121results about How to "Realize continuous tracking" patented technology

The invention discloses a human body target tracking method in a nonoverlapping vision field multi-camera network, which comprises the following steps of target detection of a single camera, target tracking of the single camera, and target association and continuous target tracking of multi cameras, wherein the target association of the multi cameras comprises the following substeps of target characteristic extraction and topological relation establishment, the target characteristic extraction is extraction of each human body movement target HSV (Hue, Saturation and Value) characteristic, andthe topological relation establishment is establishment of the space-time topological relation of the multi-camera network according to the obtained association relation of targets. By the human bodytarget tracking method disclosed by the invention, the continuous tracking of human body movement targets in a large scene is realized, and the problem of the monitoring of the limited vision range of a single camera is solved.

The invention relates to method and system for realizing video intelligent track navigation. The method for realizing video intelligent track navigation comprises the following steps: establishing a camera association relation table for storing camera association relations among all cameras; acquiring information of a newly added camera from a camera association relation maintenance module, and setting basic attributes of the newly added camera; generating associated camera information according relevant camera attribute information; displaying associated cameras according to the associated camera information; carrying out association relation adjustment and confirmation on the associated cameras; and storing the camera association relation information in the camera association relation table. The system for realizing video intelligent track navigation comprises a control terminal, a server, a video matrix and a plurality of cameras and can intelligently prompt monitoring personnel with the possible moving direction of a monitored target and display relevant images in advance so that the monitoring personnel can switch to relevant images quickly after the monitoring target enters into another image, thereby realizing the continuous track and improving the security quality of video monitoring systems. The system for realizing video intelligent track navigation has the advantages of simply operation, convenient control and the like and can significantly improve the working efficiency of the monitoring personnel.

The invention discloses a channel concrete lining non-destructive testing system, which includes a computer, and is characterized in that: the computer is connected to a dedicated ground-penetrating radar subsystem, and the computer includes a radar data processing subsystem and an ultrasonic data processing subsystem, and the radar The data processing subsystem includes a file management module, a profile data processing module, a lining disease simulation analysis module, a layer tracking and disease identification module, and a lining physical property analysis module. The ultrasonic data processing subsystem includes an ultrasonic wave velocity calculation module and a coagulation intensity Judgment module. This system improves the data processing efficiency and accuracy of the construction quality inspection of canal concrete lining, and provides scientific, accurate, advanced and effective technical support for the construction quality inspection of canal concrete lining.

The invention discloses an all-weather multipath channel pedestrian flow monitoring system based on wireless infrared monitoring. The system comprises a video acquisition terminal, wherein the video acquisition terminal comprises active infrared monitoring cameras and video acquisition modules both installed at the inlets of various pedestrian flow monitoring channels, a codebook background module, an image preprocessing module and a wireless transmission module; and the system further comprises: a central server terminal for receiving acquired pedestrian flow data and performing projection, tracking and counting, wherein the central server terminal comprises a wireless reception module, a new pedestrian flow detection module and a pedestrian flow tracking and counting module. The invention provides an all-weather multipath channel pedestrian flow monitoring system based on wireless infrared monitoring, with excellent self-adaptive ability and outstanding reliability.

The invention relates to the technical field of image processing, in particular to a target tracking method, a target tracking device, a computer readable medium and electronic equipment. The method comprises the steps of obtaining a to-be-tracked video, and performing target detection on the to-be-tracked video to obtain a key frame image containing a tracking target; performing image recognitionon the key frame image to obtain an object area containing a tracking target, and performing key point extraction on the object area to obtain key point data of the tracking target; extracting a to-be-detected image of a next frame adjacent to the key frame, and performing feature extraction on the to-be-detected image to obtain a second feature map of the to-be-detected image; inputting the second feature map and the key point data into a prediction model as input parameters to obtain corresponding prediction key point data of each key point in the to-be-detected image; and determining the tracking target in the to-be-detected image according to the predicted key point data.

The invention discloses a distributed tracking and monitoring method and system. The monitoring method includes the step of central control, the step of selecting a specific video camera and the step of monitoring through the specific video camera, wherein in the step of central control, a central service station sends a control instruction to a video camera group and receives feedback information of the video camera group; in the step of selecting the specific video camera, the central service station selects at least one video camera from the video camera group to serve as the specific video camera according to the feedback information and performs linkage monitoring; in the step of monitoring through the specific video camera, the specific video camera takes the initiative to recognize, track, report and monitor a target to complete continuous tracking and monitoring of the monitored target according to the control instruction of the central service station. To overcome the defects in the prior art, the distributed tracking and monitoring method and system can solve the problems that in an ultra-high definition image, a mobile target is large in recognition and calculation amount and difficult to track, the monitoring efficiency can be greatly improved particularly for the situation that the mobile target is tracked all over the city, and the distributed tracking and monitoring method and system has important significance for fast tracking and monitoring the target in real time.

The invention relates to a vehicle object information detection and identification system and a signal processing method thereof. Two lane calibration aided leading marks of are placed in the same lane line, distance and phase information between the two lane calibration aided leading marks and a radar detector is measured and obtained by the radar detector, object detection is carried out on an object vehicle, the distance and speed of the object vehicle are obtained, and a signal processing system resolves lane information according to the distance and phase information of the two lane calibration aided leading marks. According to the invention, vehicle objects of intermediate or long distance can be tracked continuously under the condition that the radar detector is placed at random, the distance, speed and lane information of the vehicle object can be detected in high precision, detection of low false alarm rate and high precision can be realized for the moving vehicle object on the basis that the hardware processing complexity of the system is reduced, interference and clutters can be filtered effectively in complex application, and accurate lane identification is realized for the vehicle objects of intermediate or long distance under the condition that the direction of radar wave beams is not calibrated via professional calibration equipment.

