277 results about "Industrial inspection" patented technology

The invention discloses an industrial inspection intelligent decision-making method and system based on large and small model collaboration. The industrial inspection intelligent decision-making method comprises the following steps: dividing an inspection task into a plurality of sub-tasks by using a cloud large model, and performing dynamic task planning based on the sub-tasks; calling an edge small model to execute the subtask to obtain multi-modal original data; performing cross-modal feature fusion on the multi-modal original data by using a cloud large model to obtain cross-modal fusion features; and reasoning based on the cross-modal fusion features by using a cloud large model, extracting abnormal factors, and formulating a new inspection task based on analysis of the abnormal factors. Through the comprehensive method of comprehensive multi-mode perception, conditional diffusion model enhancement and cloud-edge collaborative decision, the problems of fault sample scarcity and perception deviation in an extreme environment are effectively solved, high precision and low time delay are considered, and a new breakthrough of intelligent inspection is brought.

The invention discloses a visual algorithm self-training method based on multi-agent collaborative optimization, and the method comprises the following steps: constructing a multi-agent system architecture which comprises a user interaction layer, an intelligent scheduling layer, an A2A protocol communication layer and a professional agent cluster layer; the user interaction layer analyzes a user task intention and generates an execution plan; the scheduling agent calls the professional agent to complete data processing, model construction, training, testing and deployment; a task process is coordinated through a standardized communication mechanism, and task execution is supported by combining an MCP tool set, a knowledge base module and a memory system; and when the task fails, automatically executing rescheduling operation, and finally outputting a self-training result. According to the method, the development efficiency, the self-adaptability and the intelligent level are remarkably improved, and the method is suitable for computer vision tasks such as industrial detection, intelligent security and protection and automatic driving.

The invention discloses a fan metal surface defect detection method and device based on lightweight YOLO11 and a medium, and relates to the technical field of computer vision and industrial detection. The method comprises the following steps: acquiring fan metal surface defect image data, and preprocessing to obtain a training data set; yOLO11n is used as a basic model, and a lightweight StarNet adopting a four-level layered architecture and a star operation feature fusion mechanism is used as a model backbone network; combining a bottleneck module with a multi-scale convolution block, and constructing a neck network by applying a global heterogeneous kernel selection mechanism and an efficient up-sampling module; a heavy parameterized detail enhanced convolution and group normalization GN strategy is used to construct a detail enhanced lightweight shared convolution detection head; and sequentially connecting the backbone network, the neck network and the output layer of the detection head to form the lightweight metal surface defect detection model. On the premise that the detection efficiency is guaranteed, high-precision identification of metal weld defects can be achieved.

The invention relates to an automatic image segmentation technology based on a convolutional neural network, and is suitable for the field of medical image and industrial detection. In order to solve the problems of rigid feature fusion, insufficient context capture, low efficiency of boundary optimization and poor small target segmentation precision in the existing method, an adaptive multi-scale feature fusion network is constructed: an encoder adopts a progressive expansion strategy and gated attention to intensify cross-scale features; the decoder optimizes hierarchical feature contribution through a dynamic weighted fusion module; the end-to-end boundary optimization is realized by integrating the lightweight differentiable CRF; and designing a composite loss function balance category weight. The segmentation recall rate of the fine structure is obviously improved by more than 18%, the boundary sawtooth rate is reduced by 41%, and the calculation efficiency is improved by 76%.

The invention discloses an aluminum profile surface defect detection method based on a DMSA-Swindow-Unet network model, and relates to the technical field of industrial detection, and the method comprises the steps: carrying out the preprocessing of a collected aluminum profile surface image; dividing the image into image blocks by using a convolution operation with a predetermined step length, and performing linear projection to generate an embedded feature sequence; the features are input into a Swin Unet encoder module to extract multi-layer local and global features, and an alternate window attention and shift window attention mechanism is adopted by a multi-layer nested Swin Transform module; in the decoding process, coding features of corresponding scales are fused through jump connection; a DMSA module is configured after the jump connection and is used for enhancing the context information sensing capability after fusion; and step-by-step up-sampling is carried out through image block expansion, and a pixel-level defect segmentation image consistent with an original image in size is output. The method is high in detection precision, strong in robustness, good in adaptability and strong in deployability, and is suitable for automatic identification and intelligent quality inspection scenes of surface defects of the aluminum profile on an industrial site.

