1193 results about "Feature point matching" patented technology

The invention provides a battery film production automation control method and system based on visual inspection, and relates to the technical field of production control, and the method comprises the steps: collecting a film image, carrying out the fusion, generating a fused film image, carrying out the noise reduction, constructing a feature point matching relation, removing wrong point pairs, carrying out the geometric correction, carrying out the sub-pixel-level alignment, and obtaining a registration image; edge enhancement is carried out based on an improved Frangian filtering algorithm, an edge region is repaired and segmented to obtain a film edge contour, three-dimensional reconstruction is carried out, three-dimensional point cloud coordinates are solved, a film three-dimensional contour model is reconstructed, a film defect detection model is constructed, and a film defect distribution diagram and defect confidence are obtained through forward propagation; performing association analysis with a production process, constructing a feature mapping model, generating a defect knowledge base and matching a process parameter optimization strategy, generating a dynamic instruction and sending the dynamic instruction to a station control unit, monitoring operation data, performing closed-loop feedback control, and obtaining a current optimal control scheme.

Disclosed in the present invention are a digital image calculation method and system for deformable template-based RGB-D camera multi-view matching. The method comprises the six steps of data acquisition, point cloud data preprocessing, feature point matching, re-registration of deformable templates, calculation of a point cloud data transformation relationship between large-view images, and point cloud fusion. The size of a non-adjacent image matching template is adjusted on the basis of a registration result of adjacent views to achieve correct registration of feature points of non-adjacent view images, thereby improving the matching precision, eliminating the accumulative error of an image set, providing a relatively accurate initial value for subsequent iteration of point cloud fusion, reducing the number of iterations, and implementing three-dimensional reconstruction of images.

The invention discloses a close planting farmland growth vigor assessment method and system based on image processing, and relates to the field of agricultural information, and the method comprises the following steps: S1, multi-source data collection and preprocessing; s2, improving image segmentation, and extracting crop features; and S3, multi-dimensional growth vigor evaluation. According to the method, the field block level, the plant level and the whole growth period are covered through multi-source data collection, a generative adversarial network is used for repairing and shielding the plant image and restoring complete form information, the segmentation problem in a close planting scene is solved, the accuracy of close planting crop image analysis is improved, accurate registration of multi-modal data is achieved by means of feature point matching, and the accuracy of close planting crop image analysis is improved. The graph neural network optimizes image segmentation, effectively distinguishes overlapped leaves and stalks, deeply fuses multi-modal features and dynamically selects a fusion strategy, improves feature distinguishability, constructs a dynamic adaptive evaluation model, improves generalization ability and evaluation precision, identifies and intervenes abnormities in real time, and improves crop anti-risk ability and yield prediction accuracy.

The invention relates to the technical field of intelligent detection in the electronic manufacturing industry, in particular to a PCB warping degree intelligent evaluation system integrating visual positioning and multi-modal sensing, which comprises a multi-modal sensing unit, a visual positioning unit, an intelligent evaluation engine and a closed-loop control unit, the multi-modal sensing unit integrates laser displacement, infrared thermal imaging and strain sensors to acquire three-dimensional deformation, temperature and stress data; the visual positioning unit realizes sub-pixel-level positioning by using a high-resolution industrial camera and a feature point matching algorithm, and compensates vibration errors; the intelligent evaluation engine fuses data based on a time-space synchronization protocol, predicts a thermal deformation trend through an improved multi-modal convolutional neural network, and dynamically adjusts a qualified threshold value; and the closed-loop control unit executes sorting and rechecking according to an evaluation result, and optimizes warping and leveling parameters. According to the system, multi-dimensional accurate detection and intelligent control are realized, the PCB warping degree detection accuracy is effectively improved, the process can be dynamically optimized according to the production working condition, and the equipment fault risk is reduced.

