7321 results about "Depth map" patented technology

The invention discloses an unmanned aerial vehicle routing inspection line adaptive obstacle detection method and system based on a monocular camera. The method comprises the steps of obtaining initial flight parameters, evaluating and optimizing the flight parameters based on scene complexity, collecting an original image sequence and carrying out quality evaluation and exposure compensation. Constructing a multi-resolution feature pyramid, extracting feature descriptors, and performing feature matching and verification; performing multi-constraint pose optimization and depth estimation based on a feature matching result, and reconstructing a three-dimensional point cloud; calculating direction enhancement response to extract a power line candidate area, performing multi-view line segment matching and three-dimensional reconstruction, and analyzing dynamic characteristics of the power line; and based on the power line three-dimensional model and the depth map, carrying out obstacle segmentation, calculating an obstacle distance, evaluating a risk level, and generating early warning information and a monitoring scheme. According to the invention, different scenes can be adaptively processed, the obstacle risk can be accurately identified and evaluated, and the inspection efficiency and reliability are improved.

The invention relates to the technical field of navigation, and particularly provides a quadruped robot autonomous navigation method and system for a special environment, and the method comprises the steps: generating a semantic tag for depth point cloud data corresponding to a depth image through employing a recognition result of a target detection network for the depth image; calculating local odometer information according to the laser point cloud data and the inertial sensor data by using a laser radar mileage calculation method; fusing the depth point cloud data with the semantic label, the laser point cloud data with the semantic label and the local speedometer information to obtain positioning data in a pre-constructed point cloud map; constructing a multi-layer cost map according to the pre-constructed point cloud map, the newly acquired depth image, the laser point cloud data and the inertial sensor data by using a sensing method of a region planner; and generating navigation path information according to the positioning data and the multi-layer cost map by using a global path planning algorithm and a local path planning algorithm. The method improves the navigation precision.

The invention provides a bridge structure crack position identification method based on deep learning and computer vision, and the method comprises the steps: obtaining a multi-view visible light image and a depth image, fusing the visible light image and the depth image according to registration parameters, and generating a fusion image containing the texture and depth information of a bridge; for the fused image, extracting a crack region by adopting an adaptive threshold segmentation algorithm, and dynamically adjusting the threshold to obtain a crack binary image; according to the crack binary image, calculating three-dimensional coordinates of crack pixel points through a three-dimensional reconstruction algorithm, and performing reconstruction to obtain three-dimensional point cloud data containing crack space position and form information; on the basis of the three-dimensional point cloud data, three-dimensional local feature descriptors are adopted to describe geometric attributes of the surface of the crack, crack fragment parameters are calculated, and a three-dimensional crack model is constructed; based on a deep learning model, a transfer learning strategy is adopted, a self-adaptive transferable bridge crack three-dimensional detection method is formed, and model parameters are updated so that the model parameters can adapt to the characteristics of a target bridge.

The invention relates to the field of defect detection, in particular to a PCB defect detection method and system based on image recognition. The method comprises the following steps: collecting a multidirectional PCB detection image, carrying out pixel-level registration correction and adaptive pixel stability compensation, and constructing a space-time stability compensation image sequence; performing reverse pyramid structure division on the space-time stability compensation image sequence, performing normalized similarity probability calculation, and constructing an initial region classification result; based on an initial region classification result, depth image visual analysis is carried out, pseudo defect comprehensive elimination optimization is carried out, and a pseudo defect purification high-confidence image is constructed; pCB connection defect identification is carried out on the pseudo defect purification high-confidence image, global defect point distribution marking is carried out, and a defect point space distribution diagram is constructed. According to the invention, high-credibility, high-precision and high-closed-loop PCB defect detection is realized.

