the suspected crack area to quantify the severity, the judgment is improved from qualitative to quantitative. The two work together to ensure the accuracy and scientific nature of crack judgment and provide a reliable basis for subsequent processing.

[0017] Preferably, it also includes a step of generating subsequent processing suggestions: based on the crack determination results in step 7, generating suggestions for repairing, discarding, or adjusting the processing technology of the printed wood.

[0018] Through the above technical solution, the subsequent processing suggestion generation step is based on the crack judgment result in step 7, and corresponding suggestions such as repair, scrapping or process adjustment are output. This breaks the limitation of traditional detection that only outputs results, realizes the connection between detection and production processing, not only improves the practical value of the method, but also reduces crack defects from the source through process adjustment suggestions, and helps to improve production quality and efficiency.

[0019] Preferably, the multimodal data fusion in step 6 adopts a weighted fusion algorithm, which assigns weights based on the detection confidence of acoustic emission signals and digital images.

[0020] Through the above technical solution, step 6 adopts a weighted fusion algorithm that assigns weights based on the credibility of acoustic emission signals and digital image detection, rather than simply superimposing data. This can highlight the guiding role of high-credibility data, achieve accurate coordination of internal and external crack information, avoid the limitations of single-modal data, improve the comprehensiveness and reliability of fusion results, and provide core basis for accurate crack judgment.

[0021] Preferably, the system calibration step is also included: after step 1, the acoustic emission signal acquisition system and digital image acquisition device are calibrated using standard crack samples to ensure detection accuracy.

[0022] Through the above technical solution, the system calibration step uses standard crack samples to calibrate the equipment parameters after the test preparation. This can correct the parameter deviations of the acoustic emission signal acquisition system and digital image acquisition equipment caused by the environment or aging, ensure the accuracy of the original test data, and provide basic support for the accuracy of subsequent signal processing, data fusion and crack judgment from the source, thereby improving the stability and reliability of the detection method.

[0023] This invention provides a method for detecting cracks in printed wood based on acoustic emission and digital image correlation. It has the following advantages: 1. This invention acquires characteristic parameters of internal cracks in wood through acoustic emission signal acquisition and processing, and identifies suspected crack areas on the wood surface using digital image correlation algorithms. Then, it achieves crack determination through multimodal data fusion. Compared with single detection methods, it effectively makes up for the limitation of single methods that can only detect surface or internal cracks, improves the accuracy and comprehensiveness of crack detection in printed wood, and further ensures the reliability of fusion results by allocating weights according to the detection credibility through a weighted fusion algorithm.

[0024] 2. This invention, by setting up a system calibration step and using standard crack samples to calibrate the parameters of the acoustic emission signal acquisition system and digital image acquisition equipment, ensures the accuracy of the detection equipment parameters and provides a foundation for the reliability of subsequent detection data. Simultaneously, it can generate subsequent processing suggestions such as repair, disposal, or adjustment of processing technology based on the crack judgment results, achieving a connection between detection and subsequent processing guidance, improving the practicality of the method and its guiding value for actual production. Furthermore, it can adjust relevant parameters for printed wood of different materials, textures, or printing processes, enhancing the method's adaptability. Attached Figure Description

[0025] Figure 1 This is the overall flowchart of the present invention; Figure 2 This is a flowchart of the acoustic emission and digital image analysis process of the present invention. Detailed Implementation

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see the appendix Figure 1 - Appendix Figure 2 This invention provides a method for detecting cracks in printed wood based on acoustic emission and digital image correlation, comprising the following steps: Step 1: Test preparation, debugging the test equipment, and pre-processing the printed wood samples; Step 2: Acoustic emission signal acquisition. Acoustic emission signals generated by the printed wood during the inspection process are acquired using an acoustic emission signal acquisition system. The acoustic emission signal acquisition system in step 2 includes an acoustic emission sensor and a signal acquisition module. The acoustic emission sensor is attached to the surface of the printed wood sample in a preset layout.