The invention relates to the technical field of stamping part detection, and discloses a stamping part size and defect synchronous detection method and system. The method comprises the steps that the multi-sensor measurement module is used for collecting line laser scanning morphology data, infrared thermal image strain data and structured light projection contour data of a stamping part; multi-sensor data registration is achieved through a feature point matching algorithm, and a three-dimensional space coordinate mapping relation is generated; calculating the thermal expansion compensation amount of the material in combination with a thermal deformation correction model; filtering the structured light projection contour data, and extracting key contour feature points and defect region boundaries; inputting the related data into a multi-source data fusion model to obtain a dimensional deviation and defect fusion detection result; based on this, a measurement path is updated through a dynamic path planning algorithm, and a synchronous detection scheme is output. The device can synchronously detect the size and defect of the stamping part, improves the detection precision and efficiency, achieves the quality grade classification, and is high in adaptability.

The invention relates to a steel structure component virtual pre-assembly method based on three-dimensional laser scanning. Designing a self-adaptive hierarchical scanning strategy, and dynamically adjusting parameters of scanning equipment to obtain data of each region of the component; abnormal points are detected and eliminated by fusing scanning data of different levels; datum points are extracted, coordinate calibration is carried out, data of all levels are spliced, and a full-component three-dimensional point cloud model is formed; classifying the point clouds by using a multi-dimensional data analysis algorithm; establishing a closed-loop data verification system; designing a multi-feature point matching algorithm; performing iterative docking and position calibration through an algorithm; quantitative analysis is carried out on spatial deviation between the components in the assembling process; after the deviation is detected, a fine adjustment scheme is generated, and the position of the component is adjusted; a virtual pre-assembly path is generated according to the shapes, the connection modes and the installation sequence of the components, and the assembly sequence and the layout of spatial positions are controlled; and the assembly state of each component is monitored in real time, and the path is optimized and adjusted through a feedback mechanism.

Disclosed is a digital image calculation method and system for RGB-D camera multi-view matching based on a variable template, the method includes six steps: acquiring data, preprocessing point cloud data, performing feature point matching, re-registering a variable template, calculating point cloud data transformation relationships among large-view images, and performing point cloud fusion. A size of a non-adjacent image matching template is adjusted based on registration results of adjacent angles of view, and correct registration of feature points of images from non-adjacent angles of view is accordingly achieved, which improves matching accuracy, eliminates cumulative errors in image sets, and provides more accurate initial values for subsequent iterations of point cloud fusion, such that the number of iterations is reduced, and three-dimensional reconstruction of images is implemented.

The invention relates to a non-standard special-shaped part high-precision measurement and verification method based on three-dimensional vision, and belongs to the field of three-dimensional reconstruction and computer vision. Point cloud data of a non-standard special-shaped part are obtained through multi-angle scanning by means of multi-camera-projector collaborative layout in combination with a structured light coding technology and a phase shift code technology; the method comprises the following steps of: acquiring point cloud data, ensuring comprehensiveness and accuracy of the data, realizing high-precision alignment of the point cloud data through matching an underivative optimization algorithm with image feature points, and performing fusion, denoising and registration processing on the acquired point cloud data to ensure that point clouds acquired from a plurality of angles can be effectively fused so as to reconstruct a high-precision three-dimensional model. In the measurement and verification stage, the deviation between the point cloud data and the CAD model is measured through a standard coordinate system, multi-stage registration and a high-precision algorithm, rapid detection and size verification of the geometric deviation of the industrial workpiece are achieved, and reliable technical support is provided for quality control.

The invention provides a camera extrinsic parameter calibration method, an image stitching method and an apparatus thereof. The calibration method can include: acquiring a plurality of images captured by the camera based on a plurality of sets of extrinsic parameters to be calibrated, wherein each of the plurality of sets of extrinsic parameters to be calibrated include a horizontal rotation angle and a vertical rotation angle; obtaining at least one matched image pair based on feature point matching among the plurality of images, wherein each matched image pair includes matched feature point pairs with a quantity satisfying a threshold condition; and calibrating the plurality of sets of extrinsic parameters to be calibrated according to a first difference representation between coordinate representations of each matched feature point pair of each matched image pair in a reference coordinate system.

