368 results about "Posture correction" patented technology

The invention provides a physical training posture correction method based on machine vision, which comprises the following steps: acquiring an athlete training image sequence through a multi-view image acquisition system and preprocessing the athlete training image sequence, extracting three-dimensional posture key points of a human body by utilizing a depth posture recognition model, constructing a skeleton model and calculating real-time kinematics characteristic parameters, and performing multi-dimensional comparison with standard parameters to generate a posture deviation evaluation report, thereby generating a personalized multi-modal correction feedback scheme, and establishing a personal athlete movement feature database to realize adaptive standard parameter optimization. According to the method, the training postures of athletes can be accurately recognized and corrected in real time, the training effect can be improved, sports injuries can be prevented, and training individuation and scientificity are improved.

The invention discloses a horizontal five-axis turning and milling composite center with a self-adaptive matching system. The horizontal five-axis turning and milling composite center comprises a tool turret, a boring tool mechanism, a tool changing area and a control system. A radio frequency identification reading device is arranged in a tool changing area of the tool turret and the boring tool mechanism, and an RFID chip tag is attached to the outer surface of each tool and used for storing tool serial numbers, specifications and length compensation parameters. And the control system realizes self-adaptive matching and sequential error correction of the cutter and the working procedure through radio frequency identification. The tool changing area is provided with a visual monitoring assembly which comprises a plurality of sets of industrial cameras, an adjustable light source and a structured light stripe illumination module and is used for detecting the clamping posture, the abrasion state and the type of a tool. The system has self-learning and digital twinning simulation functions, and can simulate the motion trail of the cutter and predict the interference risk before the cutter is changed. According to the method, integrated control over automatic tool recognition, posture correction, abrasion detection and path prediction is achieved, and the tool changing precision, the operation stability and the automation level of the horizontal five-axis turning and milling composite center are remarkably improved.

The invention relates to the technical field of automatic control, in particular to a detection equipment attitude real-time correction system which comprises an attitude trajectory extraction module, a trajectory stability evaluation module, an interference area recognition module, a light spot trajectory reconstruction module and an attitude correction control module. According to the invention, through comprehensive extraction of the direction change trend, the angle difference and the trajectory overlapping condition in the light spot image sequence, the trajectory trend of the attitude change between continuous frames is established, so that the identification of the attitude offset evolution process is more detailed, the attitude change stability analysis is realized in the time sequence dimension, and the accuracy of the attitude change is improved. Interference area identification is carried out by means of luminance continuity and spatial distribution consistency of light-emitting pixels, real light spots and lagging residual signals can be effectively distinguished, a light spot track path is reconstructed on the basis of interference removal, and real-time calculation of a track dynamic offset direction and an angle change amplitude is combined with track correction rule setting, so that the accuracy of track correction is improved. And the specific correction action matching of the deviation state is realized.

The invention belongs to the technical field of welding automation, and provides a welding gun posture self-adaptive adjusting method and system for a complex welding seam, and the technical scheme is that point cloud data of a to-be-welded welding seam is converted into a welding seam track, key features of the welding seam are extracted based on the welding seam track, and key points of the welding seam are determined; calculating the tangential direction and the normal direction of each welding seam key point; aiming at different types of welding seams, performing segmented path planning in combination with the key characteristics of the welding seams to obtain an ideal movement track of a welding gun, and determining the posture of the welding gun based on the ideal movement track of the welding gun in combination with the tangential direction and the normal direction of each welding seam key point; calculating the deviation between the welding gun and the welding seam according to the point cloud data collected in real time and the ideal track of the welding seam, generating a posture correction instruction according to a deviation calculation result, and adjusting the path and posture of the welding gun to ensure that the welding gun always keeps an optimal angle with the surface of the welding seam. And the welding quality and adaptability are improved.

The invention provides a muscular-bone rehabilitation posture correction method and system based on visual identification, and the method comprises the steps: firstly collecting the posture video data of the muscular-bone rehabilitation training of a rehabilitation object, and covering the joint movement, muscle contour change and body balance state change processes; analyzing the posture video data to generate joint movement tracks, muscle force application association and body balance state features, corresponding the joint movement tracks, the muscle force application association and the body balance state features to a preset muscle-bone rehabilitation stage target to establish a multi-dimensional adaptation relationship, and generating layered and staged correction guide information according to the multi-dimensional adaptation relationship, joint, muscle and balance adjustment layers are divided according to priorities, a specific adjustment mode is provided, new posture video data after adjustment is obtained, repeated operation is conducted, correction guide information is updated according to deviation changes, accurate and dynamic muscle-bone rehabilitation posture correction is achieved, and the rehabilitation effect and efficiency are improved.

