510 results about "Distortion correction" patented technology

The invention relates to the technical field of artificial intelligence and computer vision, discloses a subway key component fault detection method and system based on an AI visual large model, and aims to solve the problems of low detection precision, weak generalization ability, insufficient multi-mode understanding, poor real-time performance and lack of state evolution modeling in the prior art. The method comprises the following steps: acquiring images of key components through a multi-view industrial camera array, and performing distortion correction, illumination normalization and noise suppression; a pre-trained visual large model is utilized to extract deep space features, and modeling is carried out on a continuous frame feature sequence through bidirectional LSTM to capture a time sequence change trend. By introducing the large-scale visual large model and spatio-temporal joint modeling, the identification capability of tiny defects is improved, the discrimination stability is enhanced, high-precision and low-delay automatic detection is realized, the false alarm rate and the omission ratio are remarkably reduced, and the detection efficiency and the system maintainability are improved.

The invention discloses a railway comprehensive defect detection method and system based on multi-source sensor fusion. The method comprises the following steps: carrying out joint calibration on static external parameters by adopting a multi-modal calibration plate; a high-precision timestamp and vehicle multi-source motion parameters are utilized to construct a motion compensation model to carry out adaptive dynamic distortion correction calibration; synchronously acquiring multi-source sensor data, and based on static external parameters and adaptive dynamic distortion correction, obtaining a corrected visible light image, an infrared image and a laser radar three-dimensional point cloud; complementarity features are extracted from the data of the three modes, adaptive fusion and optimization are carried out through an attention mechanism, and multi-scale fusion features are output; segmenting different railway facility areas by using multi-scale fusion features, extracting corresponding area features, and calling detection algorithms of corresponding facilities to identify defects of different railway facilities; according to the method, high-precision calibration, feature-level fusion and railway facility special defect detection are fused, and the automation level and the detection accuracy of railway inspection are improved.

The invention provides an optical and acoustic image registration method for underwater structure crack detection, and the method comprises the steps: collecting a multi-mode image through the vision field and time synchronization, building a one-to-one correspondence relation according to the pose and time metadata, and improving the image quality through the denoising, distortion correction and other means; optical and acoustic multi-scale features are respectively extracted through a dual-channel network, and cross-modal semantic alignment is realized in combination with a shared weight and a channel attention mechanism; generating a high-resolution scale offset field by using local cross-correlation and micro-upsampling, and introducing manifold regularization constraint to ensure that a vector field is smooth and continuous; according to the method, the local scale field and the affine parameters are combined, non-rigid space mapping is achieved through thin-plate spline interpolation, feedback iterative optimization based on edge structure consistency is assisted, the registration precision of a key structure is enhanced, the automatic registration effect of the underwater multi-modal image is improved, and the method has high robustness and practical application value.

The invention relates to the technical field of inductor winding monitoring, and discloses an inductor winding equipment abnormity intelligent monitoring system based on image processing. The system comprises an industrial image acquisition module, a real-time feature processing module, a defect identification module, a dynamic decision-making module and a data tracing module. The industrial image acquisition module acquires a multi-angle dynamic image through a high-frame-rate industrial camera and completes distortion correction and timestamp marking; the real-time feature processing module extracts the edge gradient, surface texture and geometric morphology features of the wire turn, and integrates the features into a multi-dimensional feature vector. And the defect identification module analyzes the feature vector through a pre-trained deep convolutional network, and outputs a form anomaly probability, a spacing deviation index and a surface defect level. The dynamic decision-making module generates a judgment result according to the judgment result and triggers grading alarm if the threshold value is exceeded; and the data tracing module is used for storing time sequence associated data and supporting abnormal process backtracking. The system realizes real-time accurate monitoring of winding abnormity, and helps to improve inductor production quality and flow stability.