The invention discloses a power transmission image defect detection and defect duplicate removal method and system based on a deep learning image segmentation algorithm, and belongs to the technical field of intelligent inspection of power equipment. According to the method, real-time tower identification and adaptive shooting are realized through a lightweight YOLO model deployed at the edge end of an unmanned aerial vehicle; pixel-level segmentation is carried out on the infrared image by using an MSAN-Net network, the network integrates a ResNet encoder, a cross-scale attention mechanism and a multi-level feature pyramid, and boundary learning is enhanced by using a composite loss function; based on a multi-view three-dimensional reconstruction technology, two-dimensional defects are mapped into space rays through feature point matching and pose estimation, and defect de-weighting is achieved through ray intersection judgment. Through the MSAN-Net network, the segmentation precision of the infrared component under a complex background is remarkably improved through an attention mechanism and multi-scale feature fusion, and the problem of repeated defect detection in multi-view inspection is effectively solved in combination with a three-dimensional space mapping method.

The invention relates to a wharf sensing method and system based on data fusion of a laser radar and a camera, and the method comprises the steps: firstly employing a fisheye camera to shoot a checkerboard calibration plate at multiple angles, employing a Zhang's calibration method to calculate an internal reference matrix and a distortion coefficient of the fisheye camera, and synchronously collecting the data of the laser radar and the camera in real time; the method comprises the following steps: performing distortion removal processing on a camera image, calculating an external parameter matrix between a laser radar and a camera through a feature extraction algorithm, a feature point matching algorithm and a nonlinear optimization algorithm, and performing data fusion through an internal parameter matrix, the external parameter matrix and a bilinear interpolation method to generate fusion point cloud data with color information; and finally, identifying a key target of the wharf by marking and training a CNN model, and calculating three-dimensional parameters of the key target by adopting a specific calculation method, so that a sailor can quickly distinguish the target in the fused point cloud data, the accuracy and efficiency of wharf environment perception are remarkably improved, and accurate environment perception and decision support are provided for ship berthing and departing operation.

The invention provides a multi-source image registration method based on deep learning, and relates to the technical field of image processing. The method comprises the following steps: firstly, respectively carrying out image preprocessing on infrared and visible light images; then constructing and training a feature point extraction model, and respectively extracting feature points and descriptors from the preprocessed infrared and visible light images; carrying out feature matching on the extracted feature points by adopting a feature point matching model to obtain high-precision matching point pairs; robust matching point screening and registration optimization are carried out on the matching point pairs, and a geometric transformation matrix between the infrared image and the visible light image is calculated; and finally, performing geometric transformation on each pixel point of the infrared image through the geometric transformation matrix between the infrared image and the visible light image, and converting the pixel points into a coordinate system of the visible light image to obtain a final registration result of the infrared image and the visible light image. According to the method, when the modal difference between the multi-modal images is large, a stable registration effect can still be obtained, and the robustness of multi-modal image registration is greatly improved.

The invention provides a method, device and equipment for determining the needle insertion depth in a fundus needle insertion image and a medium, and the method comprises the steps: obtaining a target fundus needle insertion image, and determining a plurality of first key points of a surgical needle in the target fundus needle insertion image based on a FAST corner detection algorithm; determining a plurality of second feature points of the target fundus needle insertion image based on the plurality of first key points, a plurality of first feature points of the reference image and a feature point matching algorithm; based on the plurality of second feature points and the first feature points, carrying out perspective transformation on the plurality of second feature points to a plane where the reference image is located, and obtaining a plurality of third feature points of the target fundus needle insertion image after perspective transformation; and based on the plurality of third feature points and the first feature point, determining the needle insertion depth of the surgical needle in the target fundus needle insertion image. The fundus needle insertion images of different shooting visual angles are converted to the same shooting visual angle as the reference image, the proportion of image pixels to the actual distance is unified, and the accuracy of needle insertion depth measurement in the fundus needle insertion images is improved.