The invention discloses an aviation transient electromagnetic pod coil attitude correction method and system, and belongs to the field of aviation geophysical exploration, and the system comprises a surface scanning industrial camera module which is used for obtaining the image information of a suspension coil in real time; the industrial-grade integrated navigation module is used for outputting inertial navigation data; the time synchronization and data acquisition module is used for establishing a time unification mechanism; the visual processing module is used for extracting the image processed by the time unification mechanism to obtain visual features; the inertial navigation resolving module is used for obtaining an inertial navigation state of a coil attitude angle through angular velocity integration; the data fusion and attitude estimation module is used for performing fusion calculation by combining the visual features and the inertial navigation state to obtain a fusion result; and the projection area and magnetic moment calculation module is used for calculating the ground projection area and the effective emission magnetic moment component of the coil according to the fusion result. According to the method, the system complexity and the electromagnetic interference risk caused by multi-inertial navigation layout are reduced, and the detection precision and stability of the aviation transient electromagnetic data are improved.

The invention belongs to the technical field of three-dimensional reconstruction, and particularly discloses a three-dimensional human body and scene interaction reconstruction method and system. The method is a three-dimensional human body and scene interactive reconstruction method for monocular video input, not only can quickly complete three-dimensional Gaussian reconstruction of a human body, a static scene and a moving object, but also can quickly complete the three-dimensional Gaussian reconstruction of the human body, the static scene and the moving object by introducing technical means such as posture correction, camera external parameter space regularization, joint reconstruction consistency loss and layered fusion rendering. And accurate alignment of the human body, the scene and the object in spatial positions, shielding relations and illumination styles is ensured, and dynamic interaction reconstruction conforming to real physical logic is realized. Finally, a joint rendering image with high fidelity, continuous time sequence and consistent structure can be generated, a stable and reliable three-dimensional expression capability is provided for complex human body actions, object interaction and scene understanding, and the three-dimensional interaction reconstruction quality and application value under the monocular video condition are greatly improved.

The invention relates to the technical field of tunnel safety monitoring, in particular to a strong noise tunnel point cloud monitoring method based on improved RANSAC and dynamic residual, which comprises the following steps: firstly, collecting two-dimensional point cloud data of a cross section of a tunnel, and performing attitude correction through ground baseline fitting to unify a coordinate system; afterwards, de-noising is carried out by adopting a vault-dominated improved RANSAC algorithm, and the algorithm effectively filters out strong noise and retains key lining points through constraint sampling, model geometric constraint and dynamically adjusted residual error threshold values; and performing nonlinear least square complete circle fitting on the denoised point cloud, and accurately extracting circle center and radius parameters. And finally, based on the absolute deviation, the relative deviation, the abrupt change increment and other multi-dimensional indexes of the radius, safety evaluation and early warning are carried out in combination with a standard threshold system. According to the invention, the robustness and precision of tunnel contour monitoring under complex working conditions of strong noise, equipment installation deviation and the like are obviously improved.

The invention discloses a GIS in-place robot coordination control method with a posture correction function, and belongs to the technical field of robot coordination control. The method comprises the following steps: establishing a 2D-3D coordinate mapping corresponding relation; fusing the corresponding relation with an operation state predicted by an IMU-based inertial measurement and system dynamics model; performing error calculation on the obtained 6D pose information and target pose information; constructing a nonlinear kinetic equation of the robot, and converting a control target into an optimization problem of various constraints; and converting the obtained control instruction into a torque signal for actually driving the robot to move. According to the method, the influence of mismatching on pose estimation is effectively inhibited by establishing the 2D-3D coordinate mapping relation; through a joint torque PID control algorithm, the torque required by each joint is calculated in real time, and high-precision tracking of an expected trajectory is realized; an MPC optimization control framework with multiple constraints is constructed, and precise, smooth and safe coordination control over the GIS in-place process is achieved.

