111 results about "Image tracking" patented technology

Disclosed are approaches that may provide image-guidance to interventionalists by providing true 3D visualization and quantitative feedback in real-time. A guidance system may allow a physician to manipulate a medical device and see a 3D rendering with quantitative feedback floating in mixed reality, next to standard monitors. Image tracking may detect and co-register the medical device's 3D position using, for example, bi-plane C-arm X-ray fluoroscopy and provide a 3D trajectory as quantitative feedback. Patterns in a fluoroscopic image may be used to accurately determine an object's z-position from a single angle projection.

Systems and methods for tracking a location of an object of interest through a sequence of medical images are provided. First and second input medical images of a patient are received. The first input medical image comprises an annotation of a location of an object of interest. Features are extracted from the first and the second input medical images. A location of the object of interest in the second input medical image is determined using a machine learning based location predictor network based on the annotation of the location of the object of interest in the first input medical image and the extracted features from the first and the second input medical images. The location of the object of interest in the second input medical image is output. The machine learning based location predictor network is trained based on a comparison between 1) locations of a particular object in a sequence of training images determined during a forward tracking of the particular object through the sequence of training images and 2) locations of the particular object determined during a backward tracking of the particular object through the sequence of training images.

The present invention provides a method for quantifying moving heterologous nanoparticles in a suspension by imaging. The quantifying method comprises acquiring at least one z-stack of images within the suspension; tracking the nanoparticles in the images to identify unique nanoparticles; and enumerating the unique nanoparticles. Also provided is a method for characterizing size distribution of moving heterologous nanoparticles in a suspension by imaging. The characterization method comprises acquiring time lapse images; tracking the nanoparticles in the images to identify unique nanoparticles; determining the locations of each of the unique particles; determining the size of each of the unique nanoparticles; and aggregating the sizes of the unique nanoparticles.

The invention discloses a non-motor vehicle passenger-vehicle identification method and system in a crowded scene, and belongs to the technical field of intelligent traffic management. In order to solve the problem of accurate recognition of non-motor vehicles, persons and vehicles in a crowded scene, image data of a traffic intersection is collected through a traffic camera, target detection processing is performed on the image data by using a target detection model, and a person target and a non-motor vehicle target are detected to obtain a person target frame and a non-motor vehicle target frame; judging whether the scene category to which the image data of the traffic intersection belongs is a crowded scene or not to obtain crowded scene image data; personnel key points and non-motor vehicle key points in the crowded scene image data are extracted; associating the personnel key points with the non-motor vehicle key points to determine personnel key points associated with the non-motor vehicle key points to obtain a personnel-vehicle association target; and identifying whether the person-vehicle associated target has traffic violation behaviors or not through an image tracking technology, and if so, identifying the identity information of the person target for traffic violation early warning and processing.

The invention relates to the technical field of image tracking, in particular to an abnormal person tracking method and system based on a face image, and the method comprises the steps: carrying out the static identity feature extraction and dynamic behavior feature extraction of a stable face feature basic image, and obtaining a static identity feature set and a dynamic behavior feature set; performing space-time alignment processing on the static identity feature set and the dynamic behavior feature set to obtain a space-time aligned feature set; performing feature decoupling processing based on the space-time alignment feature set to obtain independent identity features and independent behavior features; and carrying out abnormal mode identification on the independent behavior characteristics to obtain an abnormal behavior type classification result. According to the invention, cross-scene and cross-time-frame personnel tracking can be realized by combining the static identity features and the abnormal behavior classification result; an abnormal person moving track is drawn according to the timestamp and the space coordinate of the abnormal behavior, track coherence is improved based on continuity scores, and tracking interruption caused by view angle switching or time intervals is avoided.

The invention relates to a robot path planning method and system based on optical vision. The robot path planning method comprises the following steps: acquiring an environment image of an environment where a robot is located and robot motion information; based on the environment image, a dense three-dimensional point cloud map is obtained through contrast enhancement, signal extraction, decoding and interpolation optimization; based on the dense three-dimensional point cloud map and the color RGB visible light environment image, generating a three-dimensional point cloud map with semantic tags through background elimination, image tracking, semantic recognition and clustering; estimating the motion state of the robot based on the three-dimensional point cloud map with the semantic tag and the motion information of the robot, carrying out global and local path planning, and obtaining a local path through track smoothing; and converting the local path into the actual action of the robot in combination with the motion state of the robot. According to the invention, through the collaborative design of the active optical visual perception architecture and multiple algorithms, the balance between cost control and performance improvement is effectively realized.