[0028] Specifically, step 1, the preparation for testing, involves debugging the testing equipment and pre-processing the printed wood samples. This provides a foundation for the stability of the subsequent testing process and the reliability of the test data. Equipment debugging ensures that the acoustic emission signal acquisition system and digital image acquisition equipment are in optimal working condition, while sample pre-processing removes impurities or interfering factors from the surface of the printed wood that may affect the test results, preventing deviations in subsequent signal acquisition and image acquisition. Step 2, the acoustic emission signal acquisition, is the core step in obtaining information about internal cracks in the wood. It employs an acoustic emission signal acquisition system composed of acoustic emission sensors and a signal acquisition module. The acoustic emission sensors are attached in a preset layout. By integrating the acoustic emission sensor with the surface of printed wood samples and utilizing its sensitivity to elastic waves generated by crack propagation energy, the system efficiently captures acoustic emission signals generated during the printing wood inspection process. The preset layout ensures that the sensor can fully cover the inspection area, reducing signal acquisition blind spots. The signal acquisition module converts the analog signals captured by the sensor into digital signals that can be processed later, providing high-quality raw data for subsequent acoustic emission signal processing and feature parameter extraction. This step works in conjunction with the inspection preparation steps to effectively improve the targeting and accuracy of acoustic emission signal acquisition, laying a crucial foundation of internal crack information for subsequent multimodal data fusion and accurate crack determination.

[0029] Step 3: Acoustic emission signal processing. The acquired acoustic emission signal is processed using signal processing algorithms to extract characteristic parameters related to the crack. The signal processing algorithms in step 3 include a noise reduction algorithm and a feature extraction algorithm. The noise reduction algorithm is used to remove interference signals from the acoustic emission signal, and the feature extraction algorithm is used to extract peak value, energy, and rise time characteristic parameters from the acoustic emission signal.

[0030] Specifically, step 3, acoustic emission signal processing, is a crucial link between acoustic emission signal acquisition and multimodal data fusion. Its core principle lies in performing layered processing on the acquired raw acoustic emission signal through a combination of denoising and feature extraction algorithms. The denoising algorithm filters out unwanted signals inevitably mixed in during acquisition, such as environmental noise and equipment operation interference. By suppressing signal interference from non-crack sources, it significantly improves the signal-to-noise ratio of the raw signal, providing a clean signal foundation for subsequent feature extraction. The feature extraction algorithm, based on the specific characteristics of acoustic emission signals generated by crack propagation in dimensions such as peak value, energy, and rise time, accurately extracts the aforementioned feature parameters directly related to the crack, realizing the transformation from continuous signal to discrete crack feature information. This step, through the synergistic effect of the two algorithms, effectively eliminates invalid interference information and filters out key features that directly reflect the existence and development state of the crack. It solves the problem of difficult crack information identification caused by the disorder of the raw acoustic emission signal, providing high-quality, highly targeted internal crack feature data for subsequent fusion processing with surface crack information obtained by digital image correlation methods, thus providing core data support for the accuracy of crack determination.

[0031] Step 4: Digital image acquisition. A digital image acquisition device is used to acquire a digital image of the surface of the wood for printing in a preset manner. The digital image acquisition device in step 4 is an industrial camera. The preset methods include preset shooting angle and preset lighting conditions. The preset lighting conditions are uniform white light illumination.

