11 results about "Contour tracing" patented technology

Disclosed herein is intravascular ultrasound (“IVUS”) system is disclosed which includes an IVUS console for generating a vessel contour trace and a blood flow image from an IVUS image. The IVUS console includes a communication interface; a memory circuit; an image output display; and a processor circuit configured to, when an interim vessel contour trace meets or exceeds accuracy criteria, display on the image output display a complete vessel contour trace comprising the interim vessel contour trace with user-interactable key nodes or control points and merge or superimpose a blood flow IVUS image with the complete vessel contour trace.

The invention relates to the technical field of indoor design, in particular to an indoor space structural partition constraint generation method based on a language model, which comprises the following steps: acquiring a four-channel vector house type image and a reference space data set of a target indoor space, the reference space data set comprises room parameters and connection relation data of various house types; polygonal vertex coordinates of all rooms are extracted from the four-channel vector house type image through a contour tracking algorithm, and an initial vertex data set is obtained; and preprocessing the initial vertex data set and the reference space data set. According to the method, the geometric information of the room can be automatically extracted from the complex house type image by using the contour tracking algorithm, gray stretching, contrast enhancement and other technologies, and standardization processing is carried out, so that partitioning, connection relationships and the like in the interior design process can be efficiently and automatically completed, manual intervention is reduced, and the efficiency is improved.

The invention discloses a gravitational lens critical curve contour tracking method and system based on nearest neighbor distance and geometric constraint, and relates to the technical field of image processing. The method comprises the steps that after a contour point set is obtained, double tolerance is carried out through a nearest neighbor distance method, an initial base line is determined through two-step tracking search, and the contour tracking process is accurately controlled by setting a direction tolerance parameter and a distance tolerance parameter. In the iteration process, combining a direction constraint (an included angle cosine value is greater than a preset threshold value) and a distance constraint (a distance proportion is smaller than the preset threshold value) to screen a next contour point, and performing nearest neighbor search again when a point conforming to the constraint is not found; meanwhile, a double-contour synchronous tracking method is provided, and the consistency of associated contours is ensured by synchronously exchanging corresponding points. The system comprises a geometric calculation module, a nearest neighbor searching module, a tolerance tracking module and a critical pair tracking module. The method improves the continuity and accuracy of contour tracking, is especially suitable for high-precision tracking of lens curve and critical-caustics peer-to-peer complex contours and multi-contour associated scenes, and can realize high-resolution and smooth contour extraction. Compared with a traditional method, the method has the remarkable advantages in the aspects of tracking precision, robustness and complex geometrical shape processing capacity.

This invention discloses a dynamic image analysis and measurement method and system for cassiterite mineral particles, specifically relating to the field of industrial vision analysis technology. It addresses the problem that existing methods struggle to accurately distinguish particle motion states and separate motion trails in dynamic flow fields, leading to inaccurate segmentation, tracking, and parameter measurement. By acquiring continuous frame images and extracting flow velocity characteristics of the flow field region, the method determines cross-frame displacement trends based on the overlap of neighboring pixel grayscale distribution and flow velocity characteristics. It extracts the stability of candidate particle contour morphology and performs matching verification. For matched particles, it constructs a flow velocity field constraint range and obtains a motion vector field. Combining the motion vector amplitude corresponding to the cassiterite mineral density, it uses density clustering to distinguish cross-frame displacement from independent multi-particle motion. High-speed moving particles are selected, and inter-frame grayscale gradient difference is used to separate motion trails from the true contour. Based on the motion type and the true contour, it completes target segmentation, contour tracking, and parameter measurement, improving the accuracy of particle analysis under dynamic conditions.

This invention relates to the field of optical component inspection technology, specifically disclosing a method and system for detecting prism manufacturing defects based on contour feature matching. The method includes: acquiring a surface image of the target prism, sequentially performing grayscale and binarization segmentation to obtain a binarized prism image; extracting the original contour point set through contour tracking; applying multi-level Gaussian filtering to the point set to calculate curvature characterization values, thereby segmenting the original contour point set into a straight line segment contour subset and a corner point contour subset; establishing straight line feature descriptors and corner point feature descriptors in parallel, and concatenating the features to obtain a contour feature vector; simultaneously acquiring the theoretical contour feature vector, comparing the differences between the two item by item to identify abnormal edge length regions and abnormal edge angle regions; finally, integrating the spatial locations of these two types of abnormal regions to generate a manufacturing defect distribution map, thereby marking the defect location and defect type of the target prism; this invention can improve the accuracy of prism manufacturing defect detection.