The invention discloses a video transmission image stitching data enhancement method based on deep learning, and relates to the field of video image processing. The method comprises the following steps: S1, acquiring and screening images; s2, continuously screening structures and selecting key frames; s3, correcting image distortion; s4, splicing and fusing the images; s5, performing image enhancement output; firstly, a sliding time window mechanism is adopted, multi-dimensional image quality screening is combined, fuzzy, underexposure or severely-shielded inferior frames are accurately removed, and high quality of input key frames is ensured; through intelligent splicing and enhancement of a dynamic adaptive threshold strategy and semantic guidance, the image splicing precision and efficiency of the unmanned aerial vehicle and the multi-view camera in a complex environment are remarkably improved; besides, semantic segmentation guided feature extraction is combined with a multi-band fusion technology, seamless splicing is realized, the quality of an output image is improved, and the reliability of automatic analysis and decision making is remarkably improved.

The invention discloses an optical lens coating defect detection method based on image detection, and relates to the technical field of image processing, and the method comprises the steps: collecting a multi-illumination multi-band reflection image of a lens, carrying out dark white field and distortion correction to form a correction image, carrying out the segmentation and pose solving of the correction image, then carrying out the curved surface expansion remapping, and carrying out the image processing. A reflection observation sequence is formed according to channel stacking, a judgment threshold value is calculated based on the white field image and the dark field image, and pixel positions meeting the judgment threshold value are marked as low credibility and converged to generate a quality mask; coating reflection representation fitting is carried out on the reflection observation sequence, binary weighting is carried out according to a quality mask, range and continuity checking is carried out on a fitting result to obtain a credible mark, and an ideal reflection response is synthesized according to the credible mark. According to the method, ideal reflection response and residual error construction are performed on the reflection observation sequence, so that separation characterization of imaging disturbance and real coating abnormality is realized.

The invention relates to a visual guidance robot automatic control method, system and device and a medium. The method comprises the following steps: firstly, acquiring a target image sequence and extracting initial visual features, and processing the initial visual features through a deep learning algorithm to output a high-resolution image; calculating a distance switching value based on the high-resolution image, and if the distance switching value exceeds a preset threshold value, performing scale distortion correction on the high-resolution image to obtain a target image; multiple frames of time sequence data are fused through a Kalman filtering algorithm, image distortion real-time compensation and feature point track stability judgment are synchronously completed, and a target positioning precision estimated value is output; and finally, according to the target positioning precision estimated value, determining the position of a grabbing point through a posture mapping rule of mapping preset grabbing feature points in a target coordinate system to a robot-based coordinate system, and based on the position of the grabbing point, optimizing and fusing track parameters through a Kalman filtering algorithm to generate adaptive track parameters. The method improves the positioning precision and track following effect of automatic control of the robot.

The invention relates to the technical field of wave-absorbing material quality detection, and discloses a pyramid wave-absorbing material coating quality online detection system based on machine vision, which comprises a conveying unit, a vision acquisition unit, an image preprocessing unit, a coating quality analysis unit, a result output unit and a feedback control unit. The coating quality analysis unit is internally provided with a pyramid surface image geometric correction core model, a coating thickness quantitative calculation core model and a coating uniformity and defect judgment core model, cooperation is achieved through a feedforward correction and feedback optimization linkage mechanism, and the visual collection unit achieves full-view coverage through three sets of industrial cameras. The image preprocessing unit reduces distortion correction errors through conical surface constraint iterative correction; the thickness calculation model is combined with quadratic polynomial fitting and temperature compensation; the defect discrimination model fuses multiple parameters and comprehensive deviation values, and accurately recognizes three types of defects. According to the system, production parameters can be dynamically adjusted, the detection real-time performance and the quality traceability are both considered, and the large-scale production quality control requirement is met.