The invention discloses a robot vision inertial navigation method and device and a storage medium, and belongs to the technical field of mobile robot navigation, and the navigation method comprises the following steps: measuring motion information of a mobile robot by using an IMU sensor; collecting image data of the mobile robot by using a camera; inputting the feature points and the descriptors into a LightGlue network model to obtain a local feature point matching relation of the image; calibrating the offset of the gyroscope by optimizing and recovering visual scale information, and estimating the IMU coordinate system speed, the gravitational acceleration and the offset of the IMU; introducing a visual feature re-projection error, an IMU pre-integration error and marginalized residual constraint, and then integrally converting into a target function to obtain a nonlinear optimization result; further repositioning is carried out to obtain a repositioning result; and optimizing the global pose map to obtain a final global pose map optimization result, and carrying out navigation operation according to the global pose map optimization result. The method can improve the positioning precision and operation efficiency under view angle change and illumination change, and is high in practicality.

The invention provides an unmanned aerial vehicle real-time road condition monitoring method and system and a medium, the method relates to the field of image data processing, and the system comprises an unmanned aerial vehicle which comprises an image acquisition device and a sensor group and is used for acquiring aerial videos and flight parameters of the unmanned aerial vehicle in real time; the streaming media server is used for receiving and storing the aerial video transmitted by the unmanned aerial vehicle; analyzing the video stream into a frame-by-frame image sequence with aligned timestamps; performing target detection on each frame of image, and identifying a target vehicle by using a pre-trained deep learning model; carrying out target tracking on the target vehicle in the continuous frames based on an intersection-to-union ratio matching and trajectory prediction algorithm, and constructing a target motion trajectory; and calculating the displacement distance of the target vehicle in the image based on an optical flow method, feature point matching and a deep learning algorithm, and estimating the actual running speed of the vehicle in combination with the flight parameters of the unmanned aerial vehicle. According to the system, computer vision and sensor technologies are combined, and accurate speed measurement is realized through target detection, tracking and speed estimation.

The invention discloses a method for determining a preset position of a video monitoring camera. Camera presetting positions usually need to be manually set one by one, and time and labor are wasted. The method is used for a video monitoring system, and automatic setting of the preset position of the camera is realized by utilizing a group of scene images and the to-be-monitored target image, so that the efficiency is greatly improved. The method comprises the following steps: acquiring feature points based on a scene image set by adopting a computer vision technology, generating a three-dimensional model of a scene by adopting a motion recovery structure technology and a surface reconstruction algorithm, and obtaining the position of a to-be-monitored target in a three-dimensional space by matching the feature points of a to-be-monitored target image with the feature points of the scene image set. Therefore, a proper camera is selected, a corresponding preset position is calculated, and finally the preset position is corrected according to an actual monitoring effect.

The invention discloses a slope deformation monitoring method and system, a medium and equipment, and relates to the field of slope deformation monitoring, and the method comprises the steps: obtaining a slope image sequence with a fixed time interval; extracting features of the first image, tracking the features of the first image in the to-be-detected image by using a Lucas-Kanade optical flow method, estimating a view field translation amount and a rotation angle of the to-be-detected image relative to the first image, and performing view field motion correction on the to-be-detected image; performing semantic segmentation on the first image and the corrected to-be-detected image to obtain an image pair with a slope region mask; obtaining a feature point matching set of the image pair based on LoFTR; performing feature point mismatching elimination on the feature point matching set by using a two-stage mismatching elimination method to obtain a matching point pair set M [beta]; and calculating the displacement of each point pair in the M [beta], and obtaining a sub-pixel-level slope deformation field through Gaussian interpolation. The scheme is low in cost, small in calculation amount and high in precision.