The invention discloses a three-dimensional Gaussian sputtering scene reconstruction method based on structure perception refined Gaussian, and aims to solve the problems of Gaussian drift, edge blur, structure artifacts and the like of a reconstruction model due to the fact that sparse point cloud contains outliers, Gaussian morphology and normal are mismatched and a multi-dimensional optimization target is lacked in an existing three-dimensional Gaussian sputtering reconstruction method. A key frame is extracted by collecting target scene video data, sparse three-dimensional point clouds are reconstructed by using an SfM algorithm, a depth map and a normal map are generated through a Lotus model, three-dimensional Gaussian distribution is initialized after the sparse point clouds are filtered, a Gaussian covariance matrix is adjusted by using a normal consistency regular term, and the sparse point clouds are extracted. And after structure attribute analysis is carried out, a comprehensive scoring function is constructed to screen Gaussian points, and finally, a combined training framework including luminosity, normal consistency and structure continuity loss is adopted to optimize and generate a three-dimensional Gaussian scene model. The method is mainly applied to the field of three-dimensional reconstruction and multi-view rendering, and scene reconstruction precision and geometric consistency can be improved.

A multi-scale cascaded hourglass depth map completion method guided by RGB image is presented. [Solution] The multi-scale cascade hourglass depth map completion method includes step S1 of obtaining a sparse depth map to be completed and a corresponding RGB color guide image, performing channel-dimensional pre-processing on the sparse depth map and the RGB color guide image and connecting them in series to obtain a target RGB_D image to be processed, step S2 of inputting the target RGB_D image to be processed into a trained depth map completion network model including an early fusion encoder, a sparse depth map teaching module, a multi-scale hourglass completion module and an optimization reinforcement module, and step S3 of obtaining the output of the optimization reinforcement module as a completed dense depth map.

The invention discloses a multi-view panoramic point cloud splicing method, which comprises the steps of performing joint calibration on a binocular camera and a laser radar through a calibration algorithm, and aligning a coordinate system; a binocular camera is used for collecting a two-dimensional image and carrying out distortion correction, a depth map is generated based on parallax calculation, and three-dimensional point cloud reconstruction is carried out in combination with a triangulation principle; performing deep learning target detection on the two-dimensional image, and screening a line rod assembly area through non-maximum suppression; multi-view point cloud data are collected through a laser radar, point cloud segmentation processing is carried out, and a telegraph pole assembly point cloud subset is reserved; matching the two-dimensional pixel area of the telegraph pole assembly with a laser radar point cloud projection result, and screening point cloud data belonging to the telegraph pole assembly; and carrying out alignment and fusion on the multi-view point clouds through initial registration and fine registration by adopting an improved point cloud splicing algorithm to generate a panoramic point cloud image. According to the method, the point cloud splicing processing time is shortened, and the robustness, precision and integrity of point cloud splicing are improved.

The invention relates to an optimization method and device for sparse view angle three-dimensional Gaussian splash, and belongs to the technical field of three-dimensional reconstruction in computer vision, and the method comprises the steps: collecting a sparse view angle image; a multi-view stereoscopic vision model based on deep learning generates a geometrically consistent depth map for the sparse view image, converts the depth map into point clouds and fuses the point clouds to obtain dense point clouds; sampling dense point clouds by adopting voxel-guided farthest point sampling to obtain initialized point clouds, and constructing a three-dimensional Gaussian field; rendering the three-dimensional Gaussian field through an enhanced geometric renderer to obtain a rendering depth and a rendering normal; constructing a multi-level geometric regularization loss function, and optimizing the three-dimensional Gaussian field; and performing optimization adjustment on the three-dimensional Gaussian field based on a shape-scale constraint criterion and a two-stage adaptive opacity constraint strategy to obtain an optimized three-dimensional Gaussian field. According to the method, the problems of initialization failure, insufficient geometric supervision and element out-of-control of 3D Gaussian splashing under the sparse view angle are solved.