The invention provides a method and a device for mapping image coordinates to desktop coordinates, equipment and a storage medium, which are used for improving the positioning precision of a projection touch system in a real environment. The method comprises the steps of performing distortion correction on original gesture video data according to internal parameters of a camera and a lens distortion coefficient to obtain a corrected image frame sequence, and performing interaction space calibration according to the corrected image frame sequence to obtain a distortionless interaction plane, performing perspective change matrix solving on the distortionless interaction plane according to the target vertex coordinate set to obtain orthogonal mapping data, and when gesture key points are detected in the distortionless interaction plane, performing perspective transformation and normalization processing on gesture coordinates of the gesture key points according to the orthogonal mapping data to generate initial desktop control coordinates; and performing display mapping processing on the initial desktop control coordinates according to the physical attribute parameters of the target display to obtain desktop control coordinates.

The invention relates to the technical field of computer vision and industrial automation, and particularly discloses a MARK point high-precision PCB positioning and mechanical mapping method based on multi-coordinate system fusion, which is composed of PCB positioning and mechanical mapping in a template stage and PCB positioning and mechanical mapping in a production stage. Wherein the PCB positioning and mechanical mapping in the template stage comprises the steps of coordinate system establishment and image coordinate distortion correction; the PCB positioning and mechanical mapping in the production stage comprises Mark point real-time detection and offset calculation, real-time attitude solution and dynamic matrix calculation, and mechanical coordinate calculation of any PCB point. According to the method, a three-coordinate system collaborative modeling mechanism of an image coordinate system, a PCB coordinate system and a mechanical coordinate system is introduced, so that board attitude change and equipment fixing errors are separated, and the system robustness and the mapping precision are greatly improved; rigid body affine transformation is constructed by adopting double Mark points, the rotation angle, the proportion and the translation amount are solved in real time through two-point geometric constraint, and self-adaptive compensation of the posture change of the PCB is achieved.

The invention discloses a wavefront distortion correction method based on a dynamic adaptive MESSA algorithm and a coherent free space optical communication system. The problems that an existing method is prone to falling into a local optimal solution, the correction effect is poor under dynamic turbulence, and the adaptive capacity is weak are solved. According to the method, an individual position is initialized through a Circle chaotic mapping method, iterative optimization is carried out in combination with a Levy flight strategy and a variable spiral search strategy, and a step size control parameter, a safety threshold, spiral strength and a spiral adjustment coefficient are dynamically adjusted through a double-discrimination dynamic adaptive parameter adjustor. And reducing the population scale according to a nonlinear population reduction strategy, finally outputting an optimal individual position, and driving the deformable mirror to perform wavefront correction according to a control voltage vector corresponding to the optimal individual position. According to the invention, the dynamic adjustment of the algorithm parameters and the population structure is realized, and the adaptability, convergence speed and stability of the wavefront distortion correction method in the dynamic turbulence environment are enhanced.

The invention discloses a method for automatically correcting radial distortion of a wide-angle lens, and relates to the technical field of computer vision and image processing, and the method comprises the steps: collecting an original image, calculating the distance from a pixel to an imaging principal point, and extracting a multi-scale image and edge features; analyzing the edge direction and radial change and linear extension of the structure, extracting a linear structure candidate region, obtaining a distortion compensation parameter, generating a multi-scale candidate linear parameter set, and calculating a multi-scale linear consistency index for adaptive adjustment; extracting a point set from the candidate straight line parameter set, executing forward and reverse mapping to calculate a dual-space consistency index, and optimizing local parameters and the point set; calculating a minimum linear complexity index based on the line segment curvature, the curvature change rate and the length, and implementing local optimization; and performing weighted fusion on each index to construct a comprehensive objective function, performing step-by-step shrinkage optimization to obtain an optimal compensation parameter, performing global geometric correction on an original image, generating a distortion correction image, and realizing improvement of the structure recognition precision and the geometric correction effect.

