12 results about "Partial occlusion" patented technology

This application belongs to the field of vision system calibration technology, specifically relating to a target, a target center localization method, and a target calibration plate calibration method. The target provided in this application includes a circular positioning area and an annular recognition area; the annular recognition area surrounds the circular positioning area, which has a triangular mesh, comprising triangles or sectors symmetrically arranged around its center; the target center point is located at the centrally symmetrical point of the triangles or sectors; an annular coding band is provided inside the annular recognition area, comprising multiple arc-shaped coding strips; these arc-shaped coding strips are centrally symmetrical around the target center point. The target center of this application is the intersection point of the legs of multiple triangles in the triangular mesh. Even if the target is partially occluded, or the target center is occluded, the target center point can still be accurately located by connecting the intersection points of the legs of the unoccluded triangles in the triangular mesh.

Systems and methods are provided for treating conditions such as heart failure and / or pulmonary hypertension by at least partially occluding flow through the superior vena cava for an interval spanning multiple cardiac cycles. A catheter with an occlusion device is provided along with a controller that actuates a drive mechanism to provide at least partial occlusion of the patient's superior vena cava, which reduces cardiac filling pressures, and induces a favorable shift in the patient's Frank-Starling curve towards healthy heart functionality and improved cardiac performance. The system may include sensors to determine the degree of occlusion of the superior vena cava. The occlusion system may be used to reduce volume in a heart and facilitate a cardiac procedure. The occlusion system may be used to relieve an overloaded chamber during and / or after deploying a VAD.

The present application belongs to the field of visual tracking technology, and discloses a kind of processing existing partial occlusion flexible linear object visual tracking method, based on RGB-D camera obtains color image and depth image data, from first frame image, by mask segmentation, skeletonization processing operation generates initial state point, in each subsequent frame, object point cloud is generated by mask in conjunction with depth image;Then, the density of its surrounding current point cloud is counted, and the state point is marked as "visible" or "occlusion";By establishing the analytical model based on the spatiotemporal continuity of flexible linear object, the closed form solution of the occlusion position is derived, and the iterative optimization process of the occlusion position is eliminated, thereby reducing the computational complexity and achieving efficient real-time tracking;Meanwhile, using the inherent spatiotemporal geometric evolution law of flexible linear object, a position estimation model that can be directly analytically solved is constructed, avoiding the stability risk brought by iterative optimization, and significantly improving the reliability and tracking stability of the occlusion position estimation.

This invention discloses a computer vision-based fatigue detection system and trajectory analysis method for ICU nurses, relating to the field of medical and health monitoring. It aims to address the problems of subjective inaccuracy, interference with nursing operations, and lack of longitudinal trajectory analysis in existing detection methods. The system includes modules for image acquisition, preprocessing, eye feature extraction, fatigue assessment, data storage, and trajectory analysis. It acquires video streams via a USB camera and adaptive LED fill light, performs preprocessing such as HSV conversion and grayscale enhancement, extracts 12 eye feature points based on MediaPipe, and combines sliding window filtering and posture calibration to compensate for occlusion and posture errors. Three levels of fatigue are determined using EAR and PERCLOS values. The method collects data longitudinally at nine time points, uses an LGMM model to analyze fatigue trajectory heterogeneity, and combines Logistic regression to uncover key influencing factors. This invention features a non-invasive design, high detection accuracy in partially occluded scenarios, precise location of intervention targets, and convenient deployment, providing scientific support for nursing management.

This invention relates to the field of fatigue driving detection technology and discloses a driver fatigue driving detection method based on two-stage detection and eye feature analysis. The method includes: acquiring a real-time video stream containing the driver's face; performing face and key point detection, accurately locating the face bounding box and extracting three-dimensional face key points; performing posture-adaptive eye feature analysis, calculating and comparing the aspect ratio of both eyes to determine the direction of face deflection and selecting the eye with the clearest features as the optimal analysis object for single-eye image cropping; inputting the cropped single-eye image into an image classification model to determine the eye opening and closing state; and finally, cumulatively calculating fatigue metrics such as the percentage of eyelid closure and blinking frequency per unit time, comparing them with preset thresholds to determine the fatigue state and output a warning. This invention overcomes interference caused by driver head deflection or partial occlusion by adaptively selecting the optimal single-eye analysis, improving the robustness and accuracy of fatigue detection in complex real-world driving scenarios.