The invention belongs to the technical field of intelligent traffic, particularly relates to a tunnel global tracking method based on an old camera re-identification technology, and aims to accurately realize global tracking and trajectory generation of a tunnel traffic target in a low-cost manner. The method comprises the steps of receiving an RSTP data stream, processing the RSTP data stream through a detection algorithm to obtain a sensing result, and feeding back the sensing result to the image tracking module. And processing a time sequence sensing result, adding vehicle speed information and a vehicle ID, and generating a tracking result including the vehicle ID, a target frame, a vehicle speed, a vehicle type, a color, a timestamp, a base station number and a camera number. On the basis of vehicle pixel picture data, upstream sample library candidates are matched for a current point position sample library target, if matching succeeds, the target inherits the global ID, if matching fails, the point position ID is reserved, and the target is output to a downstream point position and a simulation platform after the sample library is updated. And the trajectory generation post-processing module constructs a target matching relationship between adjacent camera sections through a frame insertion mode according to a feature re-identification result, and generates a tunnel global complete trajectory chain.

This application proposes a human body recognition system management method, apparatus, storage medium, and electronic device. First, the human body registration identity of the corresponding user is determined based on the target facial image. If no corresponding human body registration image exists for the registered identity, any target human body image from the target human body image tracking set is selected as the corresponding human body registration image. By automatically using any target human body image from the human body image tracking set as the human body registration image when no corresponding human body registration image exists, this overcomes the technical problem of requiring the user to actively provide a human body registration image, reduces user intervention, saves manpower and time, and is more convenient.

The present disclosure provides a method and device for tracking a moving object, a computer-readable medium, and an electronic device, and relates to the field of computer technology. The method comprises: obtaining an original video frame, which is acquired by a real camera with a fixed shooting angle; determining a moving object in the original video frame; creating a virtual camera based on the attribute parameters of the real camera and the position of the moving object in the original video frame; and cropping the original video frame based on the virtual camera to obtain a target video frame in which the image follows the movement of the moving object. The present disclosure can complete the image tracking of a moving object using only a real camera with a fixed shooting angle, achieving an automatic camera movement effect, and does not require manual intervention or the participation of a camera gimbal or motor, effectively reducing labor costs and hardware costs.

The invention discloses an adaptive multispectral target image tracking system and method. The system comprises an input layer, a processing layer and a transmission layer, the input layer is used for acquiring at least two paths of video signals; the processing layer is used for performing signal conversion, intelligent identification and fusion output on the video signals; and the transmission layer is used for respectively executing local video transmission and network RTSP stream distribution on the fused and output video signals. Decoding and format conversion are carried out on an access layer through a GS2971 / GM7150 chip, so that the RV1126 chip can receive different types of videos; in a processing layer, RKMedia is adopted to schedule video streams, YOLOv5 + NPU is adopted to perform intelligent identification, and a fusion module is adopted to perform picture enhancement; and in the output layer, low delay is ensured through local SDI transmission, and remote access is realized through RTSP flow. The system is compatible with complex scenes by adopting a layered decoupling thought.

The application provides an image tracking and collecting method, an imaging system and a computer device. The image tracking and collecting method comprises: initially collecting N pieces of image data of an imaging part, recording a position parameter of the imaging part in each piece of the image data, fitting a motion curve according to the position parameter, wherein N is greater than or equal to 3; predicting a predetermined position of the imaging part when collecting a next frame of the image data according to the motion curve; and controlling the movement of a ray generating device and a signal collecting device according to the predetermined position. Through the image tracking and collecting method, the system can predict the predetermined position of the imaging part when collecting the next frame of the image, and control the signal collecting device to move according to the predetermined position, so that the imaging part is always located in the signal receiving range of the signal collecting device. In the process of photographing the human body, human intervention is not required, the quality of the dynamic image is ensured, and the efficiency of the examination is improved.