The invention discloses a high-flux zebra fish behavior analysis device and method, and aims to solve the problem of compatibility of plate edge distortion and a multi-type pore plate, a telescopic mechanism is arranged in a shell of the device, an isolation cavity is formed in the shell when the telescopic mechanism is retracted, a micro-lens array and a light filter plate are sequentially arranged in the isolation cavity from top to bottom, and the micro-lens array and the light filter plate are arranged in the shell. A stimulation module is arranged at the bottom of the light filter plate, a pore plate placement area is arranged between the stimulation module and the light filter plate, and a universal adaptive bracket for accommodating a standard pore plate and a non-standard pore plate is arranged in the pore plate placement area; a camera module comprising at least one high-frame-rate camera with a fixed focal length is correspondingly arranged above the porous plate placing area, the camera module is connected with a control and analysis module, distortionless track diagrams, heat diagrams and behavior parameters are output online, and the optical mechanical integrated distortion correction and porous plate combined universal adaptation structure is used for correcting the distortion of the porous plate. When the same set of light path is switched between standard plates and non-standard plates of different models, the distortion of the edge view field is automatically compensated.

The invention relates to the technical field of image processing, and discloses a laser wavefront distortion correction system and method based on adaptive optics, and the system comprises a first module which is used for calculating a noise covariance matrix and an interaction matrix; the second module is used for solving a control matrix and a phase reconstruction matrix; the third module is used for calculating to obtain a current slope vector; the fourth module is used for updating a deformable mirror command; the fifth module is used for calculating to obtain a point spread function; and the sixth module is used for calculating an image quality index and outputting a correction result. According to the method, a noise covariance matrix and an interaction matrix are accurately constructed under reference light, and a weighted least square method with double regularization is combined, so that the influence of measurement noise is effectively inhibited, the stroke constraint of an actuator is balanced, and the robustness of a correction model is greatly improved; by means of combination of integral control, leakage processing and amplitude limiting design, saturation of an actuator due to command accumulation is avoided, and correction error superposition is prevented.

The invention discloses a wind turbine generator full-blade load monitoring method based on tower wall vision measurement, and the method comprises the steps: S1, arranging an observation platform, and installing a camera; s2, arranging a blade mark group; s3, synchronizing image acquisition and time; s4, camera calibration and distortion correction; s5, feature detection and tracking; s6, performing three-dimensional reconstruction; s7, centrifugal force and gravity compensation calculation; s8, blade net displacement calculation; and S9, carrying out aerodynamic force load inversion and result output. According to the invention, monitoring of the load in the edge direction of the whole blade of the wind turbine generator can be effectively realized, and reliable technical support is provided for safe operation, optimal design and economical efficiency improvement of the blade of the large-scale wind turbine generator.

The invention provides a nameplate printing defect detection method and system based on image segmentation. The method comprises the following steps: acquiring a nameplate surface image and carrying out distortion correction and brightness homogenization preprocessing; generating a pixel-level segmentation mask by using the semantic segmentation network; template registration is carried out based on the segmentation result, and a deformation field vector and a registration difference chart are calculated; performing defect area extraction and type identification on the difference image; according to the method, deep learning semantic segmentation is adopted to realize pixel-level analysis of a nameplate printing area, deformation field registration is combined to accurately quantify overprinting offset, defect quantitative evaluation is realized through multi-dimensional scoring, the detection rate reaches 99% or above, and the detection time of a single piece is less than 0.5 second.

The invention provides a power grid equipment monitoring method and system based on multi-source satellite power grid image data processing, and the method comprises the steps: firstly obtaining heat radiation temperature data and contour data of equipment from an infrared satellite and a visible light satellite, and carrying out the time-space synchronization processing; secondly, extracting temperature gradient distribution information from the heat radiation temperature data, and identifying and sequencing local overheating areas in combination with radiation intensity difference; enhancing equipment contour edge identification according to the sequence to obtain an enhanced contour image; superposing the data with thermal radiation temperature data, and extracting comprehensive distribution characteristics to determine an initial fusion weight; identifying a temperature distribution imbalance region, evaluating the reliability of the two types of data sources, and adjusting the weight in combination with the operation state of the equipment to obtain a final ratio; the two types of data are weighted and fused according to the proportion, and a final temperature distribution image is obtained through temperature field distortion correction; and extracting the geographic position of the overheat area and mapping the geographic position to the equipment layout diagram to complete power grid equipment thermal anomaly monitoring. Through the scheme of the invention, the thermal anomaly of the power grid equipment can be monitored.