The invention relates to a wearable fitness monitoring system that uses advanced sensor technology to track real-time muscle engagement, posture, and fatigue levels during physical activity. The system comprises a capacitive sensor system, incorporating elastic silver fiber cloth and medical-grade silicone, for precise muscle movement detection; a motion sensor for tracking movement along multiple axes; and an AI-powered processing unit to provide ergonomic feedback, prevent injuries, and optimize performance. It features dynamic self-calibrating sensors that adjust to user-specific movements using machine learning algorithms. The system offers personalized workout plans, adjusting in real-time based on biometric feedback, environmental conditions, and muscle fatigue. Additionally, it integrates Bluetooth Low Energy (BLE) for wireless communication, haptic feedback for posture correction, and AR / VR for immersive workout guidance. The system includes cloud-based data storage with encryption and supports blockchain for secure data management, offering a comprehensive, personalized fitness experience that improves workout efficiency and safety.

The invention relates to the field of three-dimensional reconstruction, and provides a pig weight estimation method, device and equipment, a medium and a computer program product, and the method comprises the steps: obtaining multi-view image data of a to-be-detected pig; generating sparse point cloud data based on the multi-view image data; generating a multi-view rendering image based on the sparse point cloud data and the three-dimensional Gaussian data; the three-dimensional Gaussian data is obtained by converting the sparse point cloud data; and inputting the three-dimensional Gaussian data into a weight estimation model to obtain pig feature information, and estimating the weight of the to-be-detected pig based on the pig feature information. According to the method, the accuracy of non-contact weight estimation of the pigs is improved and the implementation cost is reduced through an individual weighing posture correction, weighing scene three-dimensional reconstruction and posture semantic fusion associated weight estimation method.

The invention relates to the technical field of tunnel monitoring, in particular to a construction tunnel laser scanning point cloud attitude correction method based on multi-feature fusion, and the method comprises the steps: S1, synchronously extracting a ground baseline feature, a haunch symmetry axis feature and a vault circle center feature from a tunnel section laser scanning point cloud; s2, constructing confidence quantitative models for the three types of features respectively; s3, calculating the weight of each feature through a normalized exponential function based on the confidence of the three types of geometric features obtained in the step S2; setting a confidence threshold value, and forcibly returning to zero weight to shield failure features when the confidence threshold value is lower than the threshold value; when all the weights return to zero, activating a degradation protection mechanism to give equal weights; and finally, weighting and fusing the horizontal rotation angles of the three types of features according to weights, and outputting a correction angle. According to the method, the problem of systematic deviation caused by interference of a single reference is solved, and the correction precision and robustness in a complex construction environment are remarkably improved.

The invention relates to the technical field of pipeline deformation monitoring, and discloses a buried steel pipe deformation degree detection method and system based on multi-sensor fusion. The method comprises the steps that data of an accelerometer and a magnetometer are collected and processed to obtain an initial space orientation angle estimated value, the initial space orientation angle estimated value is compared with a preset steel pipe axis model, and if deviation exceeds a threshold value, dynamic deviation compensation is conducted through gyroscope data to obtain a real-time posture correction vector; extracting a multi-point detection sequence fitting deformation curve to obtain a continuity parameter set, and screening a low-consistency paragraph fusion neural network to obtain a multi-point deformation data set; and determining a final deformation evaluation value through underground interference simulation, noise modeling and particle adjustment, matching the final deformation evaluation value with historical records, and generating an alarm and risk early warning report when the matching degree is low. The method can improve the detection precision and data consistency, and guarantees the safe operation of the buried steel pipe.

The invention discloses a construction method and system of a three-layer inclined quadrangular pyramid grid structure, and relates to the technical field of building steel structure construction. The method comprises the steps of building an intelligent monitoring control system, carrying out BIM collaborative assembly and calibration, carrying out multi-dimensional graded lifting, carrying out adaptive attitude correction, carrying out BIM collaborative in-place recording and the like. The corresponding system comprises an intelligent monitoring control module, a BIM collaborative management module, a hydraulic synchronous lifting execution module, a posture correction module and an unloading in-place module. Through deep integration of a distributed control network and a BIM technology, dynamic monitoring, self-adaptive adjustment and visual management of the construction process can be realized, the accuracy, safety and efficiency of construction of the three-layer inclined quadrangular pyramid grid structure are improved, and the method is suitable for large-span steel structure grid installation engineering.