The invention discloses a robot grabbing posture generation method and a related device, and the method comprises the steps: obtaining an RGB image, a depth image and a position coordinate of a target object, and generating a spatial calibration matrix through data preprocessing; target object segmentation is carried out on the calibrated RGB matrix according to the target point coordinate sequence, a segmentation mask is generated, and object mass center coordinates are calculated; generating a three-dimensional point cloud by using the calibrated depth matrix, the segmentation mask and the camera parameters, extracting a plane point set and a non-plane point set through an RANSAC algorithm, and analyzing to obtain an axial feature vector and a plane normal vector; finally, grabbing parameters are calculated according to the feature vectors and the centroid coordinates, and a grabbing posture transformation matrix is generated through vector operation. According to the technical scheme, under the condition of not depending on a preset model library, the grabbing posture of an unknown object is generated by analyzing the geometrical characteristics of the object, the accuracy problem in the process of converting two-dimensional image information into three-dimensional grabbing parameters is solved, and the adaptability and reliability of a robot grabbing task are improved.

The invention provides an access control system based on face recognition and a recognition method, and the system comprises a multi-modal image collection subsystem which integrates a visible light camera, a depth camera and an infrared thermal imaging camera and is used for synchronously collecting a visible light image, a depth image and an infrared thermal imaging image of a face; the image preprocessing unit is used for carrying out corresponding preprocessing on the image; the feature extraction network module is used for extracting image features; the multi-modal feature fusion and alignment module is used for calculating a mutual information matrix by adopting a feature fusion strategy based on mutual information maximization and adaptive weighting, determining an adaptive weight and performing weighted summation, and performing feature alignment by applying a method based on canonical correlation analysis; and the face recognition classification and verification module verifies the image by adopting a multi-level fusion strategy. According to the method, the false recognition rate and the rejection rate are effectively reduced, the strong anti-cheating capability is achieved, and cheating attacks of various fake face means can be accurately recognized and resisted.

The invention relates to the technical field of multi-target tracking, in particular to a dynamic multi-target tracking and trajectory prediction method. Comprising the following steps: acquiring a color image and a depth image, and preprocessing; constructing a dynamic threshold double-flow neural network, extracting features from the preprocessed color image and depth image to obtain a visual feature set and a depth feature set, fusing the visual feature set and the depth feature set through an adaptive attention mechanism to obtain a fused feature set, and combining a dynamic threshold to obtain an observation information list of a target; updating the observation information list through a self-motion decoupling mechanism to obtain a real information list; establishing space association and time association between the target and the track based on the real information list to obtain a newest track set; and analyzing target characteristics, environmental constraints, social behaviors and historical trajectories, generating a prediction trajectory of each target and a corresponding multi-factor score, generating a time-varying influence graph, and performing planning and decision making by adopting a double-layer decision-making mechanism.

The invention discloses a bridge appearance disease detection method and system based on laser point cloud and visual image fusion, and the method comprises the steps: carrying out the synchronous data collection through employing a calibrated laser radar and a camera, and obtaining a three-dimensional laser point cloud and a two-dimensional visual image of a bridge appearance; projecting the three-dimensional laser point cloud to a two-dimensional visual image plane through a mapping model to generate a point cloud front view; identifying diseases in the two-dimensional visual image by adopting a preset target detection model, and generating a segmentation mask for each disease; screening out points in the segmentation mask from the point cloud front view to obtain a potential disease three-dimensional point cloud; error points generated by local shielding in the potential disease three-dimensional point cloud are filtered through ray tracing, and a real disease three-dimensional point cloud is obtained; distributing the depth information of the real disease three-dimensional point cloud to corresponding pixel points, generating a depth map of a disease segmentation mask region, and performing quantitative calculation on the physical size of the disease; and displaying the real disease three-dimensional point cloud data and the disease physical size. The method is high in efficiency and precision.