The invention relates to the technical field of image processing and crack recognition, in particular to a cooling tower crack intelligent recognition system based on an unmanned aerial vehicle inspection image, which comprises an image acquisition module, a self-adaptive distortion correction module, a local texture enhancement module, a crack feature focusing module and a geometric morphology recognition and positioning module. Wherein the image acquisition module is used for acquiring an original image of the cooling tower; the self-adaptive distortion correction module is used for carrying out nonlinear geometric correction; the local texture enhancing module is used for enhancing the texture in the crack direction and inhibiting concrete background noise; the crack feature focusing module is used for extracting a pixel-level crack candidate region; and the geometrical morphology identification and positioning module outputs a crack data set with spatial position information. According to the method, through combination of image geometric correction, texture enhancement and three-dimensional space mapping, accurate extraction, spatial positioning and actual width measurement and calculation of the cooling tower cracks are realized, and the accuracy and reliability of structural defect identification are improved.

The invention belongs to the technical field of optical system aberration analysis, and discloses an experimental device for measuring and correcting image distortion of an optical imaging system, which comprises an optical axis testing device and an optical distortion testing platform, and the optical distortion test platform is used for testing optical distortion of the optical imaging system to be tested and integrating optical distortion data with optical axis deviation to obtain an image distortion fitting coefficient and a distortion correction model of the optical imaging system. According to the invention, the image distortion of the optical imaging device can be measured, the image distortion can be corrected, the imaging device can correct the deviation between the optical axis and the center of the target surface when the system is installed and adjusted, the optical distortion of the whole optical system can be corrected, and the method has guiding significance for the application of detecting and identifying the accurate position of a target by using the optical system.

The invention relates to the technical field of gem detection and three-dimensional modeling, in particular to a gem high-precision three-dimensional modeling and flaw detection method based on reverse engineering, which comprises the following steps: S1, fixing a gem to be modeled, and acquiring image data of different wavelengths by using a multispectral imaging device; s2, performing distortion correction, noise suppression and light intensity normalization processing on the multispectral image data; s3, based on the preprocessed multispectral image data, refractive index inversion is carried out in combination with a gem optical model and light propagation characteristics, and point cloud data containing space coordinates and refractive index information is generated. According to the method, the gemstone is subjected to multispectral imaging acquisition, and image preprocessing, refractive index inversion, point cloud processing and self-adaptive grid reconstruction are combined, so that high-precision three-dimensional modeling and defect marking are realized, and the problems that traditional gemstone detection mostly depends on manual observation, and due to the lack of multispectral information, the gemstone detection accuracy is low are solved. Therefore, the problem that internal defects cannot be accurately identified is solved.

The invention relates to the technical field of plate automatic production, in particular to a plate directional two-dimensional edge positioning method, and aims to solve the problems that a camera shooting and simple edge detection method cannot adapt to high-speed online measurement, image distortion influences precision, complex image textures and colors and the like, and edge positioning robustness is poor. According to the technical scheme, the method comprises the following steps: acquiring an original image of a moving plate and an image of a calibration plate through an image acquisition unit; based on a pre-generated calibration file, performing global distortion correction on the original plate image through a coordinate calibration module, establishing a mapping relation between image sub-pixel coordinates and physical coordinates, and obtaining a corrected plate image; and processing the corrected plate image, extracting a plate area through an edge positioning module, taking at least two edges of the plate in different directions as target edges, and performing directional two-dimensional edge positioning on the target edges in at least one direction under shadow interference.