The invention belongs to the technical field of industrial robot inspection, and discloses an instrument identification method for industrial robot inspection, an instrument identification system comprises a pose adjustment module, a lightweight detection module, an image double enhancement processing module and a double-branch instrument reading identification module, and the instrument identification method comprises the following steps: obtaining an instrument image, inputting the instrument image into an image dual-enhancement processing module to obtain a preprocessed instrument image; the preprocessed instrument image is input into a lightweight detection module, an instrument target detection result is output, the instrument type is judged according to the result, and an instrument reading result is output and recorded. According to the method, the improved lightweight target detection model and the adaptive image optimization processing algorithm are combined, the real-time performance and the high efficiency of instrument reading in an industrial scene are effectively improved, more intelligent and accurate instrument reading and industrial inspection can be provided, and powerful support is provided for information acquisition in a complex industrial environment.

The invention discloses a point cloud registration method based on a fast point pair feature attention module, and the method comprises the steps: firstly collecting point cloud data, and then carrying out the matching of point clouds through a point cloud registration model. The point cloud registration model is based on an encoder-decoder architecture, a fast point pair feature attention module FPFA, an Inception deep convolution module IDCM, a lightweight cross-frame attention aggregation module LCAAM and a matching matrix calculation module TSMM based on two stages are fused, and a final point cloud registration result is obtained. According to verification of a contrast experiment and an ablation experiment on a public data set and a self-built workpiece data set, the method is excellent in accuracy, speed and robustness, and especially has remarkable advantages in low-overlapping-rate scenes and industrial workpiece registration; an efficient solution is provided for factory assembly line automation and mechanical arm accurate grabbing, and the method has important application value for industrial detection scenes with limited computing resources.

The invention discloses a time stretching sweep frequency interference high-precision distance measurement method and device based on double auxiliary interferometers, and relates to the technical field of non-cooperative target absolute distance measurement. In order to solve the technical defects that in the prior art, due to initial wavelength fluctuation of a light source and frequency sweeping nonlinearity, drift of a ranging result is large, and interference signals are interfered by zero-frequency noise, the technical scheme provided by the invention comprises the following steps: acquiring linear frequency modulation optical signals output by a common-source frequency sweeping laser; the signals are respectively transmitted to a measurement interferometer, an auxiliary interferometer and a resampling auxiliary interferometer to form a homologous three-interference structure; a dispersive medium is introduced through the reference arm, so that the reference light generates time stretching and interferes with the measurement light to form a chirp beat frequency signal; k-domain resampling and cross-correlation processing are carried out on the beat frequency signals, and target distance information is extracted. The method is suitable for high-precision and wide-range non-cooperative target absolute distance measurement work, and is especially suitable for the fields of laser radar three-dimensional modeling, precision industrial detection, robot calibration and the like.

The invention relates to the technical field of machine vision industrial detection, and discloses an all-welded heat exchanger industrial detection method and system based on machine vision, and the method comprises the steps: obtaining a welding line region image of a to-be-detected heat exchanger, and extracting a boundary contour of a welding line region through an image preprocessing algorithm; calculating a welding line center line according to the boundary contour of the welding line area, and dividing the welding line center line into a plurality of detection units with equal length; in each detection unit, linear light scanning data of the surface of the welding seam and ultrasonic echo signals of a welding seam area are collected; generating a three-dimensional shape model of the surface of the welding seam according to the linear light scanning data, and extracting acoustic characteristic parameters in the welding seam; preprocessing the three-dimensional shape model and the acoustic characteristic parameters, and inputting the preprocessed three-dimensional shape model and acoustic characteristic parameters into a pre-trained defect evaluation model to obtain a comprehensive defect index of the welding seam area; and according to a comparison result of the comprehensive defect index and a specified threshold group, judging whether obvious defects exist in the welding seam area, and completing comprehensive evaluation of the welding seam quality.

The invention relates to the technical field of mixed reality and artificial intelligence fusion application, in particular to a mixed reality intelligent inspection method and system based on multi-modal fusion. According to the technical scheme, the method comprises the following steps: synchronously acquiring a user view image and voice stream data through mixed reality equipment; performing target detection on the view image by adopting an improved YOLOv10 model, and outputting a candidate target area and a confidence score # imgabs0 # thereof; voice input is transferred through a Whisper voice recognition model, and a semantic vector # imgabs1 # is generated by using a BERT semantic embedding model. According to the invention, through a multi-mode cooperative sensing mechanism of fusing image, voice and space gazing information, in combination with context semantic optimization and lightweight terminal deployment, intelligent target identification and natural interaction with high precision, real-time performance and strong robustness are realized in a complex industrial inspection scene.