An aero-engine augmented reality virtual-real fusion method based on depth prior scene reconstruction comprises the following steps: firstly, collecting data of an aero-engine look-around RGB image and a depth image, calculating a camera frame pose by using an SFM algorithm and obtaining a sparse point cloud, and optimizing the camera pose by using an ICP algorithm and the point cloud converted from the depth image to obtain an aligned prior dense point cloud; gaussian points are initialized based on the priori dense point cloud and the sparse point cloud, and random generation of 3D Gaussian is limited by using the boundary of the dense point cloud as geometric constraint; through the three-dimensional reconstruction of the adjacent multi-view enhancement strategy, the spatial and visual consistency is improved, and an aero-engine model is reconstructed; then understanding the reconstructed aero-engine model based on semantic segmentation, identifying and marking core components, and finally performing virtual-real fusion visualization on the reconstructed aero-engine through augmented reality to guide training, assembling and maintenance operations. According to the invention, rapid high-quality three-dimensional reconstruction and augmented reality virtual-real fusion visualization of the aero-engine are realized.

The embodiment of the invention provides an unhooking robot control method based on multi-source visual fusion, which is applied to the technical field of robot control and comprises the following steps: acquiring an RGB image, a depth image, an infrared image and IMU data through a multi-source sensing system mounted at the tail end of a robot; carrying out feature fusion identification by adopting a double-branch neural network, and outputting the boundary contour of the lifting hook and the three-dimensional coordinates of the optimal grabbing point; the visual coordinates are unified to a robot base coordinate system through a registration correction mechanism; a Transform prediction model is constructed based on the visual and inertial signals, and future pose changes of the lifting hook are estimated; a feedforward control track is generated to counteract swing of the lifting hook, and track correction is carried out in combination with visual servo feedback; and a joint instruction is generated through path planning and inverse kinematics solution, and the mechanical arm is driven to complete precise unhooking operation. According to the method, the recognition precision, the anti-interference capability and the operation success rate of unhooking operation in complex illumination and dynamic environments are effectively improved.

Systems and methods for generating custom art fonts with consistent style include receiving user input that identifies a base font style for a custom font and includes descriptive text that defies one or more text effects to use for the custom font. Depth maps are selected for characters to be included in the custom font. The depth maps are preprocessed to add noise to the depth maps. A generative model generates custom font images conditioned with the text prompt and the depth maps. The custom font images are then used to render text on a display screen of a computing device.

The invention provides a multi-modal image analysis-based robot-assisted wound treatment method and system, and relates to the technical field of robots, and the method comprises the steps: collecting a multi-modal image of a to-be-treated wound, determining an image registration result, obtaining a wound region image through a multi-scale fusion algorithm and wound region segmentation, and carrying out the type classification of wound tissues. A wound tissue type classification result is obtained; performing surrounding shooting on a wound area to obtain a wound multi-view image, registering subsequent depth maps frame by frame by taking a first frame of depth map as a reference, accumulating point cloud data, determining a wound three-dimensional point cloud, generating a wound three-dimensional reconstruction result by combining a curved surface evolution algorithm, and determining a wound parameter measurement result; based on the wound tissue type classification result and the wound parameter measurement result, a robot path planning algorithm is adopted, the optimal movement track of the robot is generated, the movement parameters of the robot are adjusted in real time through a robot control algorithm of vision and force feedback, and wound adjuvant therapy is completed.

The invention relates to a roadside signboard identification method based on unmanned aerial vehicle three-dimensional point cloud lightweight semantic segmentation, and belongs to the technical field of intelligent traffic. The method comprises the following steps: optimizing a point cloud acquisition path through adaptive flight control; adopting an RSPAE algorithm to enhance local geometric features of the point cloud; converting the point cloud into a three-channel fusion image (a depth image, an intensity image and a local depth variance image); extracting multi-scale features by using a double-branch neural network, and fusing the aligned features through a GFM module and a CFM module; the lightweight decoder recovers a high-precision semantic segmentation map; and generating a final identification result by combining geographical registration and multi-frame redundancy suppression. And state evaluation and anomaly detection are realized based on an MLP scoring device and a mahalanobis distance. According to the method, the segmentation precision, the reasoning speed and the positioning precision are remarkably improved in a complex scene, the method is suitable for deployment of embedded equipment, and the problems of low efficiency and high omission ratio in the prior art are solved.

