8866 results about "Visual perception" patented technology

The present application provides a method for diagnosing and / or evaluating the presence or absence, severity or degree of the improvement of a retinal disease in a subject, which comprises determining and / or evaluating circulatory parameters, retinal function, retina morphology and / or visual relating quality of life (QOL).

The invention provides a method for displaying a hotspot keyword. The method comprises the steps of obtaining the hotspot keyword, generating a display body used for displaying the hotspot keyword, implanting the hotspot keyword into the display body, and controlling the display body with the implanted keyword to move towards the bottom end of a rotated web page in a gravity falling mode after the web page where the display body with the implanted keyword is located is rotated. The invention further provides a device and a browser for displaying the hotspot keyword. The display body is generated by obtaining the hotspot keyword, the hotspot keyword is implanted into the display body, and after the web page where the display body with the implanted keyword is rotated, the display body with the implanted keyword moves towards the bottom end of the rotated web page in the gravity falling mode. Therefore, dynamic and stereoscopic display of the hotspot keyword is achieved, the display effect is improved, the visual sense is enriched, and the interaction performance is high.

According to the multi-source information fusion robot positioning method and system for the unstructured environment provided by the invention, fusion processing is carried out on multi-source information data such as laser point cloud, environment image information and acceleration which are acquired in real time, so that the multi-source information fusion robot positioning method and system for the unstructured environment can be realized in a severe weather state and a dynamic unstructured environment. Carrying out high-robustness real-time positioning under the condition of violent illumination change; in a geometrically degraded roadway area, registration positioning of laser point cloud emission fails due to lack of enough external characteristics, and at the moment, the visual positioning module can still work normally and complete robot repositioning by detecting matched image information; in an environment lacking enough illumination, the laser point cloud and acceleration information can still provide enough positioning fusion data input for the computing unit.

An apparatus and method for automatically updating a visual element on a graphical user interface. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor. The memory instructs the processor to generate a graphical user interface comprising a plurality of visual elements, wherein the plurality of visual elements comprises a plurality of interactive components, wherein each of the plurality of interactive components comprises an interaction event handler, display at least a command corresponding to an interactive component, detect, using an interaction event handler of a plurality of interaction event handlers, input corresponding to a reference datum of the at least a command, determine, using a tracking module, interaction data corresponding to the input, and generate a modified graphical user interface.

System, apparatus, and computer program products are disclosed for protected display of a shared visual data object for an audiovisual conference session. In one aspect, a method includes receiving the shared visual data object, wherein the shared visual data object comprises visual data associated with one or more shared user interfaces; determining, using user interface generation circuitry and based on the shared visual data object, one or more protected data values associated with one or more protected data fields; generating, using the user interface generation circuitry, a masked visual data object based on each masked representation and each spatial coordinate set; and causing, using the user interface generation circuitry and / or the communications hardware, the masked visual data object to be displayed using the audiovisual conference session.

The present application provides a kind of straw coverage detection method and device of cross-scale multi-source data fusion, applied to remote sensing image processing technical field, above-mentioned method includes: the ground photograph is input to visual big model and is carried out semantic segmentation, obtains the straw coverage of corresponding ground scale range;Based on the straw coverage of ground scale range, the scale range adaptation is carried out to unmanned aerial vehicle image, and unmanned aerial vehicle sample image is obtained;Based on the image similarity calculation of unmanned aerial vehicle image block corresponding ground photograph and unmanned aerial vehicle sample image, the overall similarity of each unmanned aerial vehicle image block is obtained;Select a fixed proportion of target unmanned aerial vehicle image block to generate new sample;Based on unmanned aerial vehicle image straw coverage sample set, construct straw coverage regression model;Satellite remote sensing image is input to straw coverage regression model, and the straw coverage extraction result of target area is obtained;Through the present application, the straw coverage of large area range can be realized accurate extraction.

