489results about "Flaw detection using microwaves" patented technology

The invention belongs to the technical field of microwave measurement, and discloses an outer wall hollowing microwave reflection detection method based on multi-modal fusion, which comprises the following steps: carrying out multi-modal scanning on a building outer wall to be detected, obtaining visible light image data and infrared temperature distribution data of the outer wall surface, and carrying out space registration and coordinate mapping; a unified multi-modal fusion data set is formed; performing anomaly screening on the multi-modal fusion data set, and identifying a thermal anomaly region by analyzing infrared temperature distribution data; detecting a bump or crack area in combination with texture and morphology anomaly features of the visible light image data; performing information fusion on the thermal anomaly region and the bump or crack region, and extracting candidate detection regions of suspected hollowing; high-precision recognition and quantitative evaluation of the outer wall hollowing are achieved, and the precision and stability of outer wall hollowing detection are improved.

The invention relates to the technical field of cable nondestructive testing, in particular to a cable defect detection system. The method mainly solves the problems that in the prior art, early-stage tiny defects in a cable are insufficient in recognition sensitivity, the omission ratio is high, and accurate classification cannot be achieved. According to the system, signals are synchronously collected through a multi-physics field composite sensing array, excitation parameters are optimized through a genetic algorithm, a three-dimensional defect probability graph is generated through fusion, a double-current Transform model is adopted to deeply analyze features and automatically recognize defect types, and finally, maintenance is guided through AR visualization, so that non-intrusive online accurate diagnosis and intelligent operation and maintenance of cable defects are achieved.

The invention discloses a joint test system, a silicon carbide metasurface grating detection method and a silicon carbide metasurface grating detection device, and belongs to the technical field of semiconductor detection. A synergistic excitation signal is generated by regulating a light source and a microwave source and is applied to the silicon carbide metasurface grating to be measured to form a light intensity and microwave field interference pattern. The method comprises the following steps: synchronously acquiring data through multiple channels, constructing a multi-dimensional matrix through time alignment, marker feature extraction and coordinate mapping, extracting defect features to establish a reference model, calculating deviation degree to judge quality, and carrying out abnormal point screening, candidate region construction and coupling verification on a defect grating to divide a potential defect region. For a potential defect area, excitation parameters are dynamically adjusted through a gradient descent algorithm, and defect grades are divided by combining the characteristic defect degree and the area. According to the invention, the problems of insufficient single-mode detection, low multi-mode alignment precision and blind adjustment of excitation parameters are solved, cross-band cooperative detection and intelligent grading are realized, and the detection reliability is improved.

A non-invasive liquid integrity monitoring system using dielectric fingerprinting and machine learning to detect and identify contamination in sealed containers is described. The system may employ externally-mounted sensors that measure dielectric properties through electromagnetic interrogation, comparing measurements against baseline signatures to detect deviations indicating contamination, tampering, or degradation. Industry-specific ML models enable identification of specific contaminants with confidence, providing alerts without breaching container integrity.

The invention discloses a fire-fighting non-destructive detection method based on multi-modal detection, and relates to the technical field of fire-fighting engineering, and the method comprises the following steps: carrying out the rapid scanning of fire-fighting equipment through an ultrasonic array, analyzing and positioning a suspected defect region based on a time domain reflection signal, and generating a three-dimensional coordinate mapping model; and performing multi-angle X-ray projection acquisition on the positioning area, and generating an internal structure cross-sectional image through a three-dimensional reconstruction algorithm. According to the fire-fighting non-destructive detection method provided by the invention, by integrating various technical means such as ultrasonic array scanning, X-ray tomography, thermal imaging analysis and microwave sealing detection, comprehensive and accurate identification of potential defects of fire-fighting equipment is realized, and the dimensionality and precision of defect detection are improved; the structural integrity of the equipment is kept to the maximum extent, extra risks caused by destructive detection are avoided, and the reliability and accuracy of detection are improved through a data fusion algorithm.