The invention provides a posture recognition algorithm for any object under a monocular camera and an application system, and the algorithm comprises the steps: S1, constructing a target three-dimensional model, carrying out the multi-view annular shooting image collection of a target, and generating a dense grid model through feature extraction, matching, posture calculation and a multi-view geometric method; s2, generating an image depth map, and predicting depth information of a target in a motion process based on a monocular image sequence; s3, extracting a target image mask, and generating a target area mask graph through an image encoder, a prompt encoder and a mask decoder; and S4, executing attitude estimation, performing attitude initialization, correction and screening by combining the three-dimensional model, the depth map and the mask map, and outputting a six-degree-of-freedom attitude result of the target. According to the method, the target is subjected to annular shooting modeling through the method based on multi-view geometry, the three-dimensional model of the target is generated, attitude estimation is achieved in combination with the image mask and the depth map, the generalization ability of an attitude estimation algorithm in an actual scene is improved, and the application range of the attitude estimation algorithm in the actual scene is widened.

The invention discloses a weld joint key point recognition method, device and equipment, a medium and a program product, and the method comprises the steps: generating a data set based on training image data and annotation data, constructing an original recognition model, improving the original recognition model, and training and optimizing the improved original recognition model based on the data set. Obtaining a target recognition model; acquiring a depth image and an RGB image of the welding part through image acquisition equipment on the welding robot; the RGB image is input into the target recognition model for weld joint key point recognition, weld joint key point coordinates of the welding part are determined based on the recognition result and the depth image, accurate recognition of the weld joint key points is achieved, a welding robot is guided to accurately reach the weld joint key point coordinate positions for welding, key point recognition autonomy and adaptability are improved, and the welding quality is improved. And welding tasks with different requirements are effectively adapted, the welding efficiency is greatly improved while the welding quality is ensured, and the time cost is reduced.

The invention provides an interaction method and system based on deep learning, and relates to the technical field of computer vision, and the method comprises the steps: collecting a continuous image sequence, extracting feature points, calculating the position variation and gradient variation of the feature points, constructing a two-dimensional feature vector, and carrying out the clustering through a Gaussian mixture model, thereby obtaining a stable feature class. And weighting the feature points based on a feature weight distribution function, calculating feature similarity and constructing a matching relationship. And calculating a dynamic parallax threshold in combination with the vehicle movement speed, screening reliable matching feature pairs, and solving pose transformation parameters. A feature change sequence is extracted through a feature detection network, a motion feature mapping matrix is constructed, and a compensation depth map is generated in combination with depth information. And scene reconstruction is carried out based on the compensation depth map and the pose transformation parameters, the scene confidence score is calculated, the pose parameters are calibrated, a high-precision pose parameter sequence is obtained, and the robustness of feature matching and the accuracy of pose estimation can be improved.

The present application relates to a real-time online cargo volume recognition method, system, and device, and a medium. The method comprises: collecting, by means of a camera, a depth image and a color image of a set area comprising a cargo; introducing a median filter to remove noise pixels in the depth image, and aligning the color image with the filtered depth image on the basis of internal and external parameters of the camera to obtain an aligned color image; on the basis of color, light, and grayscale differences between the cargo and a non-cargo object, using a pre-trained YOLACT model to perform instance segmentation on the aligned color image, and generating a cargo mask; and performing edge fitting, surface extraction, and background separation of the cargo by means of the aligned color image, the filtered depth image and the cargo mask, so as to determine the volume of the cargo. Thus, the present application has high stability and robustness, reliable data can be obtained regardless of the placement location and the placement direction of a cargo, and the working efficiency can be improved.