The utility model belongs to screening device technical field discloses a kind of machine vision industrial raw material screening devices with adjustable light source, including the conveying belt for conveying industrial raw material, the conveying belt is equipped with two groups of rotating mechanism, two groups of rotating mechanism are adjacent installation, the working end of two groups the rotating mechanism is equipped with adjusting mechanism, the output end of one the adjusting mechanism is equipped with light source, the output end of another the adjusting mechanism is equipped with camera, the utility model can adjust the position of light source and camera by rotating mechanism, and then ensure that light source does not cause influence to the data acquisition of camera under specific condition;By parameter and program control, realize the adjustment of automation, and the efficiency and detection accuracy of full-automatic detection are guaranteed.

This invention discloses a method for accurate identification of the moment of an accident in a new energy vehicle based on multimodal data fusion. The method acquires vehicle motion sensor data, dynamic parameters, and visual image data, and then performs the following parallel processing: The physical layer processes the data to obtain motion features and compares them with preset trigger conditions; when a suspected collision is detected, it outputs the physical layer trigger time and acceleration amplitude; the dynamic layer processes the data based on the deviation between the vehicle's dynamic model and its actual motion state, outputting a vehicle trajectory status label; the visual layer processes the data to detect targets and analyze their motion state; when visual evidence of a collision is detected, it outputs the visual layer trigger time and the type of collision object; the fusion layer performs spatiotemporal alignment of the three layers' outputs and makes a comprehensive judgment, outputting the fused accurate collision trigger time and accident scene classification information. This invention effectively filters false alarms and improves the detection rate of low-speed accidents through multimodal fusion, achieving accurate identification of the moment of an accident and scene reconstruction.

This invention discloses a multimodal acupoint targeted stimulation and feedback rehabilitation device. The device integrates focused ultrasound, transcutaneous electrical stimulation, and thermotherapy modules to form a composite stimulation unit. It monitors the physiological response of acupoints in real time using bioimpedance spectroscopy and PPG optical sensors, and calculates the "Deqi Index" (DI) based on a weighted fusion of normalized impedance decrease rate and blood flow increase rate. It combines RGB-D visual dynamic tracking of acupoint coordinates and provides location confirmation through a wearable force feedback device. The central control unit adaptively adjusts various stimulation parameters based on the DI value, forming a closed-loop control of perception-decision-execution. This invention solves the problems of existing devices having a single stimulation modality, lacking objective feedback, and inaccurate positioning, significantly improving the accuracy, effectiveness, and intelligence level of acupoint rehabilitation therapy.

The application provides an intelligent visual detection method and device for component defects, which comprises the following steps: loading a light standard three-dimensional model corresponding to a target component from a backend server; synchronously collecting real-time image data and depth data of the component on a conveying mechanism or a clamping mechanism; generating a real-time point cloud model for describing surface geometric features of the component based on the real-time image data and the depth data; matching the real-time point cloud model with the light standard three-dimensional model to locate a point area of interest in a high-defect area of the component; spatially mapping and fusing texture features in the real-time image data and the depth data in the point area of interest; and identifying defects in the point area of interest of the component through a joint probability model. The application can realize high-speed and accurate defect detection of components without complex feature engineering, and greatly reduces the maintenance difficulty and implementation cost of the algorithm.

The application relates to the technical field of image processing, in particular to a two-dimensional code re-encoding recognition detection method based on machine vision. The method comprises the following steps: acquiring ink jet pressure, electrostatic intensity and vibration data of a code spraying device in a code spraying process; analyzing an ink jet pressure fluctuation time domain curve to determine a pressure influence coefficient; analyzing an electrostatic intensity fluctuation time domain curve to determine an electrostatic influence coefficient; analyzing a vibration time domain curve to determine a vibration influence coefficient; calculating a code spraying risk characteristic value; determining an abnormal tendency of the code spraying process; locating a re-encoding area with a strong abnormal tendency; comparing real-time material winding speed in the code spraying process with a preset shooting speed to determine a re-encoding detection synchronization value; analyzing code spraying environment data to determine illumination uniformity; combining the code spraying risk characteristic value to calculate a re-encoding specific characteristic value; determining a code spraying re-encoding state; and executing alarm output and / or removing re-encoding content. The application comprehensively analyzes the code spraying process and the detection process, recognizes code spraying re-encoding, and guarantees the consistency of two-dimensional codes.