The utility model discloses a badminton racket for swing correction, which belongs to the technical field of sports auxiliary appliances and comprises a racket rod component, an auxiliary racket surface movably mounted at the upper end of the racket rod component and a front racket surface rotatably mounted on the auxiliary racket surface, and symmetrical return springs are connected between the auxiliary racket surface and the front racket surface. The racket rod assembly comprises a rod body, a grip rotatably installed at the lower end of the rod body and a handheld groove formed in the grip, the auxiliary racket face is movably installed at the upper end of the rod body, the handheld groove can be used for correcting and training the racket holding posture of a player, and the racket can be adjusted by adjusting the racket face and the grip, so that the racket can be used for correcting and training the racket holding posture. The device is suitable for correcting the racket holding posture of the left hand or the right hand, so that the applicability of the device is improved; the rotatable front racket surface is matched with the auxiliary racket surface, so that the device can be used for correcting the racket swinging strength and the racket swinging action of a training player, and the training efficiency of the player is improved.

The invention discloses a shoulder joint rehabilitation action image recognition and real-time posture correction feedback method and system, and belongs to the technical field of medical image processing and computer vision, and the method comprises the steps: collecting a rehabilitation training image frame sequence through an RGB camera; inputting the image into a human body analysis model, extracting skeleton key point coordinates and calculating a joint angle sequence; constructing the skeleton key points into a space-time diagram structure, and inputting the space-time diagram structure into a space-time diagram convolutional network for action classification and recognition; aligning and comparing the actual joint angle with the standard action template by adopting a dynamic time warping algorithm to calculate a deviation vector and an action compliance score; a standard posture skeleton line and an angle deviation indicating arrow are overlaid and rendered on an image, the compliance degree of each joint is displayed in a color gradient mode, an action compliance degree score and personalized correction suggestions are output, the error posture recognition accuracy is larger than 92%, the feedback delay is smaller than 100 ms, and the technical problem that home rehabilitation lacks professional action guidance is solved.

The invention relates to the technical field of foot health and motion monitoring, in particular to a multi-motion-scene-oriented foot motion data acquisition and intelligent analysis guidance method, which comprises the following steps of: acquiring foot three-dimensional static form and plantar pressure data through a foot correction shaping box, and establishing a foot geometric model with an anatomical mark; then, a nine-axis inertial measurement unit and a pressure sensor at a shoelace knob in the shoe are used for collecting time sequence foot motion data of multiple motion scenes; extracting multi-modal features after fusion, inputting the multi-modal features into a pre-training multi-task model, and outputting exercise habit classification, technical action scoring and injury risk assessment; finally, a sports biomechanics knowledge base is adjusted, and personalized suggestions of posture correction, muscle strength training and equipment adjustment are generated. The method is suitable for various exercise scenes, accurate in data acquisition, comprehensive in analysis dimension and high in guidance suggestion pertinence, the exercise normalization of the user can be effectively improved, and the exercise injury risk can be reduced.

The invention relates to the technical field of image processing, in particular to an identity recognition method and system based on image processing. When identity recognition is carried out, image illumination compensation operation is carried out on a collected image through a closed-loop process of semantic perception, type classification and mode-based compensation fusion, the type classification is divided into a low-light scene and an overexposure scene, and for the low-light scene, the overexposure scene is divided into an overexposure scene and an overexposure scene; for an overexposure scene, performing image enhancement processing on the pre-screened face image by adopting a face region distribution guided Retinex algorithm, and for the overexposure scene, pertinently adjusting cliplimit parameters of a CLAHE algorithm by sensing gradient features of a highlight region and a key texture region to realize image enhancement processing on the pre-screened face image; according to the invention, the method achieves the precise adaptation of the face structure, retains the illumination restoration of key details, solves the global defects of an existing method, and lays a high-quality image foundation for the subsequent shielding restoration and posture correction.

The invention provides an astronaut physical exercise process intelligent guide feedback method and system, and relates to the technical field of data feedback, and the method comprises the steps: obtaining three-dimensional motion track data in an astronaut training process, and carrying out kinematics analysis calculation to obtain a joint angle sequence and limb space posture parameters; the method comprises the following steps: obtaining a deviation vector by comparing and analyzing with a standard action template, generating a posture correction instruction and feeding back in real time, and regulating and controlling load parameters of training equipment, thereby realizing accurate guidance and intelligent correction in the astronaut training process, and improving the training effect and safety.