The invention discloses a steel surface defect detection method based on a YOLOv8 multi-scale convolution attention mechanism, and relates to a steel surface defect detection method. Firstly, a multi-scale convolution attention mechanism MSCA module is introduced into a backbone network, accurate expression of small target features is enhanced, and the detection precision is improved; besides, a brand-new deformable convolution C2fDCNv2 module is provided to replace an original C2f module, so that the capability of capturing complex shape and irregular target features by the model is enhanced, and the efficiency of feature extraction is further improved; and finally, replacing the original loss function with a Wise-IoU (WIoU) loss function, accurately measuring the similarity between the target frames, and improving the defect detection performance of the model. According to the model, an experiment is carried out on an NEU-DET data set, an experiment result is compared with different-scale models of YOLOv8, the steel defect detection precision is effectively improved, and the industrial detection requirement is better met.

The invention belongs to the technical field of computer vision, and discloses a target identification tracking method based on a self-supervision mechanism, and the method comprises the steps: enhancing a self-supervision pre-training module through causality, constructing a causal sample pair through unlabeled video data, learning universal features through combining with comparison loss, and achieving the high-precision tracking without large-scale manual labeling. A multi-modal feature fusion and dynamic calibration mechanism further reduces dependence on annotated data, is especially suitable for industrial inspection, field monitoring and other scenes where data acquisition is difficult, significantly reduces time and labor costs in a data preparation stage, and broadens the application range of the technology in resource limited scenes; a causal reasoning and physical constraint mechanism is introduced, a dynamic relation between targets is modeled through a space-time causal graph, unreasonable tracks are filtered in combination with a physical rule, and complex conditions such as shielding, rapid movement and extreme weather are effectively dealt with; the dynamic feature calibration module corrects feature drift in real time, and ensures stable model performance in long-term tracking.

The invention discloses a target tracking method and device based on a rolling and pitching type holder, and the method comprises the steps: building a polynomial response model through adaptive color calibration, and achieving the dynamic color correction; an improved YOLOv8 model is combined with a bidirectional feature pyramid network to carry out multi-scale target detection, and template matching is assisted to improve a low-confidence-coefficient scene recognition rate; predicting the position and the speed of the target by using Kalman filtering to cope with a shielding condition; and high-precision smooth rotation of the holder is realized through quaternion attitude solution in combination with PID (Proportion Integration Differentiation) and FOC (Fiber Operating Control) double-loop cooperative control. The device adopts a rolling and pitching type holder structure and comprises a main controller, an angle sensor, a visual sensor and a magnetic encoder, the mechanical structure is simplified, the weight is reduced, and higher response speed and higher control precision are achieved. The target tracking accuracy, stability and anti-interference capability of the intelligent mobile platform in a complex environment are remarkably improved, and the method is suitable for various dynamic scenes such as aerial photography and industrial inspection.

The invention discloses an inspection method of an industrial inspection system based on multi-modal fusion, which belongs to the technical field of industrial inspection, and comprises the following steps: S1, obtaining positioning data of a node to be positioned by using an HDS-TWR distance measurement algorithm and a Taylor full centroid algorithm, and outputting a positioning coordinate; s2, using a sensor to measure the acceleration and angular velocity of the inspection system so as to align the motion and posture for evaluation; s3, improving a Kalman filtering algorithm to enable an output prediction result to be close to a true value; s4, infrared-vision dual-mode cooperative tracking is adopted, and meanwhile, a time synchronization displacement compensation mechanism is utilized to calculate a relative displacement difference between two tracking modes, and dynamic calibration is carried out; s5, a YOLOv8 model is adopted to process video frames or pictures collected in the inspection process, a target box and category probability are output, and finally an inspection result is obtained; according to the inspection method of the industrial inspection system based on multi-modal fusion, a high-precision and low-delay automatic inspection solution is provided for severe environments such as mines and nuclear power stations.