A fibre-reinforced polymer composite component 2 formed from a polymer matrix material containing fibre reinforcement. The component 2 comprises a main portion 4 and at least one raised feature 6 extending from a surface plane s of said main portion 4. The at least one raised feature 6 is arranged to incur visually perceptible damage when the component 2 is subject to an impact with an energy above a predetermined impact energy threshold and to resist an impact with an energy below the predetermined impact energy threshold.

The present application belongs to the technical field of industrial vision detection, and discloses a kind of copper foil adhesive tape bubble and pinhole defect multi-light source detection method;In the process of copper foil adhesive tape transmission, through the time division multiplexing illumination of coaxial light source and multi-azimuth oblique light source, combined with the line scanning acquisition technology of transmission speed synchronization to obtain multi-lighting mode image set.Process the image distortion caused by high-speed transmission using motion blur compensation technology, and execute local contrast anomaly detection on motion compensation image to extract candidate defect area.Through calculating radial morphological features and stereoscopic protrusion discrimination features, accurate discrimination of bubble and pinhole defects is realized.The present application has self-adaptive optimization capability, can dynamically adjust light source parameters and reacquire for the area with low discrimination confidence, and output defect position, type and severity level.The present application overcomes the influence of motion blur, improves the accuracy of defect classification, and provides accurate basis for production process parameter adjustment.

This invention discloses a zero-shot embodied target navigation method and system based on a prior semantic map. The method utilizes scene prior data and infers the probability of a target object's presence in different room types using a large language model to construct a prior value map. During navigation, based on the currently observed image, a visual language model is used to calculate the similarity score between the observation and the semantic description of the target object, generating an observation value map. The prior value map and the observation value map are dynamically fused, with the prior value weights adaptively adjusted as observation information accumulates, resulting in a fused value map. Based on the fused value map, room-level semantic prior points and frontier points are evaluated, executing a two-layer search navigation that alternates between upper-level room guidance and lower-level frontier exploration. When a target object is detected, the system switches to target navigation. This invention, through value fusion and a hierarchical navigation architecture, effectively utilizes environmental prior semantic information, improving the efficiency and stability of target search under zero-shot conditions.

This invention discloses an efficient 2D dynamic soft shadow rendering method and system based on symbolic distance fields, belonging to the field of computer graphics and image processing technology. It includes extracting displacement vectors to determine active occlusions and generating bounding boxes to be updated; downsampling and reconstructing within the bounding boxes and concatenating with an external buffer to generate a low-resolution symbolic distance field; calculating a local original resolution symbolic distance field within the same region; calculating and normalizing the difference matrices of the two types of distance fields to generate a fusion weight matrix; and using the weight matrix to weightedly superimpose the two types of distance fields, performing ray propagation to output shadow textures. This invention uses bounding boxes established based on inter-frame displacement vectors to lock dynamic regions and constructs pixel-by-pixel fusion weights through multi-level resolution distance difference matrices to achieve smooth superposition of heterogeneous field data. This significantly reduces scene reconstruction overhead while eliminating visual seams caused by incremental updates, outputting high-quality and spatiotemporally continuous 2D soft shadow effects.

The application relates to the technical field of robots, and discloses an AI vision positioning method and system for robot automatic assembly, which comprises a composite perception system, an active physical excitation module and a central processing unit; the composite perception system is integrated with a polarization camera, a thermal imager and a structured light projector; and the central processing unit is internally arranged with a physical causal decoupling network and a health state manifold model. The method comprises the following steps: in an offline stage, a physical characteristic enhanced point cloud of a standard part is collected through a detection, excitation and response protocol, so as to train the physical causal decoupling network and construct a health state manifold representing the state of qualified materials; in an online stage, the state of a material to be assembled is diagnosed, and an expected process path connecting the current state and the target state is generated on the manifold. The application solves the positioning and control problems in complex scenes such as high reflectivity and transparency, and realizes high-precision in-situ closed-loop assembly.

