10 results about "Point localization" patented technology

This invention relates to the field of measurement data analysis technology, specifically to an engineering measurement data analysis method and system based on image recognition. The method includes the following steps: using an industrial camera to identify structural edges, measurement markers, and calibration rulers to form a reference set of measurement point coordinates; establishing corresponding quantities of directional scales based on the calibration ruler direction; combining the displacement relationship of the measurement points to form a measurement point interval characterization set; determining the edge deviation reference rate; and finally forming a boundary positioning association result. This invention uses grayscale abrupt changes to define edges and construct measurement point positioning references; calculates the lateral and longitudinal displacements of the calibration ruler to extract corresponding directional scales; combines measurement point displacement and edge direction to form interval characterizations; analyzes the normal span and surface deviation state to form edge references; and combines the boundary extension orientation to define the associated segments of the arrangement trajectory. This reduces interference from manual interpretation, maintains consistency of multi-directional conversion benchmarks, controls the error accumulation process, and improves the coherence and stability of engineering monitoring data processing.

This invention belongs to the field of intelligent medical image recognition technology, specifically relating to a method and system for locating key points in hand bone X-ray medical images. The method includes the following steps: S1, collecting and preprocessing hand bone X-ray images and corresponding key point annotation data; S2, constructing a key point localization model for hand bone X-ray medical images; S3, training and validating the model using an auxiliary loss function; S4, performing end-to-end prediction using the validated model. This invention features high localization accuracy and the ability to locate a sufficient number of hand bone key points at once.

Example aspects of the present disclosure are directed to computing systems and methods for hand tracking using a machine-learned system for palm detection and key-point localization of hand landmarks. In particular, example aspects of the present disclosure are directed to a multi-model hand tracking system that performs both palm detection and hand landmark detection. Given a sequence of image frames, for example, the hand tracking system can detect one or more palms depicted in each image frame. For each palm detected within an image frame, the machine-learned system can determine a plurality of hand landmark positions of a hand associated with the palm. The system can perform key-point localization to determine precise three-dimensional coordinates for the hand landmark positions. In this manner, the machine-learned system can accurately track a hand depicted in the sequence of images using the precise three-dimensional coordinates for the hand landmark positions.

This invention relates to the field of image feature recognition technology, specifically to a method and system for locating key surgical points in orthopedic surgery based on anatomical structure feature recognition. The method includes the following steps: constructing a prior skeletal topology map representing biomechanical connectivity based on human anatomical standards; utilizing a graph attention mechanism to perform semantic mapping and feature aggregation between the multi-scale anatomical feature map and the prior skeletal topology map to generate node-enhanced feature vectors; and performing spatial coordinate regression calculations based on these node-enhanced feature vectors to output the coordinates of the surgical key points. In this invention, by introducing prior anatomical topological knowledge as a global constraint for feature extraction, the information transmission mechanism of the graph structure effectively captures the geometric dependencies between key skeletal points. This solves the problem of local feature ambiguity caused by traditional methods in complex pathological scenarios such as fracture displacement, bone hyperplasia, or tissue occlusion, achieving high-precision and robust surgical key point localization.

The invention provides a human body key point detection method and device, electronic equipment and a storage medium, and relates to the technical field of computer vision, key point features of different granularities in a to-be-detected human body image are extracted through a plurality of key point decoding layers, the performance and speed of human body key point regression can be optimized in a coarse-to-fine mode, and the accuracy of human body key point regression is improved. And the human body key point positioning accuracy is improved. Each key point decoding layer applies the bottom-layer semantic features and the position coding features of the bottom-layer semantic features, and a large amount of position guidance information can be provided for human body key point positioning, so that the regression difficulty of the human body key points is reduced. The key point detection model introduced through the method not only can improve the positioning accuracy and positioning precision of the human body key points, but also can avoid the problems of high calculation cost and low detection speed caused by a thermodynamic diagram method.