The present embodiments provide systems and methods for tracking objects using refined bounding boxes and bounding box association. The system can include image sensors, a detector, a tracker, and a server.

The application provides an automatic delivery system of an interventional medical instrument based on image tracking, comprising: module M1: acquiring in-situ interventional medical instrument image information in real time, and acquiring instrument feature information based on the acquired interventional medical instrument image information; and module M2: delivering the interventional medical instrument based on the instrument feature information until the interventional medical instrument is delivered to a target position.

The present application relates to a human image tracking method and apparatus, and a non-volatile storage medium and an electronic device. The method comprises: acquiring a current image frame, and determining tracking information of the current image frame; determining a target face recognition region and a target head and shoulder recognition region from the current image frame; on the basis of a degree-of-overlap indicator between the target face recognition region and the target head and shoulder recognition region, and a first preset condition, respectively determining a first face recognition region and a first head and shoulder recognition region, and a second face recognition region and a second head and shoulder recognition region; acquiring a historical image frame corresponding to the current image frame; on the basis of a historical recognition region, determining a prediction region of a tracking object in the current image frame; matching a second region with the prediction region, so as to obtain a similarity indicator; and determining, from the second region, a target region with a similarity indicator between same and the prediction region meeting a second preset condition, and on the basis of the target region, updating tracking information of the prediction region.

Systems and methods for panoramic imaging in 2D and 3D ultrasound imaging are disclosed. The techniques disclosed herein use image registration to combine the 2D images into a 3D volume. Ultrasound images collected in different locations are combined using image tracking. When the transducer moves along a lateral direction (in line with a longitudinal axis of the transducer), a portion of the image is overlapped between frames and the overlapping frames can be patched. When the transducer moves along an elevational direction (non-parallel to the longitudinal axis of the transducer), there is no overlap between frames and the non-overlapping frames are not patched together but are stored side-by-side to create a volume. The storage and patching processes can be adaptive (e.g., the images are not stored or patched if there is minimal difference between neighboring frames).

The embodiment of the invention provides a target image tracking control method and system for a photoelectric turret, computer equipment and a medium, and belongs to the technical field of automatic control, and the method comprises the steps: S1, obtaining the altitude of a task area; s2, calculating an angular velocity compensation value; s3, selecting an interest target; s4, calculating an azimuth tracking miss distance and a pitching tracking miss distance; s5, calculating an initial angular velocity; s6, calculating the compensated angular velocity; and S7, the angular velocity difference value is input into a stable controller, the stable controller outputs a motor driving amount after receiving the angular velocity difference value, and a stable platform torque motor is driven to rotate. According to the method, the relative movement of the aircraft and the target is overcome, the visual axis translation caused by the linear displacement angular velocity in the flight state is counteracted, the target capturing and tracking difficulty in the flight is greatly reduced, the tracking performance of the target in the dynamic state is ensured, and the target capturing efficiency is improved. The method is small in calculation amount, low in cost, high in portability and suitable for being used on the photoelectric turret.

The invention provides an image acquisition system for tracking imaging of a rotor blade. The image acquisition system comprises a light source tracking system, a high-speed imaging system and a synchronous control system, the light source tracking system comprises a light source tracking high-precision turntable, a light source tracking reflector group, a light source tracking protective cover and an excitation light source; the high-speed imaging system comprises an imaging high-precision turntable, an imaging reflector group, a high-speed imaging protective cover, a high-speed camera and a coaxial light splitting device; the synchronous control system comprises a photoelectric coded disc, a delay signal emitter and an image tracking correction plate. The device is reasonable in structural design, all-phase 2D temperature and pressure dynamic measurement of the surface of the rotor blade can be realized by combining the pressure-sensitive (PSP) / temperature-sensitive (TSP) coating, and image blurring can be eliminated, the signal-to-noise ratio of the image can be improved and high-precision temperature and pressure measurement results can be obtained under the condition of long exposure time based on the light source and the image tracking system of the high-speed rotating mirror.