[0032] Specifically, step 4, digital image acquisition, is the core step in capturing information about cracks on the surface of printed wood. Its principle lies in using an industrial camera as the digital image acquisition device. Leveraging the high resolution and high stability of the industrial camera, combined with preset shooting angles and uniform white light illumination, accurate acquisition of digital images of the printed wood surface is achieved. The preset shooting angle ensures that the camera can fully cover the wood surface detection area, avoiding missed images of surface cracks due to shooting angle deviations. Uniform white light illumination effectively weakens the texture interference of the printed pattern on the wood surface, improves the contrast between the wood surface and potential crack areas, and reduces image distortion caused by uneven illumination. This step, through the collaborative design of the industrial camera and the preset acquisition method, solves the problem of unclear imaging of fine surface cracks under the printed layer. The acquired high-definition, high-contrast surface digital images provide a high-quality image data foundation for the subsequent digital image-related algorithms in step 5 to accurately identify suspected crack areas. This complements the internal crack information acquired in steps 2 and 3, providing a key surface feature data source for multimodal data fusion to achieve comprehensive crack detection.

[0033] Step 5: Digital image analysis. The acquired surface digital image is analyzed using digital image correlation algorithms to identify suspected crack areas on the wood surface. The digital image correlation algorithm in step 5 includes an image preprocessing sub-algorithm, a sub-pixel matching sub-algorithm, and a displacement field calculation sub-algorithm. The displacement field calculation sub-algorithm is used to obtain the displacement distribution on the wood surface, thereby identifying suspected crack areas.

[0034] Specifically, step 5, digital image analysis, is the core step in identifying suspected crack areas on the surface of printed wood. Its principle is based on the precise perception characteristics of digital image correlation algorithms for changes in the surface morphology of materials. The detection target is achieved through hierarchical collaboration of image preprocessing sub-algorithms, sub-pixel matching sub-algorithms, and displacement field calculation sub-algorithms. The image preprocessing sub-algorithm performs denoising and enhancement on the surface digital image obtained in step 4, effectively suppressing interference factors such as printing layer texture and residual ambient light, and improving the grayscale contrast between crack and background areas, providing a high-quality image foundation for subsequent accurate matching. The sub-pixel matching sub-algorithm optimizes the matching of feature points in the preprocessed image, breaking through the accuracy bottleneck of traditional pixel-level matching, and accurately capturing minute displacements on the wood surface, providing high-precision data support for displacement field calculation. The displacement field calculation sub-algorithm solves for the global displacement distribution on the wood surface based on the matching results, utilizing the abrupt displacement characteristics caused by structural discontinuities in crack areas to accurately locate and identify suspected crack areas. This step, through the collaborative action of multiple sub-algorithms, solves the technical problems of low accuracy and weak interference resistance in identifying fine surface cracks under the printed layer. The output information of suspected crack areas complements the internal crack feature parameters extracted in step 3, providing a highly reliable basis for surface crack characterization for subsequent multimodal data fusion and accurate crack determination, and significantly improving the overall detection method's ability to identify surface cracks.

[0035] Step 6: Multimodal data fusion, which involves fusing the acoustic emission feature parameters extracted in Step 3 with the information on suspected crack areas identified in Step 5; The multimodal data fusion in step 6 adopts a weighted fusion algorithm, which assigns weights based on the detection confidence of acoustic emission signals and digital images.

[0036] Specifically, step 6, multimodal data fusion, is the core step in connecting internal and surface crack information to achieve comprehensive detection. Its principle is based on the complementarity of acoustic emission signals reflecting the characteristics of internal cracks in wood and digital image information representing the state of surface cracks. A weighted fusion algorithm is used to construct a collaborative correlation mechanism between the two types of data. This weighted fusion algorithm does not simply superimpose the two types of data; instead, it first analyzes the reliability of acoustic emission characteristic parameters (peak value, energy, etc.) and suspected crack area information under different detection scenarios. For example, acoustic emission signals have higher reliability in scenarios dominated by internal cracks, while digital image information has better reliability in scenarios with minor surface cracks. Weights are then dynamically allocated accordingly to achieve accurate fusion of the two types of data. This step, through differentiated weight allocation in the weighted fusion algorithm, effectively avoids the limitations of single-modal data and solves the detection bias problem caused by simple superposition of multiple data in existing technologies. The fused comprehensive information can fully cover both internal and external crack characteristics while highlighting the guiding role of high-reliability data. This provides a core basis for accurate crack determination in the subsequent step 7, balancing comprehensiveness and reliability, and significantly improves the overall accuracy of crack detection in printed wood.