The invention discloses a mapping and polling positioning method and system based on a laser radar inertial odometer, and the method comprises the steps: carrying out the distortion correction of current point cloud data based on inertial data, and obtaining a point set after distortion removal and a prediction global point set; adaptively selecting a target voxel level and a target neighborhood scale of the predicted global points to obtain a neighborhood point set of each predicted global point, and constructing a point-to-plane residual vector based on the neighborhood point set; based on the predicted pose, the covariance and the residual vector, error state iteration updating is carried out through iterative extended Kalman filtering to obtain a posteriori state; local map updating is carried out based on the posterior state; and obtaining a corrected transformation matrix based on the key frame pose, and obtaining a corrected release pose based on the corrected transformation matrix. According to the method, self-adaptive adjustment of voxels and neighborhoods can be realized, data pollution is effectively prevented, and smooth and global consistency of a closed-loop track is realized under the constraint of a non-re-optimization state and a non-re-mapping map.

The invention discloses a ship spherical global image splicing method and system based on image processing, and the method comprises the following steps: obtaining multi-camera images which are distributed at the top, the bottom and the annular direction, and executing camera calibration and distortion correction; performing hemispherical projection and cylindrical projection by using a hybrid projection link; feature point extraction and matching are carried out, and an initial registration relation between adjacent view fields is established; executing APAP gridding local registration of projection perception, and generating local homography by applying an anisotropic weight kernel consistent with the region projection direction to obtain a partition alignment result set; carrying out weighted fusion and stitching suppression on the overlapped region to generate a panoramic spliced image; and converting the panoramic spliced image into an interactive three-dimensional panoramic view. According to the method, mixed projection and projection perception APAP algorithms are fused, ship spherical surface global high-precision splicing and three-dimensional immersion display are achieved, and the method has the advantages of being small in distortion, high in precision and high in visibility.

The invention discloses a DAC output error correction method based on digital compensation, and the method comprises the steps: collecting the amplitude deviation, phase deviation and time sequence jitter data of a DAC output signal through a quantum sensing DAC calibration analysis platform, and constructing a multi-dimensional error original data set; calling an operational amplifier imbalance dynamic coupling prediction model to analyze a coupling relationship, and extracting dynamic association features; separating harmonic components and screening error contribution factors based on a broadband DAC harmonic distortion correction model; calculating a gradient change rule and an extreme point position by adopting a differential nonlinear gradient correction model; and integrating the characteristics and the parameters to construct a digital compensation parameter matrix, and dynamically adjusting an output code value through real-time digital modulation to realize error correction. According to the method, dynamic association and distribution characteristics of multi-source errors are accurately captured through multi-model collaborative operation and full-process closed-loop processing, multi-dimensional error synchronous correction is achieved, and DAC output precision and working condition adaptability are improved.

The invention discloses a vehicle identity accurate recognition method and system based on multi-modal image fusion, and particularly relates to the field of vehicle identity recognizing.The method comprises the steps that images are collected through visible light and infrared sensors which are synchronously calibrated, and preprocessing is conducted through adaptive filtering, CLAHE enhancement and distortion correction; a feature pyramid network and a channel attention mechanism are utilized to carry out adaptive fusion of multi-modal features, an improved residual convolutional network is combined to extract key features of a vehicle contour, a license plate, a vehicle logo and a vehicle window, a mixed matching algorithm is adopted, a cosine similarity and a weighted Euclidean distance are synthesized to calculate a feature matching degree, and the feature matching degree is calculated. Retrieval is carried out through a local-cloud distributed database; and when the fusion similarity exceeds a preset threshold value, outputting identity information such as a vehicle license plate, a brand model, a registration year and a color.

The invention discloses a single-frame unmanned aerial vehicle image pixel positioning method and system based on an elevation map, and relates to the technical field of unmanned aerial vehicle image processing and geographic positioning, and the method comprises the steps: obtaining a single-frame original image shot by an unmanned aerial vehicle, unmanned aerial vehicle attitude information during shooting, camera internal and external parameters, and a digital elevation map of a shooting region; according to the internal reference of the camera, converting the pixel coordinate of the target pixel in the original image into a normalized plane coordinate system of the camera, and carrying out coordinate distortion correction to obtain the back projection point coordinate of the target pixel; converting a camera-unmanned aerial vehicle body coordinate system and a body-geographic inertial reference coordinate system according to camera external parameters and unmanned aerial vehicle attitude information to obtain normalized ray direction vectors corresponding to target pixels; an earth surface height constraint is constructed based on a digital elevation map, and an intersection point of a ray and the ground is determined by performing iterative approximation on a space point along a ray direction. According to the invention, rapid and accurate conversion from the single-frame unmanned aerial vehicle image pixel to the geographic coordinate can be realized.