This application relates to the field of robotics technology and discloses a robot visual target detection method and system, including acquiring visual image information of the target to be detected; identifying multiple edge segments that are broken due to occlusion from the visual image information; logically connecting the edge segments located on the same straight line to construct the reconstructed edge of the target to be detected; acquiring preset geometric specification information that matches the category of the target to be detected, and determining a reference benchmark that meets preset stability conditions in the reconstructed edge; performing geometric extrapolation based on the reference benchmark and the preset geometric specification information to complete the missing contour features of the target to be detected in the occluded area; determining the working pose information of the target to be detected based on the complete contour features of the target to be detected after completion, and restoring the complete contour of the target when the target is partially occluded and the visual information is incomplete, accurately determining its working pose, and improving the accuracy and robustness of target detection in complex occlusion scenarios.

This invention discloses a robot barcode scanning method, apparatus, device, and medium. It relates to the field of automatic barcode scanning technology. The method includes: determining the real-time detection success probability of each candidate barcode scanning trajectory in a trajectory library based on the robot's current state information; the real-time detection success probability characterizes the likelihood that executing the candidate barcode scanning trajectory in the current state will successfully identify the barcode; determining the confidence-weighted success rate of each candidate barcode scanning trajectory based on the prior success probability of each candidate barcode scanning trajectory and the real-time detection success probability; selecting the optimal trajectory from the candidate barcode scanning trajectories based on the confidence-weighted success rate and the execution cost of each candidate barcode scanning trajectory; and controlling the robot to scan according to the optimal trajectory. This invention can overcome the problem of barcode scanning failure under non-ideal conditions such as posture deviation, lighting changes, or partial occlusion with a single fixed trajectory, thereby improving the barcode scanning success rate and operational efficiency.

This application provides a pose estimation method, electronic device, and medium. The method includes: acquiring an RGB-D image of an object; generating observation point cloud data of the object based on the RGB-D image; and determining the observation point cloud features of the object based on the observation point cloud data; performing coarse pose estimation on the object based on the observation point cloud features and the multi-view template point cloud features of the object to obtain coarse six-degree-of-freedom pose data of the object, wherein the multi-view template point cloud features are pre-stored in a template feature library; and performing fine pose estimation on the object based on the coarse six-degree-of-freedom pose data, the observation point cloud data, and the pre-rendered source point cloud data of the object to obtain fine six-degree-of-freedom pose data of the object, so as to improve the accuracy and robustness of the object pose estimation under different viewpoints, different lighting conditions, and partial occlusion.

The application relates to the technical field of image recognition and computer vision. A circular character recognition method, device and equipment based on circle center relocation are disclosed. The method comprises the following steps: acquiring an image to be recognized; performing contour line detection on the image to be recognized based on a pre-set contour line detection model to obtain an outer contour of a target device in the image to be recognized; determining a target character region based on the outer contour of the target device in the image to be recognized and the image to be recognized and based on a pre-set character extraction model; determining a character region rotation center based on the target character region and rotating the target character region based on the character region rotation center based on a pre-set circle center relocation model; and recognizing characters in the target character region based on a pre-set character recognition model. The application effectively overcomes the problem of circle center deviation caused by inaccurate outer contour extraction or partial occlusion, can improve the recognition rate, and realizes stable and reliable recognition of circular characters.

The application provides a distraction driving detection method based on improved YOLOv5, which detects key points by analyzing the head posture and body posture of the driver. First, an AIFI module is introduced based on YOLOv5. The module can better capture subtle changes and fine features of the driver's behavior by optimizing the interaction between deep layers of the same feature map, avoiding the complex overall performance evaluation and data analysis process in the SPPF module, and thus being more efficient in calculation and suitable for real-time application scenarios. Second, the original Concat module of the neck layer is replaced with a TFE module, which fuses feature maps through adaptive scale adjustment and channel attention weighting, thereby improving the accuracy of small target detection such as mobile phones and cups. Finally, a shape-IoU module that considers shape information is introduced, which enhances the robustness of the target with occlusion, deformation or partial occlusion, and provides more stable performance compared to the original IoU.

The application discloses a sequence mixed mask scene text recognition method based on appearance difference guidance, comprising the following steps: 1) establishing a difference-guided sequence mixed mask DSHM pre-training framework; 2) appearance difference quantization; 3) sequence mixed mask generation; 4) difference condition decoder construction: modifying a student network decoder, introducing a difference condition token DC-Token, and realizing adaptive reconstruction; 5) multi-loss joint training; 6) migrating the pre-trained student network to various scene text recognition STR tasks, and utilizing the difficult area reasoning ability learned in the pre-training stage to realize robust text recognition in scenes such as occlusion, distortion and low resolution. The application can effectively solve the recognition challenge caused by the appearance difference caused by partial occlusion or deformation.