The invention relates to the technical field of medical image processing, in particular to a target point positioning method and device based on a medical image, equipment and a medium. According to the method, a target point is manually marked on a three-dimensional cross-sectional image, then the probability that each voxel coordinate on the three-dimensional cross-sectional image is possibly the target point is predicted on the three-dimensional cross-sectional image by adopting an identification positioning algorithm, and the target point is identified according to the marked target point coordinate and the prediction probability corresponding to each voxel coordinate. And screening the final position of the target point from each voxel coordinate. According to the method, manual labeling and an automatic identification positioning algorithm are combined, so that the positioning instability caused by image quality in automatic identification positioning can be made up, and the positioning stability and accuracy of the target point are improved.

This application relates to the field of computer vision technology and discloses a method and system for detecting pain expressions. The method includes: acquiring a facial video stream, segmenting muscle regions and extracting texture features after facial key point localization and inter-frame alignment; calculating motion energy based on these features, extracting enhanced micro-expression temporal segments and multi-scale spatiotemporal features, constructing a muscle dynamics model and completing state estimation; generating an adaptive candidate spatiotemporal window set through spatiotemporal attention weighted fusion; performing temporal modeling for dynamic feature fusion, and selecting the optimal window by fusing micro-expression features; and outputting the final pain level after double consistency verification and loop closure optimization. This application overcomes the limitations of static images, accurately captures facial dynamics and temporal changes, strengthens feature correlation, and significantly improves detection accuracy.

The invention discloses a touch anti-interference algorithm scheme, which comprises the following steps of: designing and arranging M driving channels and N sensing channels, executing a scanning period, obtaining original signal intensity of all nodes, converting original signal data into differential data, traversing lines by combining a threshold value t, calculating and recording a line touch area of each line, and calculating the line touch area of each line according to the line touch area. The node signals with the same scanning time and similar scanning time are differentiated to form whole-screen differential signals, interference bottom noise is eliminated accordingly, and effective signals are reserved; the invention provides a method for calculating a point area according to a line touch area and a method for calculating point coordinates and point width and height information through a circumscribed rectangle, and provides a method for calculating point coordinates and point width and height information according to the point coordinates and the point width and height information. Therefore, various interferences are eliminated, the touch point positioning and contour information obtaining accuracy is remarkably improved, and the user control and writing experience is improved.

1. The name of the design product: positioning of the task approval information of the graphical user interface of the electronic device. 2. The use of the design product: an electronic device. 3. The design points of the design product: the part of the graphical user interface of the electronic device that requests local protection. 4. The picture or photo that best indicates the design points: design 1 front view. 5. The electronic devices of design 1-design 3 are common designs, and other views of design 1-design 3 are omitted. 6. Design 1 is designated as the basic design. 7. The use of the graphical user interface: the product interface of design 1-design 3 is an interactive interface for multi-dimensional data display of energy equipment; it is used for users to quickly locate the required auxiliary information and provides a convenient view mode; the interface of design 1 front view is an interactive interface for approval operation; in the interface of design 1 front view, after clicking the auxiliary information tab, design 1 interface change state diagram 1 is displayed; in the interface of design 1 interface change state diagram 1, after selecting the anchor point to the corresponding auxiliary information module (such as selecting the approval record anchor point), design 1 interface change state diagram 2 is displayed; the interface of design 2 front view is an interactive interface for approval operation; in the interface of design 2 front view, after clicking the to-be-confirmed X item button, the to-be-confirmed artificial inspection item is viewed, and design 2 interface change state diagram is displayed; the interface of design 3 front view is an interactive interface for approval operation; in the interface of design 3 front view, after clicking the permanent mode, design 3 interface change state diagram is displayed; the red coated part is not the part of the design that requests protection.

The application aims at the field of target counting and target positioning in remote sensing images, and discloses a remote sensing image multi-class target counting method based on point positioning. First, a feature extractor is used to extract features in the remote sensing image, then a scale adaptive feature enhancement module is used to complete rough prediction of the target scale in the remote sensing image, and the image features are fused and enhanced, next, the enhanced image features are used to predict the position, category, confidence and fine scale of the target in the image by using a prediction head, and the corresponding candidate target of the remote sensing image is generated. The candidate target contains repeated prediction of the same target, so it cannot be directly used as the prediction result of multi-class target counting and positioning. By using a non-maximum suppression post-processing means for the candidate target, the false prediction target and the repeated prediction target in the candidate target can be suppressed. The application improves the accuracy of network multi-class target counting and positioning.