The invention discloses a three-dimensional ground penetrating radar map generation method. The method comprises the following steps that S1, a three-dimensional ground penetrating radar is used for collecting selected road internal structure information to obtain original data; s2, preprocessing the obtained original data; s3, carrying out normalization processing and trilinear interpolation on the longitudinal section data of the preprocessed radar data; s4, generating a three-dimensional ground penetrating radar map, wherein the three-dimensional ground penetrating radar map at least comprises a single-channel oscillogram, a slice map, a longitudinal profile map and a three-dimensional map; and S5, based on the three-dimensional diagram, introducing a gradient enhanced adaptive perspective algorithm, realizing perspective of the three-dimensional diagram, and highlighting a disease area. According to the method, an effective and convenient standardized map is formulated in combination with data collected by the three-dimensional ground penetrating radar, the problems of section discretization and information discontinuity are solved, and through the generated three-dimensional perspective drawing, the visualization effect of a road disease area is effectively enhanced, and accurate positioning and recognition of road internal diseases are achieved.

A non-invasive liquid integrity monitoring system using dielectric fingerprinting and machine learning to detect and identify contamination in sealed containers is described. The system may employ externally-mounted sensors that measure dielectric properties through electromagnetic interrogation, comparing measurements against baseline signatures to detect deviations indicating contamination, tampering, or degradation. Industry-specific ML models enable identification of specific contaminants with confidence, providing alerts without breaching container integrity.

An RF-based material detection device includes an antenna connector, a first interchangeable antenna capable of being releasably coupled to the RF-based material detection device via the antenna connector, an RF transmitter unit operably connected to the antenna connector and configured to transmit an RF signal via the first interchangeable antenna into a first material at a specific resonance frequency for the first material, and an RF receiver unit operably connected to the antenna connector and configured to receive from the first material via the first interchangeable antenna a first modified signal in response to interaction of the RF signal with the first material, wherein the first interchangeable antenna includes one or more design characteristics optimized to detect the first modified signal from the first material.

The invention relates to the technical field of tunnel defect identification, in particular to a tunnel defect identification method based on a dielectric distribution diagram. According to the method, on the basis of AI rock-soil perspective radar field monitoring, gprMax simulation and laboratory electromagnetic data, a two-dimensional dielectric distribution diagram is generated through variational Bayesian inversion fusion. Through multi-scale spectral clustering and expert knowledge, a potential abnormal region is automatically identified, and then accurate classification and identification of a defect region are realized by adopting graph form constraint propagation and integrating a Transform-graph neural network model. And finally, projecting an identification result to an original image, and generating a visual defect labeling layer and a structured report. According to the invention, high-precision, automatic, visual and structured detection of tunnel structure defects is realized, the intelligent level and risk early warning capability of tunnel safety operation and maintenance are significantly improved, and the digitization and intelligent process of tunnel operation and maintenance management is promoted.

An RF-based material detection device includes an antenna connector, a first interchangeable antenna capable of being releasably coupled to the RF-based material detection device via the antenna connector, an RF transmitter unit operably connected to the antenna connector and configured to transmit an RF signal via the first interchangeable antenna into a first material at a specific resonance frequency for the first material, and an RF receiver unit operably connected to the antenna connector and configured to receive from the first material via the first interchangeable antenna a first modified signal in response to interaction of the RF signal with the first material, wherein the first interchangeable antenna includes one or more design characteristics optimized to detect the first modified signal from the first material.

The invention belongs to the technical field of structural damage monitoring, and particularly relates to a structural dynamic tomography and damage monitoring method using self-excitation waves, which comprises the following steps: S1, carrying out spatial discretization processing on a monitored object, and establishing a three-dimensional discrete model; arranging a plurality of sensors on a monitoring object; recording the space coordinates of each sensor; s2, applying external excitation to a known position, and inverting an initial wave velocity field in the monitoring object; s3, extracting self-excitation waves of the monitored object received by each sensor; s4, identifying the arrival time of each self-excitation wave signal on each sensor; the space coordinates of the wave source of each excitation wave signal are calculated through inversion; s5, synchronously updating the wave velocity field based on the space coordinates of the wave sources of the respective excitation waves, and processing to obtain a damage result of the monitored object; and S6, performing corresponding processing based on a damage result. According to the method, dynamic tomography and real-time monitoring of the damage position, range and evolution process can be realized.

The invention discloses a data processing method and system for intelligent road detection and related equipment, and relates to the technical field of traffic engineering and artificial intelligence crossing. In the road detection process, multi-sensor fusion equipment with dynamic adjustment is adopted to collect multi-source data; the real-time environment perception model based on deep reinforcement learning dynamically optimizes the parameters of the multi-modal sensor, and improves the data acquisition quality in severe weather and environment. Millimeter-level alignment of multi-source data is realized by adopting a spatio-temporal model, and a generative adversarial network is introduced to enhance the multi-source data after spatio-temporal alignment so as to supplement missing data details and improve data quality. Disease intelligent identification is carried out on multi-source data through a sample learning model and a meta-learning algorithm, the sample learning model solves the problem of insufficient labeling samples of micro-disease data, the meta-learning algorithm is quickly generalized on the basis of a small amount of labeling data, micro-diseases such as early cracks and hidden road structure layer defects are identified, and the accuracy of identification is improved. And the accuracy of road disease identification is improved.