The embodiment of the invention provides an object contour tracking method, an active contour line model training method, equipment, a medium and a product, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring a to-be-processed video corresponding to a to-be-tracked object in a preset scene, cutting the to-be-processed video to obtain a plurality of video frames, determining initial contour data of the to-be-tracked object, inputting the initial contour data into a pre-trained active contour line model corresponding to the preset scene to obtain result contour data, and for the second video frame and each subsequent video frame, inputting the result contour data corresponding to the previous video frame into a pre-trained active contour line model to obtain corresponding result contour data, and superposing the result contour data corresponding to all the video frames to obtain the result contour data corresponding to all the video frames. And obtaining corresponding dynamic tracking contour data of the to-be-tracked object in the preset scene. According to the method, the tracking accuracy of the object contour can be improved.

The present application relates to the field of metal 3D printing and machine vision detection, specifically to a kind of metal 3D printing defect vision detection system, comprising: image acquisition module: in dynamic processing process, the original image frame sequence of target area is collected;Characteristic evaluation module: extract features to generate basic feature map;And quantitatively analyze to generate illusion entropy representing the probability of pseudo feature generation;Strategy scheduling module: obtain the occupancy rate of computing resource;Judge the relationship between illusion entropy and occupancy rate and preset threshold;To generate downgrade or normal processing instruction;Image processing module: respond to normal instruction to generate first feature data for global pixel-level segmentation;Respond to downgrade instruction to generate second feature data for local morphological contour tracking;And generate surface feature anomaly detection result;The present application reduces the missed detection rate caused by continuous frame drop, improves the robustness of the system in the manufacturing scene.

The invention discloses an MES automatic management system based on face recognition login authentication, and relates to the technical field of face recognition. An environment darkness analysis module is used for judging illumination conditions, and if the illumination conditions are dark, personnel passing analysis is started; then face recognition login is carried out; and if the environment is bright, directly executing face recognition. And finally, returning to an initial state to form closed-loop management. According to the invention, through dual judgment of environment darkness and personnel behaviors, accurate triggering of light supplement is realized, login residence and transient passing are effectively distinguished, and disturbance and energy consumption are avoided; through dynamic contour tracking and staying analysis, fuzzy behaviors are converted into reliable instructions, and the authentication accuracy is improved; safety is guaranteed, non-inductive login experience is optimized, and humanization and energy conservation are both considered; existing hardware is fully utilized, efficiency jump is achieved through algorithm upgrading, and cost is remarkably reduced.

Based on the transverse block of super large grid fast vectorization method, through the block, in turn, each block contour tracking; and based on the contour node counterclockwise trend characteristics, using the contour line turn boundary line method contour point sequence into the contour surface; for not contact with the interface line contour surface directly output to the vector file, and the remaining other with the block boundary contact face will be with the last block with the interface line contact face merging. Mainly through a kind of end point sorting after pairing, link segment line, again merging the idea of surface, to realize the adjacent face merging between blocks; after merging two blocks, still in the block lower boundary face, save to the face cache, will be used for the next block face merging operation; through the loop iteration, repeatedly on each block contour tracking-merge adjacent face operation, ultimately realize the whole grid vectorization effect, solve the problem of direct large grid vectorization consume a lot of memory.

The application discloses a cross-section map real-time synchronous display method and system based on GIS, relates to the technical field of geological modeling, and comprises the following steps: importing multi-source heterogeneous geographic data, performing standardized pretreatment, and generating a unified format geographic information dataset; performing multi-dimensional feature extraction on the geographic information dataset by using a feature-level fusion algorithm to construct a multi-dimensional geographic feature model fusing geological structure, terrain elevation and surface cover information; performing data optimization on the multi-dimensional geographic feature model based on a spatial interpolation algorithm to generate a continuous geographic surface model; and performing parallel contour tracing and topological sorting on a discrete cross-section line segment set to generate a self-defined cross-section map supporting interactive parameter adjustment in real time.

The invention discloses an AI intelligence-based traditional Chinese medicine auxiliary diagnosis and treatment system, and relates to the technical field of intelligent medical treatment, and the method comprises the steps: firstly, processing a tongue coating image through image segmentation and contour tracking, and obtaining a tongue coating contour containing color and coordinate data; analyzing the color and the shape by using a saliva interference model, and judging whether image restoration is needed or not; and finally, the AI diagnosis module performs intelligent diagnosis on the processed tongue coating contour. According to the method, the tongue coating saliva interference model is constructed, color overexposure and feature point position abnormity are comprehensively evaluated, image distortion caused by saliva reflection or a liquid film is effectively recognized, interference detection precision is improved, cleaner and more reliable image data are provided for AI diagnosis, analysis accuracy is enhanced, and through a dynamic monitoring and repairing mechanism, the diagnosis accuracy is improved. The tongue coating state is continuously evaluated in the time window, the contour with the minimum interference is locked through the sorting algorithm, misjudgment is avoided, it is ensured that AI is based on optimal image analysis, and the clinical practicability and stability of the system are improved.