[0037] Step 7: Crack assessment. Based on the fusion treatment results, determine whether there are cracks in the printed wood and the severity of the cracks.

[0038] The crack determination in step 7 includes: when the acoustic emission characteristic parameters exceed the preset threshold and a continuous suspected crack area is identified in the digital image, a crack is determined to exist; the severity of the crack is quantified based on the magnitude of the acoustic emission characteristic parameters and the length and width of the suspected crack area.

[0039] Specifically, the crack determination in step 7, as the core output of the entire detection method, is based on the collaborative verification characteristics of internal and external crack feature information after multimodal fusion. It achieves accurate determination through a "dual verification + quantitative evaluation" mechanism. The crack existence determination employs a dual verification logic of acoustic emission feature parameters and suspected crack areas. Specifically, a crack can only be determined to exist when the acoustic emission feature parameters extracted in step 3 exceed a preset threshold (reflecting the possibility of internal cracks) and step 5 identifies continuous suspected crack areas (reflecting the possibility of surface cracks). This logic effectively avoids misjudgments that may occur from a single information source through mutual verification of internal and external crack information. For example, relying solely on acoustic emission signals can easily lead to misjudgments of environmental interference as cracks, and relying solely on images can easily lead to misjudgments of printed textures as cracks. Crack severity quantification combines the numerical value of acoustic emission feature parameters (reflecting the severity of internal crack propagation) with the length and width of suspected crack areas (reflecting the geometric dimensions of surface cracks), achieving a comprehensive quantitative evaluation of crack severity and avoiding the one-sidedness of a single-dimensional evaluation. This step ensures the accuracy of crack identification through dual verification and achieves the scientific nature of crack severity assessment through multi-dimensional quantification. It solves the problems of easy misjudgment and inaccurate severity assessment in existing detection methods, and provides a reliable basis for generating targeted follow-up treatment suggestions, significantly improving the practical value of the detection method.

[0040] It also includes a follow-up processing suggestion generation step: based on the crack determination results in step 7, suggestions for repairing, discarding, or adjusting the processing technology of printed wood are generated.

[0041] Specifically, the follow-up processing suggestion generation step is a crucial link connecting crack detection results with actual production applications. Its principle is based on the correlation between the crack judgment results (including quantitative information on the presence and severity of cracks) output in step 7 and the production and processing needs of printed wood, establishing a matching mechanism between the judgment results and processing strategies. This step classifies crack severity into different levels, corresponding to targeted suggestions such as repair, disposal, or processing technology adjustments—for example, minor cracks correspond to suggestions for local repair process parameters, severe cracks correspond to suggestions for disposal, and common crack problems found in batch testing correspond to suggestions for adjusting processing technology parameters. This step overcomes the limitations of traditional detection methods that only output detection results, solving the technical problem of poor connection between detection and production processing. It enables detection results to be directly transformed into specific solutions to guide production practice, not only enhancing the practical application value of detection technology but also reducing the generation of crack defects from the source through process adjustment suggestions, helping to improve the quality and efficiency of printed wood production, and enhancing the practicality and industry adaptability of the entire detection method.

[0042] It also includes a system calibration step: after step 1, standard crack samples are used to calibrate the parameters of the acoustic emission signal acquisition system and the digital image acquisition device to ensure detection accuracy.