The invention discloses a rapid quantitative differential interference microscopy system and method based on holographic recording. The system comprises a partially coherent illumination module, and a microscope objective, a tube lens, a transmission grating, a first linear polarizer, a first thin lens, a half-wave plate, a shading plate, a phase-type spatial light modulator, a second thin lens, a second linear polarizer and an image acquisition module which are sequentially arranged along the optical axis direction of the partially coherent illumination module. According to the invention, the phase-type spatial light modulator is utilized to realize shearing offset in any direction and at any degree, so that the device is suitable for different types of application occasions; the system realizes holographic recording of differential interference by utilizing grating carrier frequency, realizes quantitative differential interference detection on a to-be-detected sample only by utilizing two acquired images, has very high time resolution, spatial resolution and imaging precision, and has a wide application prospect in biological imaging and industrial detection.

The invention provides an imaging optical system for industrial visual inspection, which is characterized in that a first lens is mainly composed of nine lenses and has focal power, the object side surface of the first lens is a convex surface, and the image side surface of the first lens is a concave surface; the second lens has focal power; the third lens has focal power; the fourth lens has focal power; the fifth lens has positive focal power; the sixth lens has focal power, and the object side surface of the sixth lens is a convex surface; the seventh lens has focal power, and the object side surface of the seventh lens is a concave surface; the eighth lens has focal power; and the ninth lens has positive focal power, the image side surface of the ninth lens is a convex surface, and through the combination design of nine lenses and the reasonable configuration of the refractive power and the surface type of each lens, high-resolution imaging under a compact structure and a large aperture is realized, the aberration correction capability is particularly enhanced, and a relatively high MTF value is ensured in the whole zoom range. Accuracy and reliability of industrial detection are remarkably improved, and the system has the advantages of zooming, low distortion and low cost.

The invention discloses a mobile robot autonomous search navigation method based on a visual language large model. The method comprises the steps of receiving a natural language instruction of a user, and synchronously collecting an RGB image and a laser point cloud; two-stage visual language reasoning is adopted, and target semantic analysis and existence judgment are completed in sequence; when the target does not exist, autonomous exploration is executed based on the point cloud data and is parallel to target judgment, and an exploration-judgment closed loop is formed; after determining that the target exists, switching to a visual language navigation model, generating an incremental motion instruction, and guiding the robot to approach the target; target state monitoring is continuously carried out during navigation, if the target is lost, a dynamic backtracking mechanism is triggered, the robot is driven to return to the historical state of the recently confirmed target, and navigation is restarted. The method does not need to depend on a prior map, can execute efficient and robust autonomous search and navigation tasks on edge equipment through task collaboration, model quantification and state backtracking, and is suitable for complex scenes such as industrial inspection and emergency rescue.

The invention provides an industrial inspection-oriented defect autonomous decision-making method, device and equipment and a medium, and belongs to the technical field of intelligent manufacturing. The equipment defect autonomous decision-making method comprises the following steps: acquiring equipment operation state information of industrial equipment by using a multispectral fusion terminal, and uploading the equipment operation state information to a cloud server; a built-in cross-modal deep analysis model of the cloud server is used to carry out cross-modal learning on the equipment operation state information according to the double-flow heterogeneous deep network architecture, and joint distribution of visual-text features of the industrial equipment is learned to obtain defect feature vectors of the industrial equipment; performing multi-dimensional matching on the defect feature vectors and similar cases in a maintenance knowledge base, and screening treatment schemes corresponding to the defect feature vectors to obtain a maintenance strategy; and issuing the defect information corresponding to the defect feature vector and the maintenance strategy to a communication terminal of a maintainer. According to the method and the device, the problem of low cross-modal information utilization rate caused by multi-modal data splitting and no feature level fusion in the prior art can be solved.

The invention relates to the field of AI visual detection, in particular to an AI visual detection method based on supervised and unsupervised deep learning, which comprises the following steps: establishing a sampling image visual feature by using real-time sampling image data; establishing an AI visual feature analysis model based on supervised and unsupervised deep learning by using the visual features of the sampled image; according to the AI visual feature analysis model, an AI visual detection result is obtained, high-precision and high-robustness visual detection is achieved under a small amount of annotation data through a supervised and unsupervised mixed learning architecture in combination with the discrimination ability of deep learning and the anomaly detection advantage of unsupervised learning, a supervised main model is trained based on the annotation data, and a supervised auxiliary model is trained based on the unsupervised learning. Precise recognition of known defects is ensured, limitation of a traditional supervised model is broken through, generalization ability is improved while high precision is ensured, and the method is suitable for complex and changeable industrial detection scenes.