The invention discloses a method and a system for automatically calibrating and correcting a preset position, and relates to the technical field of preset position calibration. In order to solve the defects that an existing preset bit correction technology is relatively large in correction error and cannot effectively process when an observation target deviates from an observation range, a preset bit reference image of which the definition accords with a preset confidence coefficient is obtained; acquiring a to-be-detected image of which the definition accords with a preset confidence coefficient; respectively utilizing a feature point matching algorithm process, a template matching algorithm process and / or a target detection method to calculate each offset; the weight coefficient of each offset is adjusted, and the final offset is calculated according to each offset and the corresponding weight coefficient; and according to the final offset, carrying out preset position correction. The method and the device are mainly used for automatically calibrating and rectifying the preset position.

The invention discloses an industrial image anomaly detection method and device, equipment and a storage medium, and relates to the field of image recognition. Obtaining a defect-free product graph, making a comparison data set based on the defect-free product graph, and performing pre-storage processing; performing feature point shape matching on the comparison data set based on the received to-be-detected image and the extracted feature point data, and determining a comparison image of the to-be-detected image; taking the to-be-detected image as a target object, correcting a product image area in the image, and aligning the product image area with the product image area of the to-be-detected image according to a pixel shift operation; and comparing the aligned product image areas, and identifying abnormal defects in the to-be-detected image. According to the scheme, the comparison data set of the defect-free product image is constructed, the feature point matching and pixel-level alignment technologies are combined, the problem of false detection caused by product position deviation or form change in a traditional method is effectively solved, the adaptability to illumination change is improved through brightness adjustment and difference matrix analysis, and the detection accuracy is improved. The method has the advantages of high detection precision, high environmental adaptability and high processing efficiency.

The invention relates to the technical field of UV printing, and discloses a packaging box UV printing method and device and computer device.The method comprises the steps that feature extraction is conducted on CAD model data of a packaging box, and a feature data set of the packaging box is obtained; carrying out coordinate mapping on the working space of the robot to obtain a conversion matrix and space calibration data of the mounting position of the UV nozzle; edge detection and feature point matching are conducted, and position coordinate data and posture angle data of the UV printing area are obtained; performing multi-section track planning on the ink jet track to obtain initial jet printing track control data; and performing synchronous control calculation on the UV light source to obtain a UV light source power modulation parameter and UV light source trigger time sequence data. According to the method, by dynamically adjusting the power output and the triggering time sequence of the UV light source, uniform curing of the ink droplets is achieved, the problem that curing of the ink droplets at different positions is not uniform is effectively solved, and the printing quality is improved.

The invention discloses an air-ground image matching method, device and equipment and a storage medium, and the method comprises the steps: carrying out the sparse reconstruction of an aerial image, obtaining a sparse point cloud model of an aerial image pose and a scene, generating the camera visual angle information of a new camera according to the sparse point cloud model, and enabling the camera visual angle information to serve as a middle visual angle of the scene, the visual angle difference between the space images and the ground images can be reduced, three-dimensional reconstruction is carried out through camera visual angle information, the rendered image of the middle visual angle after rendering is obtained, feature points corresponding to the rendered image are further removed after feature point matching is carried out on the ground image, the aerial image and the rendered image, and the image quality is improved. At the moment, a matching relationship between the ground image and the aerial image can be obtained, and an air-ground joint scene graph is generated according to the matching relationship to complete matching of the air-ground images. According to the embodiment of the invention, the homonymous connection points between the air-ground images can be automatically generated, and automatic air-ground joint model construction is realized.

The invention discloses an unmanned aerial vehicle-satellite scene matching method based on two stages. Precise scene matching and positioning are completed from coarse granularity to fine granularity. An existing satellite base map retrieval task is used as a first-stage task, and a local feature point matching method is introduced in a second stage. Compared with an existing method, the method has the advantages that the matching robustness in a complex scene is remarkably improved, meter-scale accurate positioning of the unmanned aerial vehicle can be realized on an existing satellite map, and a new universal solution is provided for the field of scene matching.