[0043] Specifically, based on the characteristic that standard crack samples have clearly defined crack parameters (such as size, location, and morphology), this step is introduced after the detection preparation in step 1. Standard crack samples are used to calibrate the parameters of the acoustic emission signal acquisition system and the digital image acquisition device, establishing a correspondence between the output data of the detection device and the true parameters of the standard crack. Specifically, for the acoustic emission signal acquisition system, the known characteristic acoustic emission signals generated by the standard crack sample during detection are used to adjust parameters such as sensor sensitivity and signal acquisition module gain to ensure the system can accurately capture and convert the acoustic emission signal corresponding to the crack. For the digital image acquisition device, the clear surface image of the standard crack sample is used to adjust the focal length, exposure parameters, and light intensity of the industrial camera to ensure the device can accurately image and present the true morphology of the standard crack. This step addresses the problem of data distortion and decreased accuracy caused by drift in existing detection methods due to factors such as environmental temperature and humidity and equipment aging affecting equipment parameters. By calibrating with standard samples, the equipment is always kept in the optimal parameter state to meet the detection requirements, providing a highly reliable raw data foundation for subsequent acoustic emission signal acquisition, digital image acquisition, and subsequent processing and analysis. This improves the stability of the entire detection method and the accuracy of crack identification from the source.

[0044] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for detecting cracks in printed wood based on acoustic emission and digital image correlation, characterized in that, Includes the following steps: Step 1: Test preparation, debugging the test equipment, and pre-processing the printed wood samples; Step 2: Acoustic emission signal acquisition. Acoustic emission signals generated by the printed wood during the inspection process are acquired using an acoustic emission signal acquisition system. Step 3: Acoustic emission signal processing. The acquired acoustic emission signal is processed using signal processing algorithms to extract characteristic parameters related to the crack. Step 4: Digital image acquisition. A digital image acquisition device is used to acquire a digital image of the surface of the wood for printing in a preset manner. Step 5: Digital image analysis. The acquired surface digital image is analyzed using digital image correlation algorithms to identify suspected crack areas on the wood surface. Step 6: Multimodal data fusion, which involves fusing the acoustic emission feature parameters extracted in Step 3 with the information on suspected crack areas identified in Step 5; Step 7: Crack assessment. Based on the fusion treatment results, determine whether there are cracks in the printed wood and the severity of the cracks.

2. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, The acoustic emission signal acquisition system in step 2 includes an acoustic emission sensor and a signal acquisition module. The acoustic emission sensor is attached to the surface of the printed wood sample in a preset layout.

3. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, The signal processing algorithm in step 3 includes a noise reduction algorithm and a feature extraction algorithm. The noise reduction algorithm is used to remove interference signals from the acoustic emission signal, and the feature extraction algorithm is used to extract peak value, energy, and rise time characteristic parameters from the acoustic emission signal.

4. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, The digital image acquisition device in step 4 is an industrial camera, and the preset method includes a preset shooting angle and preset lighting conditions, wherein the preset lighting conditions are uniform white light illumination.

5. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, The digital image correlation algorithm in step 5 includes an image preprocessing sub-algorithm, a sub-pixel matching sub-algorithm, and a displacement field calculation sub-algorithm. The displacement field calculation sub-algorithm is used to obtain the displacement distribution on the wood surface, thereby identifying suspected crack areas.

6. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, The crack determination in step 7 includes: when the acoustic emission characteristic parameters exceed a preset threshold and a continuous suspected crack area is identified in the digital image, a crack is determined to exist; the severity of the crack is quantified based on the magnitude of the acoustic emission characteristic parameters and the length and width of the suspected crack area.

7. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, It also includes a follow-up processing suggestion generation step: based on the crack determination results in step 7, suggestions for repairing, discarding, or adjusting the processing technology of printed wood are generated.

8. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, The multimodal data fusion in step 6 employs a weighted fusion algorithm, which assigns weights based on the detection confidence of the acoustic emission signal and the digital image.

9. The method for detecting cracks in printed wood based on acoustic emission and digital image correlation as described in claim 1, characterized in that, It also includes a system calibration step: after step 1, standard crack samples are used to calibrate the parameters of the acoustic emission signal acquisition system and the digital image acquisition device to ensure detection accuracy.

Images