The invention provides a device and method for microwave nondestructive testing of a PE pipeline electric melting joint, the device comprises a walking module, a signal module and a control module, the walking module comprises a fixed part and a moving part, the fixed part comprises a plurality of pipeline clamps, and the pipeline clamps wrap the periphery of a PE pipeline; any two pipeline clamps are connected through a clamp connecting fitting, the moving part comprises a rotary moving mechanism, an X-axis moving mechanism and a Z-axis moving mechanism, and the signal module and the control module are connected and both installed on the walking module. According to the method, the defect detection precision is improved, the detection safety is enhanced, online real-time monitoring is realized, manual intervention is reduced, the adaptability and flexibility are improved, and comprehensive data support is provided.

The invention provides a lossless identification and quantitative evaluation method and system for railway tunnel lining defects, and particularly relates to the technical field of railway engineering detection and intelligent identification. The method comprises the following steps: firstly, collecting a multi-source detection signal, and carrying out adaptive preprocessing on the multi-source detection signal; performing multi-level fusion on the preprocessed multi-source detection signals in a unified space coordinate system to generate a high-precision data set; and finally, inputting the high-precision data set into a multi-task lining defect evaluation model so as to realize refined identification and quantitative evaluation of the railway tunnel lining defects. According to the method, the automation and objectivity of lining defect detection are remarkably improved, and a reliable basis is provided for maintenance decision-making of a tunnel structure.

The invention relates to a tunnel lining quality detection method, which comprises the following steps of: acquiring tunnel lining ground penetrating radar data, and preprocessing the tunnel lining ground penetrating radar data; carrying out image segmentation on the preprocessed radar image, and dividing the preprocessed radar image into a lining thickness region, a steel bar distribution region and other background regions; counting the number of reinforcing steel bars in the reinforcing steel bar distribution area through a polymorphic detection method; on the basis of the segmented lining thickness area, the actual lining thickness is calculated in combination with the dielectric constant difference; inputting actually measured radar data into the cavity recognition model, and outputting a cavity recognition result; and generating a tunnel lining quality evaluation report by integrating the number of the reinforcing steel bars, the actual lining thickness and the cavity identification result. According to the prediction method provided by the invention, integrated operation of lining thickness measurement, steel bar statistics and defect detection is realized through multi-task collaborative optimization, the detection efficiency and precision are improved, and a reliable technical scheme is provided for tunnel lining quality evaluation.

The invention relates to the technical field of radar signal processing, in particular to a ground penetrating radar reinforcing steel bar clutter suppression method based on physical constraints, and mainly solves the problem that a data set is difficult to obtain in an existing ground penetrating radar reinforcing steel bar clutter removal method based on deep learning. The method is an improved method based on the CUT network, the CUT network structure and a comparative learning mechanism determine that the requirement of the network for the data size of a data set is low, a waveform smoothness constraint is added on this basis, the clutter removal effect and generalization ability are improved by introducing physical prior, the physical constraint serves as a regularization item, and the regularization efficiency is improved. The problem that a CUT network is prone to model collapse under a small data set is solved. Finally, the improved model is compared with other models through different evaluation indexes, and the result shows that the improved model has more advantages.

The invention discloses a road hidden disease identification, positioning and segmentation method based on a three-dimensional ground penetrating radar, and belongs to the technical field of road hidden disease detection. The problems that the traditional method is low in road hidden disease detection efficiency and poor in accuracy and cannot perform automatic three-dimensional reconstruction on road diseases are solved. According to the invention, by establishing a YOLO-nnUNet three-dimensional ground penetrating radar disease detection-segmentation joint algorithm, a specific method for rapid positioning, identification and three-dimensional segmentation of hidden road diseases in three-dimensional ground penetrating radar data is provided. According to the method, a complete disease identification process including three-dimensional ground penetrating radar data acquisition, two-dimensional disease defect detection, three-dimensional disease clustering positioning and three-dimensional disease boundary semantic segmentation is realized, the efficiency and accuracy of road hidden disease detection are improved, meanwhile, road hidden disease segmentation data of the three-dimensional ground penetrating radar can be obtained, and the road hidden disease detection accuracy is improved. And outputting a three-dimensional reconstruction result of the road disease. The method can be applied to road hidden disease detection.