The invention provides a weld defect intelligent detection method based on high-definition image deep learning, which comprises the following steps: S1, collecting a high-definition weld image, manually marking the selected image, and constructing a weld defect image database; s2, performing first fine tuning on the pre-training weight GLIP-L by adopting a GLIP model; s3, a self-made welding test piece is made, a high-definition camera and ultrasonic detection equipment are used for collecting a weld joint surface high-definition image of the self-made welding test piece and ultrasonic echo characteristic data of internal defects, and a high-definition weld joint image and ultrasonic multi-mode database is established; s4, introducing a multi-layer perceptron module into the GLIP model, constructing an image and ultrasonic multi-modal feature alignment mechanism, and performing second fine tuning to obtain a trained GLIP model; and S5, performing defect detection on the welding seam by using the trained GLIP model, and outputting a detection result. According to the invention, through multi-modal data fusion, the accuracy and robustness of weld defect detection are improved, and the method is suitable for automatic and accurate industrial detection scenes.

The invention relates to the technical field of industrial detection maintenance and deep learning, in particular to a steel wire rope defect detection method with multi-dimensional interpretability, and the technical scheme comprises the steps of steel wire rope broken wire data set manufacturing, YOLOv8 model training and prediction and model evaluation and optimization. According to the method, the electromagnetic signal is converted into the image, the YOLOv8 deep learning model is used for training and predicting the broken wire of the steel wire rope, meanwhile, the multi-dimensional interpretability technology is used for carrying out interpretability analysis on the model from different angles, and indexes for quantifying the interpretability of the model are provided, so that the broken wire is more concrete, the identification degree is higher, and the method is suitable for popularization and application. The model detection speed is high, complex intermediate steps are not needed, model transparency can be improved, and the method plays an important role in guiding model optimization and increasing credibility.

The invention belongs to the technical field of industrial detection and reverse engineering, and particularly discloses a conduit detection method based on scanning point clouds, which comprises the following steps: a dynamic self-calibration optical system acquires an initial point cloud of a measuring tube; segmenting the initial point cloud to form a complete tube; a linear section and a bent section are distinguished through the inclination angle change of the adjacent cylinder axes, bending points are aligned through a bending point ICP algorithm, and a matrix conversion relation is obtained; correcting the bending point based on the relation and solving process data; restoring conduit characteristic data; and finally, the reconstruction tube and the model tube are aligned through a bending point ICP algorithm, and the measurement deviation is determined. The method solves the problems of low tracking precision caused by environmental interference, influence of point cloud noise on alignment, difficulty in digitalization of process data, difficulty in feature restoration and the like, is suitable for high-precision detection of complex conduits, and provides a quantitative basis for quality control and process optimization of industrial production.

The invention discloses a distributed unmanned aerial vehicle cluster cooperative control method and system, and relates to the related technical field of unmanned aerial vehicle control, and the method comprises the steps: carrying out the inspection task planning of an unmanned aerial vehicle cluster and a first robot cluster, and obtaining a cooperative plan; unmanned aerial vehicle pilot-operated macroscopic inspection is carried out based on collaborative planning, and a target detection area is obtained; robot following type microscopic detection is carried out based on collaborative planning, and a target detection result is obtained; and performing fan diagnosis by combining the target detection area and the target detection result. The technical problems that in the prior art, the wind power plant inspection efficiency is low, the fan equipment diagnosis precision is insufficient, the collaboration of an unmanned aerial vehicle and a robot is poor, and the industrial control automation degree is low are solved. The technical effects of improving the industrial inspection automation level, optimizing the fan equipment diagnosis accuracy, enhancing the multi-equipment cooperative control stability and improving the overall efficiency of industrial operation and maintenance are achieved.

The invention relates to the technical field of intelligent inspection, in particular to an industrial inspection robot applied to ultrasonic detection and a control method, and the industrial inspection robot comprises an automatic guide vehicle which is used for carrying out navigation advancing according to a preset inspection route; the sensor integration module is installed on the automatic guide vehicle in a lifting mode through the lifting mechanism and used for collecting data of equipment at different heights; the sensor control unit is used for generating optimal working parameters of the sensor according to the historical data and the environmental data acquired by the sensor integrated module and controlling the sensor integrated module to perform ultrasonic data acquisition according to the optimal working parameters of the sensor; working parameters of the ultrasonic sensor can be dynamically adjusted according to environmental data, and the accuracy and reliability of detection are improved.