The invention discloses a motion posture correction system and method based on AI drive, and belongs to the technical field of artificial intelligence and motion health, and the motion posture correction system comprises a multi-mode perception layer, an intelligent preprocessing layer, an AI decision layer, a multi-dimensional feedback layer and a user platform; the multi-modal sensing layer fuses vision, biosensing and environmental data; the intelligent preprocessing layer improves the data quality through a cross-modal calibration and enhancement technology; the AI decision-making layer combines transfer learning and biomechanical analysis to realize personalized judgment; the multi-dimensional feedback layer provides graded intervention and risk early warning; the user platform supports data management and model evolution. According to the AI driving-based motion posture correction system and method, scientific and efficient motion posture correction is realized through multi-modal perception, personalized decision and accurate feedback.

The invention discloses an attitude control method and system for continuously locking a ground target by an unmanned aerial vehicle, and the method comprises the steps: obtaining a target position and an unmanned aerial vehicle attitude data stream, carrying out the coordinate transformation to generate a target attitude relation identifier, and carrying out the deviation vector decomposition to form a visual axis deviation feature set; disassembling the image into a fast and slow response chain through attitude mapping processing, repairing a fracture node, constructing an attitude correction track, extracting a control feature mode, generating a response feature factor, and constructing an attitude compensation constraint matrix; frequency domain decomposition and cross coupling calculation are carried out on the optical axis deviation feature set to determine the coupling correlation degree, and a superposition disturbance sequence is identified and dynamically reconstructed to generate a compensation gain sequence; executing stability evaluation to determine time sequence distribution, identifying abnormal disturbance characteristics and extracting key moment control characteristics; the constraint matrix and the gain sequence are fused to generate a fusion control law, a dynamic priority control coefficient is determined, a target continuous locking attitude control instruction is output, and continuous and stable locking of the moving target is achieved.

The invention discloses a gait analysis and posture correction intelligent insole suitable for high-weight crowds and a preparation method of the gait analysis and posture correction intelligent insole. The intelligent insole comprises a composite functional layer, an electrode layer and a packaging layer, and the composite functional layer comprises a linear elastic gasket and a sensing material arranged in an opening area of the linear elastic gasket. The sensing material adopts a wedge-shaped structure design, and the effective contact area between the sensing material and the electrode can be linearly changed along with the compression displacement under the constraint action of the linear elastic washer, so that the high-linearity electrical response of a pressure signal is realized. And the adjacent sensing units are physically isolated by the linear elastic washers, so that the crosstalk phenomenon in the pressure sensing array is effectively reduced. Through linear regulation and control of the linear elastic gasket on mechanical response and cooperative modulation of the wedge-shaped structure sensing material on electrical response, stable and high-linearity plantar pressure detection is achieved under the condition of high plantar load, and the device is suitable for gait analysis and posture correction of high-weight people in the states of standing, walking, running and the like.

The utility model belongs to posture correction tool technical field discloses a kind of supporting mechanism and the posture learning desk board comprising supporting mechanism;Supporting mechanism includes support leg, limit component, unlocking piece and positioning plate, limit component includes elastic member, and elastic member two ends respectively with positioning plate and limit component abut, and limit component is close to support leg;Supporting mechanism is also provided with hinged shaft on support leg;After supporting mechanism is opened, limit component moves under the action of elastic member, and it is locked and positioned by clamping support leg;Unlocking piece is set on limit component, and limit component removes locking state by reverse movement after unlocking piece is actuated.The utility model's support leg will not easily rotate in use process, and stability is good, and practicality is strong.Posture learning desk board, including supporting mechanism and panel main body, supporting mechanism is set on the back of panel main body.Supporting mechanism is flexibly matched on panel main body, and while realizing even distribution of pressure, the balance of panel main body can be maintained.

The invention relates to the technical field of medical care training, in particular to a real-time image recognition and posture correction feedback system for nursing operation standardization training, which comprises a nursing operation specification description module, an operation image recognition model module and an operation posture correction feedback module, the technical core of the method is that a differential geometry theory and a computer vision technology are deeply fused, a posture representation architecture based on Riemannian manifold is constructed, posture deviation is quantified through geodesic distance on the architecture, a posture evolution trajectory is predicted based on manifold dynamics, a system adopts a multi-level posture representation structure, and the posture representation architecture is a multi-level posture representation architecture. Comprising a key point layer, a posture layer and a semantic layer, and deep understanding of nursing operation postures is achieved. According to the medical care training system, personalized real-time correction guidance is provided for an operator through regionalized deviation analysis and a self-adaptive learning mechanism, the standardization and learning efficiency of nursing operation are remarkably improved, and a brand new intelligent solution is provided for medical care training.