The invention discloses a multi-target tracking method combining camera motion compensation and pseudo depth estimation, belongs to the field of computer vision, and is suitable for a complex automatic driving road environment. The method comprises the following steps: constructing a training set and a test set; detecting the image by using a deep learning detector and extracting features; a Kalman filter is adopted to correct a motion modeling state vector, and the target position and size prediction precision is improved; solving a homography matrix through feature point matching, performing global camera motion compensation, and reducing camera jitter and displacement interference; target pseudo depth information is calculated, hierarchical cascade matching is carried out, and association performance in dense and shielding scenes is optimized; a three-level cascade strategy is adopted to complete high confidence degree, low confidence degree and residual target matching in sequence; and finally, outputting a tracking result with a detection frame and identity information to obtain a trained model. According to the invention, accurate detection and stable tracking of multi-category targets can be realized in a complex environment, and identity switching is effectively reduced.

The invention discloses an intelligent change detection method for non-aligned two-time-phase images of an inspection unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicle image processing and change detection, and the method comprises the steps: obtaining two-phase images collected by a low-altitude unmanned aerial vehicle under a fixed route and same sensor parameters; a lightweight registration model is constructed and trained, feature point matching is utilized to predict matching point pairs, a homography matrix is calculated, and accurate registration of non-aligned images is achieved; and constructing and training a change detection model based on image pair interaction feature fusion, analyzing the aligned image after registration, and outputting a change information binary image. The method can effectively solve the problem of non-alignment caused by position and angle differences during two-time-phase image acquisition of an unmanned aerial vehicle, and the technical problems of low precision, poor robustness and insufficient calculation efficiency of a traditional method in change detection, effectively improves the automation level of low-altitude safety monitoring of ground highways and railways, reduces the maintenance cost, and improves the safety of the unmanned aerial vehicle. The important application value is realized.

The invention provides a self-adaptive workpiece grabbing pose correction method and system based on feature point matching. The method comprises the steps that a standard image of a workpiece which is successfully grabbed to a jig and an auxiliary image of the workpiece which is not successfully grabbed to the jig are shot; the template image and the auxiliary image are processed, feature points in the images are recognized and compared, matched feature points between the template image and the similar image are found out, and three-dimensional coordinates matched with the feature points are obtained; and based on the matched feature points and the three-dimensional coordinates of the feature points, the pose of the workpiece is solved through an RANSAC PNP algorithm, the pose of the current unsuccessfully grabbed workpiece relative to the position of the template drawing is recovered, and it is ensured that the workpiece can be accurately placed on the jig by adjusting the pose of the mechanical arm. And through feature extraction and matching of the SIFT algorithm and the LightGlue algorithm, accurate and effective correction of the workpiece placement pose is achieved, important industrial application value and market potential are achieved, and powerful technical support is provided for improving the placement precision and efficiency of the mechanical arm.

The invention discloses a three-dimensional visual landscape garden construction method and device and electronic equipment, and relates to the field of garden planning. In the method, a plurality of garden images shot by an unmanned aerial vehicle are obtained, and the garden images are shot from a plurality of preset angles when the unmanned aerial vehicle flies according to a preset route; determining feature points of the garden image; obtaining a shooting time point of each garden image and pos data of the unmanned aerial vehicle at the shooting time point; according to the pos data and the feature points, performing feature point matching on each garden image by adopting aerial triangulation operation to obtain three-dimensional point cloud data of each feature point; constructing a three-dimensional point cloud model corresponding to the landscape garden; and mapping the texture information to the three-dimensional point cloud model to obtain a three-dimensional live-action model of the landscape garden. According to the technical scheme, the unmanned aerial vehicle collects the garden data according to the preset route, large-range and multi-view garden data can be rapidly and systematically collected, and the working efficiency is improved.