The present disclosure relates to systems, non-transitory computer-readable media, and methods for detecting and tracking edges in raster images using an advanced edge detection algorithm. For example, the disclosed systems generate a local histogram ranking of pixels within a sliding pixel window in a raster image, based on pixel values ​​located within the sliding pixel window. In some embodiments, the disclosed systems determine an edge thickness for an edge depicted in a region of the raster image enclosed by the sliding pixel window by comparing pixel ranks specified by the local histogram ranking of the pixels.In certain embodiments, the disclosed systems also provide the edge for display based on determining the edge thickness for the area of ​​the raster image enclosed by the sliding pixel window.

To accurately track an object to be tracked.SOLUTION: An image processing apparatus comprising: an acquisition unit configured to acquire images captured in time series and distance information in a depth direction in a plurality of regions of the images; a detection unit configured to detect a candidate region of an object to be tracked from the images based on image features of the images; With respect to the candidate area detected from the image, the image tracking apparatus includes an estimation unit which estimates a shielding state indicating whether or not the subject to be tracked is shielded by another subject different from the subject to be tracked, and a determination unit which determines the candidate area of the subject to be tracked from the candidate area detected from the image based on an estimation result of the shielding state.SELECTED DRAWING: Figure 6

The invention relates to a material in-situ strain DIC analysis method and system based on CT imaging, and the method comprises the steps: preparing speckles having significant X-ray absorption difference with a matrix, directly tracking speckle displacement through a CT image while carrying out the defect change of a sample in a deformation process through CT, and replacing a conventional camera to achieve speckle collection for DIC strain analysis. According to the method, synchronous analysis of defect expansion and a strain field in a material loading process is realized through staged CT scanning and image post-processing, and an integrated solution is provided for research on mechanical properties of materials. The space limitation of CT equipment is broken through, internal defect evolution and surface full-field strain data of the material are synchronously obtained, and the method is suitable for multi-scale mechanical behavior research of metal, ceramic, polymer and composite materials thereof.

The application discloses a kind of human shadow ammunition equipment warehouse-in and warehouse-out intelligent identification method and system, it is related to intelligent warehousing management technical field, by the inventory registration of each human shadow ammunition equipment in equipment warehouse, based on warehouse-out demand table, inventory information, inventory state and warehouse-in application information, carry out warehouse-out, warehouse-in operation, the visual image tracking of several human shadow ammunition equipment of warehouse-in and warehouse-out, based on image tracking result obtains several frames equipment directional picture, obtains trace information and analyzes according to the equipment directional picture of total frame number, extracts and obtains warehouse-out feature set and warehouse-in feature set, and compares with the warehouse-in and warehouse-out template configured in advance, based on the identification of abnormal warehouse-in and warehouse-out behavior according to comparison result, and information extraction is carried out to warehouse-in and warehouse-out behavior, establishes topological information tree and carries out linkage information push, realizes the safe, efficient, accurate ammunition equipment warehouse-in and warehouse-out management.

The present disclosure provides a method and apparatus for determining a pose of a tracked object in an image tracking process. In this method, a current frame of image and first point cloud information corresponding to the current frame of image are acquired, the first point cloud information being based on a relative relationship between an image capture device and an ambient space: and a pose of the tracked object in the current frame of image relative to the image capture device is determined according to the first point cloud information and object feature information of the tracked object if the tracked object is in a static state, the object feature information being based on a relative relationship between the tracked object and the ambient space. In this method, by using the first point cloud information to convert the relative relationship of the tracked object, it facilitates determining the pose of the tracked object.

This information processing device comprises: a plurality of imaging means that each capture an image including an object; an authentication means that authenticates the object using the image; a tracking means that tracks the object using the image; and a control means that, if authentication using the image captured by one imaging means among the plurality of imaging means has failed, controls other imaging means on the basis of tracking information relating to the tracking. According to such an information processing device, for example, even if authentication has failed because the object is partially hidden, an appropriate image can be captured using other imaging means.