The invention discloses a lossless detection mold and method for a rammed earth wall, and relates to the technical field of civil engineering and cultural relic protection. In order to solve the problem that in a traditional rammed earth wall nondestructive detection method, it is difficult to establish an effective corresponding relation between signal features and physical and mechanical parameters, the rammed earth wall nondestructive detection mold and method comprise the steps that rammed earth materials are prepared with reference to the soil texture of an ancient wall, and the rammed earth materials are filled into the rammed earth mold in a layered mode to prepare a wall-like test piece; carrying out nondestructive detection on the city-wall-like test piece by adopting a geological radar to obtain an oscillogram and a grey-scale map, and extracting a detection signal; sampling the city-wall-like test piece, carrying out a physical and mechanical test, and establishing a parameter database in which detection signal parameters are associated with physical and mechanical parameters; and finally, carrying out field detection on the ancient city wall, carrying out geological radar detection to obtain an actual detection signal, and matching the actual detection signal with the parameter database to obtain physical and mechanical parameter information of the ancient city wall. The method can effectively improve the accuracy of lossless detection of the rammed earth structure, and provides a scientific basis for ancient city wall protection.

The invention discloses a cable trench fireproof cover plate performance detection device and method based on multi-band electromagnetic waves. The device comprises a microwave detection module, a terahertz imaging module, a high-precision two-dimensional mechanical scanning platform, a synchronous control and data fusion unit and a comprehensive data analysis platform. The microwave module is used for transmitting and receiving 1-10GHz electromagnetic waves to realize deep structure detection; the terahertz module works at a frequency band of 0.1-3 THz and is used for acquiring near-surface time-domain spectral information. And the scanning platform drives the modules to perform two-dimensional rasterization scanning on the surface of the cover plate. And the synchronous control and data fusion unit coordinates the time sequence action of each module to ensure the time-space consistency of data acquisition. And the comprehensive data analysis platform performs fusion processing on the multi-source data through an inversion and imaging algorithm to generate a comprehensive evaluation report covering internal macroscopic defects, near-surface microdefects and coating thickness. According to the invention, through collaborative integration of microwave and terahertz technologies, multi-scale and integrated nondestructive testing and performance evaluation of the fireproof cover plate from a deep layer to a surface layer are realized.

The invention relates to the technical field of road detection, in particular to a multi-dimensional road disease detection method and system based on a ground penetrating radar. The method comprises the following steps: acquiring original ground penetrating radar data, performing DC component removal, gain adjustment and background denoising on the original ground penetrating radar data, positioning a suspected disease area and defining the suspected disease area as a disease entity; extracting a time domain feature, a frequency domain feature and a spatial context feature of each disease entity in parallel to form a multi-dimensional feature vector; constructing a road disease knowledge graph according to the multi-dimensional feature vectors, and distributing an initial weight for each graph relation; and performing node matching operation according to the multi-dimensional feature vector and a knowledge graph, performing logical reasoning according to a graph relation path, fusing similarity and reasoning confidence, and outputting a diagnosis tag signal and a confidence signal. According to the method, the positioning accuracy of the suspected disease area is improved, comprehensive utilization of multi-dimensional information is realized, and a visual causal link is provided for diagnosis.

The microwave detection robot comprises a first mounting seat, an electric telescopic rod is hinged to the first mounting seat, a second mounting seat is hinged to the output end of the electric telescopic rod, and machine body assemblies are mounted on the second mounting seat and the first mounting seat; the double-body assembly is designed, the robot is attached to a to-be-detected surface through airflow generated by a first impeller, negative pressure suction force is generated in a ring groove through a second impeller, and the negative pressure suction force is transmitted to the moving wheel assembly through a pipeline structure, so that the road holding force of the moving wheel assembly is enhanced, and the load capacity of the robot is improved; the angle between the two machine body assemblies is adjusted through the electric telescopic rods, so that the robot has the obstacle crossing ability, and the adaptability of the robot during operation in the complex environment is improved; the environment compensation module adopts an intelligent algorithm to compensate original microwave signals, so that the problem of low detection result precision caused by environmental factor interference is solved.

