8 results about "Epipolar geometry" patented technology

This invention relates to the field of image analysis technology, specifically a deep learning-based method for detecting surface defects in tab adapters. The method includes acquiring the three-dimensional geometric curvature parameters of the bending region through multi-view polarization imaging; decoupling specular reflection and diffuse structural features using a photometric distribution model to resolve high-brightness reflection interference; nonlinearly projecting the surface texture onto a two-dimensional plane using a local tangent space mapping matrix to generate an isotropic corrected texture map to compensate for geometric distortion; identifying suspected defect points that disrupt flow continuity through singular value decomposition of the anisotropic structure tensor; and introducing stress distribution priors and epipolar geometric constraints for multi-view spatial consistency verification to effectively filter lighting artifacts. The verified region and corrected texture are fused and input into a deep recognition network with a spatial attention mechanism to analyze the surface defect image. This invention achieves accurate analysis of surface defect images through deep coupling of three-dimensional geometry and a photometric physical model.

The application discloses a novel OCT space reconstruction method, which comprises the following steps: acquiring multiple coronary angiography image data at different angles; extracting coronary angiography images at the same moment in different cardiac cycles from the coronary angiography image data acquired at each angle; extracting a blood vessel skeleton according to the coronary angiography images; in the process of extracting the blood vessel skeleton, the coronary angiography images at different angles are included; multi-view stereo matching is performed by using deep learning, and matching correction is performed by using an epipolar geometry algorithm; a three-dimensional blood vessel skeleton is acquired according to the coronary angiography images after the matching correction; coronary OCT sequence images are acquired, and rearrangement is performed in combination with the direction of the three-dimensional blood vessel skeleton; body rendering is performed on the rearranged images, and a three-dimensional space model with a real posture is rendered. The method realizes prediction of invisible parts and guarantees the correctness of three-dimensional reconstruction of visible parts, and provides a data basis for screening of vascular deformity and the like in the later stage.

The application discloses a pose estimation method and system based on a complex dynamic scene and a storage medium, and the pose estimation method comprises the following steps: S1, feature points in an input image are extracted, and a motion consistency detection is performed on a motion state of an object by using a sparse optical flow pyramid method or an epipolar geometry method to obtain a motion consistency detection result; S2, real-time semantic segmentation is performed on the input image by using a lightweight semantic segmentation model to obtain a semantic segmentation result; S3, semantic label classification is performed on the object in the scene according to the semantic segmentation result, and potential dynamic object detection is performed according to the semantic label classification result and a corresponding depth image to obtain a potential dynamic object detection result; S4, dynamic points are adaptively removed based on geometric information and semantic information; and S5, pose estimation is performed based on the reserved feature points. The application can effectively guarantee the accuracy and real-time performance of the pose estimation of a visual odometer in a complex dynamic scene.

The invention provides a dual-camera weak texture multi-target space matching and positioning method and an application system, and belongs to the technical field of computer vision and three-dimensional perception. According to the method, based on a dual-camera epipolar geometric model, on the basis of obtaining a dual-camera synchronous image and performing multi-target detection, four corner points and a central point of a target outer surrounding frame are introduced to construct a multi-key-point combined epipolar constraint, and target space structure information is fully utilized. The multi-key-point combined epipolar constraint is constructed by selecting the four corner points and the center point of the outer surrounding frame of the target, the overall space structure information of the target is introduced into the dual-camera matching process, a matching constraint mechanism of multi-geometric clue fusion is formed, and compared with an epipolar constraint mode only based on a single image point, the matching constraint mechanism is more accurate. The matching ambiguity caused by appearance similarity or deformation is effectively reduced, so that the matching accuracy in a multi-target scene and the anti-interference capability of the system are improved, and the method is particularly suitable for a weak texture environment.

The application relates to the technical field of volume data visualization, and discloses a volume rendering Gaussian representation fast construction method based on visual-volume data alignment, which comprises the following steps: obtaining three-dimensional volume data to be visualized and corresponding multi-view images; inputting the three-dimensional volume data and the multi-view images into a feedforward neural network to directly map a three-dimensional Gaussian splat representation for representing the three-dimensional volume data through the feedforward neural network, wherein the feedforward neural network comprises: a dual Transformer network for jointly modeling two-dimensional appearance information of the multi-view images and three-dimensional geometric information of the three-dimensional volume data; and a volume geometry forcing mechanism for aligning and injecting the two-dimensional appearance information and the three-dimensional geometric information into the three-dimensional Gaussian splat representation based on epipolar geometry constraints; and the application is superior to existing direct volume rendering methods and three-dimensional Gaussian representation methods based on optimization in terms of rendering efficiency and visual quality.

This invention relates to a domain-adaptive stereo matching method based on multi-distribution color transformation and consistency, belonging to the field of computer vision technology. It includes the following steps: transferring the color style of a target domain image to a source domain image, and training a pre-defined first model using the acquired source domain dataset. Constructing and initializing a teacher model and a student model with the exact same structure as the first model. Inputting an unlabeled image of the target domain into the teacher model to generate a pseudo-disparity map, constructing mixed training data based on the epipolar geometry of stereo matching, and inputting the pseudo-disparity map and the mixed training data into the student model. Continuously optimizing the parameters of the student model using consistency learning, and deploying the optimized student model as a domain-adaptive stereo matching model. This invention effectively solves the domain difference problem in the prior art and improves the domain-adaptive stereo matching capability through the above methods.

The application relates to a robust visual odometer method, which comprises the following steps: acquiring input images of two adjacent frames based on a monocular camera, and extracting feature points in the two input images respectively; acquiring a plurality of pairs of feature points, wherein the feature points are expressed in pixel plane coordinates; constructing an epipolar geometry constraint equation based on the pairs of feature points, and solving to obtain an initial essential matrix; using the initial essential matrix, and calculating an initial rotation matrix and an initial translation vector of the monocular camera between the two adjacent input images through singular value decomposition; converting the pixel plane coordinates of the pairs of feature points into three-dimensional world coordinates based on the initial rotation matrix and the initial translation vector; calculating the difference between the feature points in the pairs of feature points based on the three-dimensional world coordinates; judging whether the feature points are static feature points based on the difference, and if yes, the feature points are reserved; re-estimating the actual essential matrix of the monocular camera based on the static feature points, and obtaining an actual rotation matrix and an actual translation vector.

This application belongs to the interdisciplinary field of computer vision and smart agriculture, and discloses a method, device, and medium for 3D reconstruction of seedlings based on active lighting and probabilistic occupancy. The method includes: acquiring image sequences under multi-angle active lighting using a single fixed camera and multiple calibrated light sources on a circular array; obtaining photometric properties unaffected by specular reflection through polarization component decoupling; constructing a probabilistic 3D voxel mesh; for each light source, locating surface points by constructing active lighting and shadow epipolar geometric constraints and calculating the geometric intersection of the line of sight and the light ray; introducing a dynamic confidence gate and a neighborhood game mechanism to construct a void topology protection strategy, updating the occupancy probability of voxels on the light path in an adaptive probabilistic manner; finally, correcting the probability field through multi-source shadow consistency verification, and generating the final 3D geometric model using a moving cube algorithm. This invention enables high-precision 3D reconstruction of seedlings under static occlusion without moving the camera or objects.