The invention discloses a robot dynamic environment adaptive sensing and navigation system based on a three-dimensional laser radar, relates to the technical field of robots, and solves the technical problems that comprehensive environment information is difficult to obtain, and the weight is difficult to adjust by fusing weather types and sensor confidence coefficients. Comprising the following steps: generating original point cloud data by combining a bionic compound eye type laser radar with a silicon photon integrated chip; marking point clouds based on a KITTI data set, performing preprocessing, training a Transform model to output a dynamic obstacle mask, and filtering background point clouds; a weather detection model is constructed, the weight is dynamically adjusted according to the weather type and the sensor confidence coefficient, and position and attitude estimation is fused; laser radar point cloud constructs a geometric map, a camera depth map generates a dense map, and semantic tags are mapped to generate an environmental semantic map; and converting the environmental semantic map into a three-dimensional grid map, and planning an obstacle avoidance path based on the grid map by using an A * algorithm.

The invention discloses a 6D pose estimation method and system based on an RGBD camera, and relates to the technical field of computer vision processing. Calibrating camera parameters based on the calibrated camera imaging; reconstructing a three-dimensional model and constructing a data set; segmenting the target object and calculating the actual size based on the three-dimensional model and the data set; establishing a 6D pose estimation network model based on the three-dimensional model and the data set; and performing pose estimation based on the actual size. The method has the advantages that the accuracy of 6D pose estimation is remarkably improved by combining the RGB image, the depth image and the point cloud data, object region segmentation is performed by using the SAM model, the actual size of the object is visually calculated by combining the depth information and the point cloud data, the precision of size estimation is improved, and the method is suitable for popularization and application. The introduction of depth information enables the pose estimation system to have better adaptability to objects with different sizes, shapes and surface features, and improves the robustness and accuracy of the system.

The invention belongs to the technical field of three-dimensional imaging and measurement, and particularly discloses a laser scanning three-dimensional imaging method and device, a medium and equipment, and the method comprises the steps: obtaining laser scanning data of a target object, and the laser scanning data comprises depth data including reflection intensity, a scanning angle, a timestamp and flight time; preprocessing the laser scanning data, and fusing a time sequence and scanning path information to obtain multi-channel input features adaptive to a neural network; constructing a three-dimensional geometric reconstruction model, and training the model; inputting the multi-channel input features into a trained three-dimensional geometric reconstruction model to obtain a dual-channel output tensor containing a dense depth map of the target object and a corresponding uncertainty map; converting the dual-channel output tensor comprising the dense depth map of the target object and the corresponding uncertainty map into a three-dimensional point cloud under world coordinates; and post-processing the three-dimensional point cloud to obtain a three-dimensional image of the target object.

The invention belongs to the technical field of image analysis, and discloses an old people health state prediction and alarm method based on AI image analysis, and the method comprises the steps: obtaining a depth image, a color image and a pressure image of an activity environment of old people, carrying out the combined analysis of the depth image and the pressure image, recognizing a terrain semantic region in a monitoring region, and carrying out the prediction of the health state of the old people; and carrying out dynamic segmentation on the gait contour of the old people. And the terrain gait coupling degree is calculated by analyzing the depth change characteristics and the motion period characteristics of the gait sub-regions, so that the abnormal gait sub-regions are screened out. And for the abnormal regions, the terrain gait coupling degree, the dynamic texture features of the color image and the pressure distribution features of the pressure image are fused, and a cross-modal health risk vector is constructed. Based on the vector, the health state of the old people is accurately predicted, and a corresponding alarm signal is generated. According to the invention, precise evaluation and timely early warning of the health state of the old people are realized, and intelligent guarantee is provided for the home safety of the old people.

The invention provides an interactive robot grabbing method based on a large language model, and the method comprises the steps: obtaining a task operation sequence according to a two-dimensional image of a current scene, dialogue interaction information and the large language model, the task operation sequence comprises a target object, component information of the target object, target grabbing position information and action information; obtaining a target mask area according to the two-dimensional image and the target capture position information in combination with an image segmentation model, and obtaining a three-dimensional point cloud image with a target capture position according to the target mask area and the depth map of the current scene; and according to the action information, the three-dimensional point cloud picture with the target grabbing position and a grabbing prediction model, a target grabbing pose is obtained, and according to the target grabbing pose, a robot is controlled to grab the target object. According to the method, the analysis capability of the robot on the complex semantic instruction is improved, and the adaptability of the robot in a refined operation task is also improved.