The invention discloses a radar and deep learning-based road surface internal disease three-dimensional morphology identification method, which integrates a novel technical path of qualitative identification and quantitative reconstruction, provides a high-precision and quantifiable reference for subsequent non-damage assessment by constructing a disease three-dimensional geometric model and reversely calibrating radar map characteristic parameters, and improves the accuracy and the accuracy of the three-dimensional morphology identification of a road surface internal disease. The problem that the three-dimensional morphology characteristics of the internal diseases of the asphalt pavement are difficult to accurately obtain at present is solved, and a calibration method is provided for obtaining the three-dimensional morphology characteristics of the internal diseases of the pavement based on a ground penetrating radar technology. And meanwhile, geometric data support is provided for mechanical response analysis and performance attenuation law research of the existing asphalt pavement structure containing internal diseases.

The invention discloses a highly-integrated miniature intelligent fuse module and a preparation process thereof, and the process comprises the steps: carrying out the high-frequency electromagnetic disturbance detection of a fusing element and a temperature sensor interface region, obtaining an electromagnetic response fingerprint, and determining an interface defect suspected point distribution diagram; micro-pulse thermal excitation is applied to the fuse element, a thermal pulse propagation process and sensor response are captured, and an interface thermal resistance distribution diagram is generated; carrying out acoustic microscopic analysis on the thermal resistance abnormal region to generate an interface defect fine classification graph; performing an electric-thermal transient response collaborative test on the fuse module to obtain a performance parameter test result; and performing targeted accelerated aging treatment on different types of interface defect samples, periodically detecting in the aging process, and determining a quality grade and a reliability prediction result. According to the method, the defects of the interface area of the intelligent fuse module are recognized and represented, the incidence relation between defect characteristics and function influences is established, the long-term reliability of products is effectively evaluated, and the quality control level of the products is remarkably improved.

The invention discloses an auxiliary tool, ground penetrating radar equipment, a lightweight tunnel detection vehicle and a control method, and relates to the technical field of tunnel detection.The auxiliary tool is used for assisting in detecting hidden tunnel diseases and comprises a base, a ground penetrating radar assembly and an obstacle avoidance assembly, and the base comprises a bottom plate and two side plates; the two side plates are connected to the two opposite sides of the bottom plate. The ground penetrating radar is arranged on the bottom plate and located between the two side plates, and the surface, away from the bottom plate, of the ground penetrating radar is exposed out of the edges, away from the bottom plate, of the two side plates. The obstacle avoidance assembly is arranged on the surface, away from the ground penetrating radar assembly, of the side plate and used for assisting the ground penetrating radar assembly in avoiding irregular parts of the inner wall of the to-be-detected tunnel. According to the auxiliary tool provided by the invention, collision between the auxiliary tool and irregular parts of the inner wall of the tunnel can be prevented, and the situation that ground penetrating radar equipment and tunnel ancillary facilities are damaged is reduced.

The invention discloses a polyethylene pipe microwave defect detection system based on a multi-dimensional movement mechanism. The method comprises the steps that firstly, full-coverage scanning of a welding area is achieved through a clamp and a control device, and microwave reflection signals of a welding joint are collected and subjected to amplification, filtering and normalization processing; converting the frequency domain signal into a time domain B-scan image by using inverse fast Fourier transform (IFFT), and effectively inhibiting direct waves and high-frequency noise by using a wavelet transform threshold denoising method so as to enhance defect reflection characteristics; on the basis, the system can identify and position typical defects such as holes, slag inclusion, incomplete fusion and the like, and visual visualization of defect positions and forms is realized through the information display module. The system is compact in structure and high in detection precision, has the advantages of high automation degree, wide application range, simplicity and convenience in operation and the like, can realize rapid and accurate detection of weld defects of the polyethylene pipeline, and provides reliable technical guarantee for pipeline quality evaluation and safe operation.

The invention discloses a method and a device for detecting the size of a road disease based on a ground penetrating radar. The method comprises the following steps: respectively acquiring relative dielectric constants of a surface layer, a base layer and a roadbed material of a target road; calculating a dielectric influence factor of the disease based on the position of the disease in the target road detected by the ground penetrating radar, the relative dielectric constant, and the operation parameter and the structure parameter of the ground penetrating radar; calculating a first critical dimension and a second critical dimension of the disease according to the dielectric influence factor of the disease and the structural parameters of the ground penetrating radar; and determining the size relationship between the horizontal size of the disease and the first critical size and the second critical size according to the change relationship between the reflected wave electric field intensity of the ground penetrating radar electromagnetic wave and the horizontal size of the disease. The pavement disease size is reversely deduced based on the electric field intensity of the ground penetrating radar electromagnetic wave reflected wave, data support is provided for formulating a reasonable and effective road maintenance and repair scheme, the road disease repair efficiency is improved, and safety and smoothness of road traffic are ensured.