The application provides a display device and a subtitle language recognition method. The method can acquire a subtitle corresponding to a media video in response to a play instruction of the media video, extract visual texture features of the subtitle in a case where the subtitle is an image subtitle, perform language recognition according to the visual texture features to obtain a language recognition result, extract syntax topology features of the subtitle in a case where the subtitle is a text subtitle, perform language recognition according to the syntax topology features to obtain a language recognition result, write the language recognition result into metadata corresponding to the media video, control a display to play the media video and the subtitle, and display a language identifier corresponding to the subtitle according to the language recognition result in the metadata corresponding to the media video. The method recognizes a language by analyzing features of different languages in physical forms and syntax structures, so that the language identifier of the subtitle can be correctly displayed when the media video is played.

This invention discloses a mixed reality data processing and interactive response method, apparatus, and system. The method includes: acquiring corresponding spatial feature points in VR and AR environments; optimizing the solution of rigid body transformation matrices to achieve sub-millimeter-level spatial alignment; continuously monitoring alignment errors during virtual-real fusion rendering, triggering a repositioning process when errors exceed a threshold; evaluating user operations through multimodal data fusion and generating real-time AR correction guidance; and dynamically adjusting rendering parameters and resource preloading based on visual attention focus to ensure end-to-end latency and interactive feedback latency are both below set thresholds. The apparatus includes modules for feature point acquisition, spatial alignment calculation, error monitoring and repositioning control, multimodal data interface, and real-time rendering control. The system includes a server and a user interaction terminal, supporting intelligent training and adaptive interactive optimization for virtual-real fusion. This application addresses problems such as low virtual-real spatial alignment accuracy, high interactive latency, and insufficient personalized adaptation.

This application provides a multi-source fusion positioning method, electronic device, and vehicle. The method includes: acquiring multi-source perception data and map curve parameters required by the vehicle positioning system during curve driving; quantitatively monitoring the health status of the vehicle positioning system based on the multi-source perception data and map curve parameters to obtain a visual quality score, an inertial measurement unit (IMU) excitation score, and a tightly coupled residual ratio in the curve scenario; weightedly fusing the visual quality score, the IMU excitation score, and the tightly coupled residual ratio to obtain a comprehensive health index of the vehicle positioning system under curve conditions; and automatically matching and executing a corresponding level of degradation strategy based on the comprehensive health index, outputting the vehicle pose including uncertainty measurement, the current system health level, and graded warning information. This application improves the continuity, positioning reliability, and safety of intelligent driving systems.

We provide the system. [Solution] A means of recognizing road signs by analyzing images acquired in real time from an imaging device mounted on a vehicle, A means for generating and providing voice notifications to the operator based on recognized signs, A means for accumulating individual flight data and analyzing flight tendencies based on that data, A means for generating improvement guidelines for the pilot based on the analyzed piloting tendencies, A means of optimizing insurance premiums based on piloting skills by linking piloting data with an external device, A means of providing visual notification via a display device installed inside the vehicle, A system that includes this.

This invention provides a visual positioning method and system, belonging to the field of visual positioning technology. The method includes: extracting query feature vectors from a query image to be located; obtaining candidate images and their corresponding candidate geographic coordinates based on the similarity between the query feature vectors and reference feature vectors in a reference database; calculating the spatial geographic distance between the candidate geographic coordinates; performing density spatial clustering on the candidate geographic coordinates based on the spatial geographic distance to obtain spatial clusters; determining the target spatial cluster containing the most candidate geographic coordinates; and calculating the geometric center of the candidate geographic coordinates within the target spatial cluster, using the center coordinates as the positioning coordinates of the query image. This invention introduces a physical scale to cluster and vote on the retrieval results, removing mismatch noise caused by homogeneous appearances in different locations, overcoming the jump error caused by relying solely on single visual similarity, and achieving high-precision UAV visual positioning with spatial consistency.