The invention relates to the technical field of remote sensing image ship motion, and discloses a wake matching-based ship motion parameter estimation method, which comprises the following steps of: constructing a ship wake model comprising an infrared sea surface model, a ship wake height field and a ship wake temperature field; setting multiple navigational speeds, multiple navigations and multiple sea condition parameters by using the model to generate a ship wake infrared image, and constructing a ship wake database; a deep learning convolutional neural network and a multi-task loss function are adopted to extract a wake to be estimated; and constructing a multi-stage image matching model based on the feature point extraction module, the feature vector description module and the feature point matching module, and matching the database and the batch to-be-estimated trails by using the multi-stage image matching model. According to the method, the infrared sea surface model mixedly calculates two sea surfaces, so that the repeatability of the sea surfaces is reduced while the authenticity is ensured; by simulating the wake temperature field and the wake height field, the authenticity of the wake is improved, the constructed wake data set can be reused, and efficient estimation of ship motion parameters is realized.

The invention discloses a cooperative surveying and mapping control method and system based on an unmanned ship and an unmanned aerial vehicle, relates to the technical field of automatic control, and realizes acquisition, processing and fusion of real-time high-precision three-dimensional image data during operation of the system. Multi-source data are synchronously acquired from an unmanned ship and an unmanned aerial vehicle by using LiDAR, sonar and camera sensors, the robustness of feature extraction is enhanced by denoising and calibration preprocessing, ensuring of data quality and application of an ORB algorithm in a feature point extraction module in combination with gradient directions and texture feature points, and an edge detection module identifies the edge in an image, so that the robustness of feature extraction is improved. And generating the boundary set B and marking the edge, thereby improving the accuracy of the edge information. And the feature point matching module performs feature point matching and data fusion by using an RANSAC algorithm to generate a fused image Ifuse, so that the phenomena of dislocation, incomplete fusion, overlapping and overlapping gaps at image joints are reduced as much as possible.

The invention discloses a waste lithium battery classification processing method and system based on image segmentation identification, and relates to the technical field of waste lithium battery recycling, and the method comprises the steps: configuring a high-resolution camera, a lighting device and an ambient light sensor, collecting multi-view images of waste lithium batteries in real time, obtaining a two-dimensional image data set, and carrying out the recognition of the two-dimensional image data set; inputting the two-dimensional image data set into an SIFT algorithm to detect and match feature points, generating a three-dimensional model through an MVS technology based on the matched feature points, and obtaining geometric information of the surface of the battery by using a triangulation method; according to the method, the high-quality three-dimensional model is generated by combining high-resolution multi-view image acquisition with SIFT feature point matching and the MVS technology, and the quality of the model is optimized through denoising, curvature enhancement and dynamic parameter adjustment, so that accurate modeling of a complex geometrical shape is ensured. And the optimized three-dimensional model is analyzed by using the deep convolutional neural network, so that the robustness and accuracy of defect detection are greatly improved.

The invention belongs to the technical field of medical image processing, and discloses a blood vessel segmentation reconstruction and bleeding analysis method and device. In the bleeding area detection model, detecting whether a bleeding area exists or not; and performing blood vessel segmentation and blood vessel three-dimensional reconstruction operation. According to the method, the time sequence image is adopted for comprehensive analysis, and the problems that blood vessel details are lost due to insufficient diffusion of a contrast agent and the image is blurred due to excessive diffusion are effectively solved. Based on an 8-neighborhood pixel tracking method, accurately identifying blood vessel cross points, and performing segmentation processing on the skeleton line by using the cross points. By constructing the topological structure of the blood vessel, the feature point matching efficiency is greatly improved, and the automatic three-dimensional reconstruction process is also realized. By introducing the architecture design of the hybrid convolutional neural network and the state space model, the local fine-grained feature extraction capability and the global long-range dependence modeling capability are effectively combined, so that the segmentation precision and robustness of the intracranial hemorrhage region are remarkably improved.