The invention discloses a self-adaptive three-dimensional scene reconstruction method and system based on a single panorama, and belongs to the field of computer vision processing. The method comprises the following steps: firstly, generating a depth map through an indoor scene panorama and constructing an initial three-dimensional grid; then generating a multi-view image and a shielding mask thereof based on view conversion, forming a training sample pair for fine tuning of the diffusion completion model, and injecting scene prior information for the model; then extracting a grid boundary contour and calculating a central axis, and adaptively constructing a camera pose set; after the integrity of the grid is optimized through iteration completion, the grid is converted into a 3D Gaussian sputtering field, and Gaussian point parameters are optimized through multi-view data; in the optimization process, an up-sampling strategy of rendering error feedback and edge Gaussian points is adopted, and finally a scene model with complete geometric textures is output. According to the method, a high-quality three-dimensional scene is reconstructed from a single panorama through a structure self-adaptive completion and Gaussian refining technology, and the method is particularly suitable for immersive roaming reconstruction of indoor scenes.

The invention relates to the technical field of topographic mapping, in particular to a DEM super-resolution reconstruction method and system combined with a depth map, and the method comprises the steps: constructing a multi-modal data set which comprises an optical image, a depth map, a low-resolution DEM image and a high-resolution DEM image which are paired; designing a multi-modal fusion super-resolution network; designing a pseudo Siamese feature extraction module to extract low-level features and high-level features of the optical image, the depth map and the low-resolution DEM image; fusing the low-layer feature map generated by the pseudo Siamese feature extraction module in a shallow layer through a multi-scale feature fusion module, and fusing the deep-layer feature map in a deep layer; outputting a high-resolution DEM image through a DEM reconstruction module; and training the multi-modal fusion super-resolution network by using the multi-modal data set and the collaborative loss function. According to the method, the depth map generated by the optical image is used as priori knowledge to guide the DEM super-resolution reconstruction task based on the basic depth large model, a complex fine adjustment process is not needed, and the DEM reconstruction precision and efficiency are improved.

A dynamic adaptive depth camera occlusion detection method, system and device, and a storage medium, the method comprising: step T1: constructing a three-frame time sequence analysis window, continuously obtaining infrared speckle image sequences at different moments and corresponding dense depth map sequences, and establishing a pixel-level space-time alignment mapping relationship; t2, calculating a motion vector between adjacent frames, and executing the step T3 when the average motion intensity exceeds a dynamic threshold value; wherein the dynamic threshold is determined by the dense depth map; t3, positioning a speckle feature point in the infrared speckle image, generating an analysis window by taking the speckle feature point as a center, calculating a gradient change rate, and when the gradient change rate is smaller than a gradient threshold value, marking the region as a primary candidate region; and T4, carrying out space-time consistency comparison on the primary candidate region and two adjacent frames, and reserving a shielding region of which the variation value is smaller than a variation threshold value in three continuous frames. The method is high in precision, good in real-time performance and high in stability.

The invention provides a mountainous area tunnel construction safety intelligent monitoring and early warning method and system, and relates to the technical field of construction safety monitoring, and the method comprises the steps: collecting visible light and depth images of tunnel surrounding rock, and carrying out the segmentation and extraction of crack features through a depth attention network after image preprocessing and data fusion; extracting parameter time sequence data based on the crack spatial position and the type feature; determining fracture evolution characteristics and critical state parameters by combining wavelet transform and stress-rate coupling analysis; and adopting deep reinforcement learning to calculate the instability probability and generate early warning information. According to the invention, intelligent identification, instability prediction and risk early warning of tunnel surrounding rock cracks are realized, and the safety monitoring accuracy and early warning timeliness are improved.