A system and method for high-confidence error overbounding of multiple optical pose solutions receives a set of candidate correspondences between 2D image features captured by an aircraft camera (102) and 3D constellation features including at least one ambiguous correspondence. A candidate estimate of the optical pose of the camera (102) is determined for each of a set of candidate correspondence maps (CMAP), each CMAP resolving the ambiguities differently. Each candidate pose estimate is evaluated for viability and any non-viable estimates eliminated. An individual error bound is determined for each viable candidate pose estimate and CMAP, and based on the set of individual error bounds a multiple-pose containment error bound is determined, bounding with high confidence the set of candidate CMAPs and multiple pose estimates where at least one is correct. The containment error bound may be evaluated for accuracy as required for flight operations performed by aircraft-based instruments and systems.

The present invention discloses a method and system for intelligent control of a cookware production line based on finished product inspection, relating to the technical field of production line control. It includes: constructing a cookware positioning model based on abnormality feature information from each section of the cookware production line, wherein the abnormality feature information is obtained from abnormality image data collected by visual machines, and combined with operating condition data of control equipment to locate the conveying section where the abnormal cookware is situated; generating control instructions based on the operating condition data output by the cookware positioning model and the abnormality feature information. In scenarios where multiple types of control equipment participate in the same processing chain, the present invention achieves cross-equipment action mapping through a collaborative control mechanism, automatically calculates adaptive parameter quantities for downstream control equipment to realize dynamic linkage, such that the action chain no longer relies on static settings but adjusts automatically according to the actual abnormality severity, significantly reducing error accumulation caused by segmented processing and improving the uniformity and stability of the overall processing effect.

A system and method for assessing gaze stability and vestibular function using a mobile device are disclosed. The system includes a mobile application configured to execute gaze-stability testing protocols, a display module for presenting visual targets, a head-movement and eye-tracking module utilizing real-time video captured via the device's camera, a speech-recognition module for processing verbal responses, and a data-processing module to analyze head-movement and visual-acuity data. The system enables remote patient assessments and includes protocols such as static visual acuity, visual processing, and mobile gaze stabilization tests to evaluate metrics like peak head velocity and visual acuity. Results can be processed in real-time and can be securely transmitted to clinicians for remote evaluation. The disclosed system provides a cost-effective, user-friendly telehealth solution for vestibular function assessment, eliminating the need for specialized equipment or clinical visits.

This invention relates to the field of decorative engineering technology, and discloses a method and system for generating virtual renderings of decorative engineering projects. The method includes: S1, parsing CAD drawings; S2, identifying professional intent; S3, defining and configuring key aesthetic elements; S4, attributing and correcting model parameters; S5, conflict detection; S6, adaptive fusion and correction; S7, material mapping and lighting calculation; and S8, outputting the virtual rendering. This invention solves the problem that the complexity of CAD drawings leads to incomplete or inaccurate geometric structures in 3D models, which in turn affects material mapping and lighting calculation, thus impacting the visual realism of virtual renderings.

This invention discloses a progressive domain-adaptive semantic segmentation method for low-light environments in underground mines, belonging to the fields of computer vision and artificial intelligence. It employs a geometrically adapted model by pre-training and fine-tuning a semantic segmentation model using source domain labeled images for prior alignment. An unsupervised illumination style mapping is used to construct and train a cyclic consistent generative adversarial network (GAN). This network utilizes the mapping between intermediate bright-light image sets and target-domain dark-light image sets to convert intermediate domain labeled bright-light images into synthetic dark-light images, forming a synthetic dark-light dataset. Uncertainty-aware dark-light self-training is performed, initialized with the geometrically adapted model, on a hybrid dataset consisting of the synthetic dark-light dataset and real unlabeled target-domain dark-light images, resulting in a semantic segmentation model adapted to the target dark-light scene. This invention achieves stable and efficient model transfer from bright source domains to dark target domains, improving the semantic segmentation accuracy and robustness of the model in low-light environments in underground mines.