1. Name of the product in this design: Posture Correction and Eye Protection Desk Lamp Study Board. 2. Purpose of this design: A writing pad or study mat. 3. The key design features of this product are the combination of shape and pattern. 4. The image or photograph that best illustrates the design's key points: 3D view 1.

The invention relates to the technical field of intelligent sitting posture correction, in particular to an intelligent humpback-preventing table mat based on multi-mode sensing. A main control unit, a forearm area pressure sensing module, a front edge center infrared distance measuring module, an electrochromic layer, a voice prompt device, a vibration motor and a rechargeable lithium battery are integrated in the body; according to the intelligent humpback-preventing table mat based on multi-mode sensing, standard sitting posture parameters of a user are recorded in a learning mode, bad sitting postures are judged by integrating hand pressure signals with three conditions including the vertical height, the horizontal distance and the body proportion in a monitoring mode, and progressive reminding from vision / touch to voice is triggered. Bad sitting postures such as collapsed sitting, head and neck protraction and overall forward leaning can be comprehensively recognized, missed judgment is avoided, the reminding mode is private and adapts to multiple scenes, different users can be adapted, early warning can be performed to help the formation of sitting posture habits, the integration degree is high, storage is easy, and the applicability and the user experience are greatly improved.

The utility model discloses a multifunctional integrated eye-protecting and posture-correcting table which comprises a bearing plate and a multifunctional integrated part, and the bearing plate is used for providing support for articles; the multifunctional integration part is arranged at the end part and / or the side part of the bearing plate, a mounting cavity for mounting at least one functional part is formed, and an opening is formed in at least one side of the mounting cavity, so that the functional part and the bearing plate are connected into a whole through the multifunctional integration part. The upper end face of the bearing plate serves as a writing face, notebook paper and the like can be placed on the upper end face of the bearing plate to facilitate writing of a user, the main function of the multifunctional integration part is that functional components and the bearing plate are connected together, so that the bearing plate integrates more functions in a learning scene instead of being only used for reading or writing, and all-around eye protection and posture correction are truly achieved; and the use flexibility and convenience are improved.

The application discloses a one-way valve component missing loading detection system and method, and relates to the technical field of visual detection.The system comprises an image acquisition and positioning end, an angle and posture correction end and a hierarchical comparison and detection end.The image acquisition and positioning end acquires a component area image of a to-be-detected one-way valve, and preliminarily positions the image with a pre-stored qualified product image.The angle and posture correction end corrects the to-be-detected image to the same detection posture as the template image by calculating the rotational deviation of the to-be-detected image and the template image and the affine transformation technology.The hierarchical comparison and detection end divides the corrected image into a plurality of component corresponding areas, calculates the feature difference of each area, and judges whether there is missing loading.The application can eliminate light interference through image quality adaptive enhancement, overcome position deviation through posture correction, realize multi-component parallel detection through hierarchical comparison and detection, improve the precision and robustness of missing loading detection, and support online switching detection of different models of products.

1. The name of the design product: posture correction belt. 2. The use of the design product: the design product is used for posture correction. 3. The design points of the design product: in shape. 4. The picture or photo that best shows the design points: front view. 5. The design product is a thin product, so the left view is omitted; the design product is a thin product, so the right view is omitted; the design product is a thin product, so the top view is omitted; the design product is a thin product, so the bottom view is omitted.

The invention relates to the field and discloses an online visual detection method for surface defects of door and window profiles. Comprising the following steps: acquiring model information of a door and window profile to be detected and calling a corresponding section template; in the continuous conveying process of the sectional materials, a plurality of line-scan digital cameras are used for collecting multi-view surface images at the same longitudinal position under alternate illumination of line light sources in different incident directions; real-time outer contour boundary and notch edge positions are extracted, and attitude correction parameters are obtained after registration with the section template; directly mapping the multi-view surface image into a strip expansion graph of a corresponding surface area according to the section anchor point line and the surface boundary line; and performing pseudo defect elimination based on a coordinate corresponding relation of the candidate abnormal areas under different illumination and an abnormal response direction, and performing continuity verification in combination with a longitudinal position coordinate so as to determine a real surface defect and output a defect level and a physical position.