The invention relates to the technical field of defect detection, in particular to a defect detection method and system based on a microwave frequency crystal resonator, and the method comprises the steps: forming a microwave probe through a microwave frequency crystal resonator array, placing a to-be-detected material below the microwave probe, and setting a scanning area and scanning step lengths in two directions on the surface of the material; controlling the probe to perform point-by-point scanning according to step length, and collecting a plurality of reflection coefficients of each sampling point to form an echo signal matrix; performing singular value decomposition on the matrix, applying an adjustable coefficient to a first singular value and a second singular value, and adaptively selecting an optimal coefficient according to imaging quality to reconstruct an echo signal; performing two-dimensional Fourier transform on the reconstructed matrix, constructing a spatial filter in combination with probe spacing and defect depth to perform phase compensation, performing two-dimensional Fourier transform to obtain a defect image, and performing threshold segmentation on the defect image to obtain a binary image only containing defects; and analyzing the connected regions to obtain the gravity center position of each defect and the defect area determined by the pixel number and the single pixel area. The problems of serious clutter interference, insufficient defect imaging contrast ratio, difficulty in quantitative evaluation of defect size and the like in the existing microwave detection can be solved.

The invention relates to the technical field of wafer pollution analysis, and discloses a wafer surface pollution analysis method, device and equipment, and the method comprises the steps: carrying out the multi-source detection data collection of the surface of a wafer through an ultraviolet module, a terahertz module and a microwave module, carrying out the information mapping of the multi-source data, constructing a pollution feature space, applying environment disturbance, and collecting disturbance data; according to the method, wafer surface pollution is identified, pollution distribution is verified, cleaning adaptability is analyzed, multi-dimensional cleaning strategy analysis is carried out based on pollution characteristic distribution, an optimal cleaning scheme is generated, and the method is beneficial for achieving accurate identification and dynamic evaluation of wafer surface pollution, improving adaptability and scientificity of the pollution cleaning scheme, optimizing the cleaning effect and reducing cost. The manufacturing yield and stability are enhanced, and the problem that the pollution distribution condition of the wafer surface is difficult to detect in a single identification mode in the prior art is solved.

The invention discloses a double-probe film thickness and defect collaborative decoupling method. The method comprises the steps that microwave reflection signals of a thermal barrier coating are synchronously collected through a coaxial resonance probe and a CSRR resonance ring probe; the microwave reflection signal comprises reflection coefficient amplitude, reflection coefficient phase and resonant frequency; constructing a frequency deviation model based on the microwave reflection signal; training the dual-channel deep learning network based on the frequency deviation model to obtain a trained dual-channel deep learning network; and obtaining film thickness information and a defect positioning result based on the trained dual-channel deep learning network. According to the method, collaborative decoupling of the film thickness and the defect signal is achieved through a physical-data fusion method, effective decoupling and independent recognition of the coating film thickness change and the defect response signal are achieved, and the resolution ratio, the accuracy rate and the practicability of a detection system are improved.

The invention provides a tunnel lining compactness defect detection method and system based on multi-modal data, and relates to the technical field of tunnel lining detection.The method comprises the steps that preliminary quality detection is conducted on a tunnel lining through a geological radar, and a compactness defect to be detected is obtained; acquiring multi-modal data of detected and to-be-detected compactness defects; performing feature extraction and coding processing on the multi-modal data to obtain a multi-modal embedded representation; carrying out heterogeneous graph modeling and cross-modal contrast learning based on multi-modal embedding representation to obtain a graph structure; performing joint processing on the graph structure based on a sparse attention mechanism and multi-modal fusion to obtain an embedded vector; and performing defect category classification on the embedded vector based on diffusion mapping to obtain a classification result of the to-be-detected compactness defect, the classification result including a concrete non-compactness defect, a filled block stone defect and a hole slag defect. According to the method, the problem that the specific type of the compactness defect is difficult to identify by an existing single-mode detection method is solved.