The invention discloses a WebRTC voice enhancement system and method based on a multi-modal large model, and belongs to the field of artificial intelligence and real-time communication cross technology, and the system comprises an audio and video collection module which is used for synchronously collecting original voice signals and corresponding video image data of a user side through a WebRTC protocol stack, high-precision alignment is realized through timestamp marking and a cache mechanism; the multi-modal feature extraction module is used for extracting audio features from the voice signals, extracting visual lip movement features from the video images and generating text semantic features through a voice recognition engine; a noise matching and updating module; a multi-modal semantic perception enhancement module; an audio reconstruction module; and a WebRTC integration module. According to the invention, high-precision noise suppression and millisecond-level delay are realized while the voice semantic integrity is guaranteed, and the requirements of industrial inspection, vehicle-mounted communication and other scenes on high fidelity, low delay and strong robustness are met.

The invention discloses an intelligent oil seal defect detection method based on improved YOLOv12, and relates to the technical field of oil seal defect detection, and the method comprises the following steps: collecting annotated oil seal image data; a YOLOv12 model is improved, a head layer is adjusted to fuse shallow semantics, a small target detection layer is added, a nearest neighbor interpolation up-sampling module is replaced by a dynamic up-sampling module, a CIoU loss function is replaced by an MPDIOU loss function, and a C3k2 module in a backbone network of the YOLOv12 model is replaced by an MSAF module; the model is trained with the data set and the adjustment is evaluated. According to the intelligent detection method for the oil seal defects, the small target detection layer is added, the MSAF module is used, and dynamic upsampling and a new loss function are combined, so that the detection precision, real-time performance and robustness of surface defects such as burrs, notches and scratches of the oil seal are improved, and the small-size defect identification problem in industrial detection is effectively solved.

The invention discloses a cage stranded wire pitch measurement method and system based on a line-scan digital camera, and belongs to the technical field of industrial detection. The method adopts an image acquisition mode of cooperative work of an encoding wheel, an encoder and the line-scan digital camera, feeds back the motion speed of a stranded wire in real time through the encoder and accurately controls the line scanning frequency of the line-scan digital camera; the image acquisition rate and the stranded wire movement speed are strictly synchronized; intelligent calculation of the pitch of the stranded wire is realized based on dynamic gray analysis, and the method comprises the following steps: establishing a dynamic conversion model of pixel-physical size by integrating multi-dimensional information including the number of single stranded wires, view parameters of a linear array camera and image resolution; and dynamically calculating an optimal segmentation threshold of a local area by adopting a gray threshold segmentation algorithm and collecting gray distribution characteristics of a complete stranded wire image. According to the invention, non-contact pitch measurement with micron-order precision can be realized, the product quality stability is greatly improved, and key technical support is provided for industry intelligent upgrading.

The invention discloses a defect classification method and device, and belongs to the technical field of industrial detection. The defect classification method comprises the following steps: acquiring a feature similarity between a collected to-be-detected image corresponding to a to-be-detected object and each first prompt image in a prompt image library; based on the feature similarities, obtaining a target defect category corresponding to a target prompt image in the plurality of first prompt images; and determining the target defect category as the defect category of the to-be-detected object. The defect classification method can adapt to diversified and complicated defect modes, can reduce storage requirements and calculation time on the basis of keeping feature diversity, uses the prompt image library to perform defect classification, does not need to depend on a large amount of annotation data, consumes few calculation resources, is stable in performance, and is high in practicability. And the accuracy of defect classification is high.

The invention relates to the technical field of computer vision detection, and discloses a PCB defect detection method based on improved YOLOv8. According to the invention, by introducing the C2f module fused with the coordinate attention mechanism, the adaptive weighted bidirectional feature pyramid structure and the lightweight decoupling detection head, the feature extraction capability and the spatial positioning precision of the model for PCB micro defects are improved, and the problem of small target missing detection is solved; meanwhile, in combination with defect area specific enhancement training and a dynamic receptive field optimization strategy, the anti-interference capability of the model to a complex background is enhanced, and the false detection rate is greatly reduced; besides, by adopting a structure re-parameterization technology, the model reasoning speed is improved while the detection precision is ensured, so that the method can be efficiently deployed on industrial detection equipment with limited resources, the balance of high precision and high efficiency is realized, and the comprehensive performance of PCB defect detection is comprehensively improved.