The invention discloses a multi-view fisheye camera real-time look-around image splicing method and device, and relates to the technical field of image processing, the method generates a lookup table (LUT) based on calibration parameters, converts a fisheye image distortion removal correction process into a table lookup operation, avoids real-time calculation one by one pixel, significantly reduces the operation overhead, and improves the image quality. And the system can keep low-delay processing under the condition of high-resolution input. According to the method, self-adaptive feathering weight distribution is introduced into a seam band, local energy matching and a multi-band fusion method are combined, and the width and the weight of a splicing area are dynamically adjusted according to scene brightness and contrast difference, so that ghosting and brightness abrupt change at the seam are effectively reduced, and the smoothness of splicing transition is ensured.

The invention discloses an industrial part surface defect intelligent detection system and method, and relates to the technical field of defect detection. When a part is fixed on a rotating carrier for full-angle scanning, a surrounding type industrial camera and a light source array cooperate, a light field is dynamically evaluated by combining part curved surface reflection characteristics, and the light source intensity and direction and camera exposure and gain are adjusted; obtaining a uniform initial image; a global image of the part is reconstructed through brightness standardization, denoising interference suppression, multi-scale texture enhancement, distortion correction, normal expansion and cross-view alignment splicing; the method comprises the following steps: extracting cross-scale features by adopting a graph convolution attention model, constructing a fusion feature vector based on geometric morphology parameters and texture features by virtue of defect regions segmented by regional reconstruction aggregation curved surface reflection to finish classification and grading, generating a simulated graph through an adversarial generative network, comparing and verifying, and outputting a detection report. The defect detection precision and efficiency are effectively improved, and the industrial high-quality detection requirement is met.

The embodiment of the invention discloses a hydroelectric slope deformation monitoring method, a hydroelectric slope deformation monitoring system and electronic equipment. The deformation monitoring method comprises the following steps: acquiring an optical image stream, a radar image sequence and a meteorological data set; performing feature extraction on the optical image, and obtaining a target displacement sequence; performing lens inclination angle distortion correction based on the target displacement sequence to obtain optical correction data; the method comprises the following steps: acquiring a permanent scatterer deformation sequence by adopting an SBAS-InSAR technology based on a radar image sequence, generating a global deformation graph sequence by adopting a spatial interpolation algorithm, carrying out space-time alignment on a target displacement correction sequence and a meteorological data set with the global deformation graph sequence, and carrying out atmospheric delay correction based on an obtained time distribution matrix. Calibrating the radar data based on the optical data to obtain corrected radar deformation data; and hydroelectric slope deformation monitoring information is output by fusing the optical displacement data and the radar deformation data. According to the invention, the precision, timeliness and environment adaptability of deformation monitoring are enhanced.

The invention discloses a non-contact screening system for motion coordination based on computer vision, which relates to the field of computer vision, and comprises the following steps: acquiring a human body motion image through a non-contact camera, extracting human body key point coordinates frame by frame by using a deep learning algorithm after noise reduction and distortion correction, and constructing a time sequence data set; performing smooth jitter removal and interpolation processing to form standardized data; calculating track, rhythm and stability multi-dimensional motion coordination quantitative characteristics based on the data, and performing comparative analysis with a personal historical database to identify a motion function degradation trend; and based on age groups and scenes, customizing screening parameters, coordinating workflows of the components, and realizing full-process automatic screening. The method has the advantages that invasive and cross infection risks are avoided through non-contact acquisition, accurate and objective assessment of the exercise coordination is realized by means of deep learning and time series data analysis, exercise function degradation can be early warned, and an efficient and reliable scheme is provided for various exercise coordination screening scenes.