This application provides a method, apparatus, device, and storage medium for target selection in adaptive cruise control, applicable to the field of adaptive cruise control technology. The method includes: acquiring image detection data of the area in front of the vehicle using an onboard camera, and acquiring an image tracking target and the vehicle's current lane based on the image detection data; acquiring radar detection data of the area in front of the vehicle using an onboard radar, and acquiring a radar tracking target based on the radar detection data; fusing the image tracking target and the radar tracking target to obtain a fused target; selecting the fused target that continuously meets target selection criteria within a preset time period as a potential following target; wherein the target selection criteria include pixel-level selection criteria and world-level selection criteria; and selecting the potential following target with the smallest longitudinal distance to the vehicle as the cruise following target. This application's method, by combining pixel-level selection and world-level selection, effectively improves the accuracy of adaptive cruise control target selection.

This disclosure provides a hand tracking system. A camera is configured to obtain a hand image of a hand of a user. A tracker is adapted to be attached to an object and configured to obtain tracker data. A processor is configured to determine the hand being touching the object based on the hand image or the tracker data. The processor is configured to determine a contact portion of the object based on the object. The contact portion is adapted to be in contact with the hand of the user. The processor is configured to determine a hand gesture of the user based on the hand image. The processor is configured to determine a hand pose of the hand and an object pose of the object based on the hand gesture and the contact portion.

The invention relates to the technical field of space debris orbit forecasting, in particular to a space debris orbit forecasting correction method based on machine learning and triangular characteristic decomposition. The azimuth angle and the pitch angle are converted into six-dimensional feature vectors through triangular feature decomposition, and a real-time mapping model of orbit forecast deviation is established in combination with a neural network. The two bottleneck problems on which traditional image tracking depends are solved: 1) in a strong laser interference scene, tracking failure caused by background light saturation of an optical camera is avoided, and real-time image feedback is replaced by deviation prediction driven by historical echo data; 2) when space debris in an earth shadow area is invisible, constructing a training set by using deviation data captured at the moment of successful laser ranging, and realizing autonomous correction of forecast errors under a non-illumination condition; the method gets rid of dependence on space debris optical visibility, realizes all-day and all-arc telescope pointing deviation correction, and improves the reliability and continuity of space debris orbit prediction under complex working conditions.

This application discloses an image tracking processing method and apparatus. The method includes: obtaining feature points and attributes of each frame image in a real-time video; performing a geometric transformation on a reference frame image according to preset transformation conditions to obtain a relation table; determining a lookup index for at least one point pair based on the attribute differences between the first and second feature points of multiple point pairs; determining an evaluation value for each lookup index based on the template image corresponding to each lookup index in the relation table; obtaining the target rotation angle and target scaling scale based on the evaluation value of each lookup index; and determining the target transformation matrix based on the target rotation angle and target scaling scale. This method can establish a relation table by performing a geometric transformation on the reference frame image, quickly calculate the evaluation values ​​of multiple lookup indices through the relation table, and use multiple evaluation values ​​to quickly find the target rotation angle and target scaling scale, effectively improving the processing efficiency of image tracking.

A system includes a sensor, a weight sensor, and a tracking subsystem. The tracking subsystem receives an image feed of top-view images generated by the sensor and weight measurements from the weight sensor. The tracking subsystem detects an event associated with an item being removed from a rack in which the weight sensor is installed. The tracking subsystem determines that a first person or a second person may be associated with the event. In response to determining that the first or second person may be associated with the event, buffer frames are stored of top-view images generated by the sensor during a time period associated with the event. The tracking subsystem then determines, using at least one of the stored buffer frames and a first action-detection algorithm, whether an action associated with the event was performed by the first person or the second person.

Embodiments of the present disclosure disclose a method, device, equipment and medium for unmanned aerial vehicle (UAV) control based on UAV vision. A specific embodiment of the method comprises: a control signal emitting device emits signals in real time; a control recognition device performs recognition processing on the unknown flying object to generate a recognition result; a monitoring UAV in a UAV group performs image tracking processing on the unknown UAV; the unknown UAV is positioned; a second unknown UAV trajectory is generated, and the first unknown UAV trajectory is corrected according to the second unknown UAV trajectory to obtain a corrected UAV trajectory; the unknown UAV is subjected to trajectory prediction processing to generate a UAV predicted trajectory; and the UAV group or the interception device intercepts the unknown UAV. This embodiment improves the regional security of the detection area.