The invention relates to an airport runway monitoring system based on intelligent video monitoring for a whole scenic spot, comprising an information processing and storing module, wherein the input end and output end of the information processing and storing module are connected with the input end and output end of a terminal service module and respectively connected with a whole scene monitoring module and a target tracking module through a switchboard, wherein the whole scene monitoring module is used for monitoring the whole scene of an airport runway, and the target tracking module is used for tracking a target; the output end of the information processing and storing module is connected with an alarm module; and the input end and output end of the whole scene monitoring module are connected with the input end and output end of the target tracking module. The airport runway is monitored in a firearm bolt and ball machine linkage way, so that large-scene and whole-scene monitoring can be realized, and meanwhile, continuous tracking and information capturing of a plurality of targets within a monitoring range can be realized; and missing inspection and false alarm in video monitoring are effectively reduced, and the intelligence of modern security and protection is really realized, therefore, the airport runway monitoring system has higher application values in the aspects of airport runway monitoring and airplane landing tracking.

The invention provides a YOLO-based specific target tracking method, and belongs to the technical field of target tracking. The method comprises the following steps: firstly, setting a to-be-tracked specific target, shooting a video in real time by using a camera, sending the video to a computer, and putting an image of the specific target as a template into a template folder; Taking the first frame of image with the specific target in the video as a first frame of image, and carrying out target detection on each frame of image by using a YOLO detection algorithm to obtain a detection result corresponding to the type of the object where the specific target is located in each frame of image; And correcting the detection result by using a Deepsort algorithm, and then tracking a specific target by using a Surf algorithm on the corrected result. According to the method, continuous tracking of a single specific target can be realized, the speed is high, the real-time performance is good, and the method is well suitable for environment changes.

The invention relates to a video tracking method and system. The method comprises the following steps: acquiring a video tracking target from an image picture of video monitoring; acquiring a continuous monitoring region image picture for the monitoring region located by the image picture in the video monitoring through a pre-constructed building information model; acquiring the continuous monitoring region image picture of the video tracking target from the continuous monitoring region image picture according to feature information of the video tracking target; and displaying the video tracking target through the image picture of the video monitoring and the continuous monitoring region image picture of the video tracking target. Therefore, the video tracking target is displayed through an image picture and the continuous monitoring region image picture, the continuous tracking of the video tracking target is realized without depending on the manual implementation, the manual implementation is avoided, and the problem of low efficiency is solved.

The invention discloses a car target trajectory tracking system which comprises multiple data collecting systems and a data control system. Each data collecting system comprises an interchanger, an industrial personal computer and millimeter-wave radar. A licence plate camera is connected with the industrial personal computer through a data line, the millimeter-wave radar is connected with the interchanger through a signal line, and the interchanger is connected with the industrial personal computer through a signal line. The data control system comprises a data server and a data displayer, the industrial personal computer is communicated with the data server through a public network data converter, and the data server is connected with the data displayer. According to the system, a targetcar can be monitored. The invention further discloses a car trajectory tracking method. According to the tracking method, the moving track of the target car can be accurately tracked and positioned,and high innovativeness is achieved.

The invention discloses the multiple vibration hydraulic buffer based on parallel six-freedom platform. It comprises the butter fixing platform, six double-functional single-freedom hydraulic buffers, the buffer movable platform and the joint hinge. The buffer movable platform connects the controlled target; the fixing platform is equipped on the base or other fixing planes; the said six hydraulic buffers connect the movable and the fixing platforms through the hinge and support the movable platform to move six-dimensionally. The invention can bear the vibrations and the strikes from six directions at the same time, can track the continuous movement of the controlled target and provides a tight and ideal assembly scheme for the multi-freedom composite vibration buffer.

The invention discloses an RFID and video technology-based analysis method and an analysis system thereof. On one hand, the method comprises the steps of 1, obtaining the data of an IC chip by an RFID reader-writer; 2, obtaining the video data of a person or an object by a video acquisition device; 3, comparing the time information and the location information in the video data with the time information and the location information in the data of the IC chip; on the condition that the time information and the location information in the video data are consistent with the time information and the location information in the data of the IC chip, judging the video data and the data of the IC chip are associated to the same person or the same object, and conducting the step 4; 4, deeming the information of the person or the object carried in the data of the IC chip as the associated information of the person or the object in the video acquisition device. On the other hand, the invention provides an analysis system for realizing the above method. The method and the system have the advantages of good data relevance, high recognition capability, good person / object tracking continuity, fast processing speed and the like.

The embodiment of the invention discloses an object monitoring method and device, a server and a storage medium. The method comprises the steps of obtaining a monitoring image collected by a first camera in a camera set; Determining a target moving object from the monitoring image, and calculating a monitoring distance between the first camera and a target object corresponding to the target movingobject; Determining whether a monitoring switching condition is met or not according to the monitoring distance, and determining a second camera from the camera set when the monitoring switching condition is met; And switching the first camera into a second camera, and calling the second camera to monitor the target object, thereby realizing continuous tracking of the target object.

Disclosed is a parallel type six-freedom-degree redundant driving antenna structural system. The parallel type six-freedom-degree redundant driving antenna structural system comprises an antenna reflective surface and an antenna pedestal, wherein the antenna reflective surface is linked on the antenna pedestal; the antenna pedestal comprises a fixed platform and a movable platform which are arranged in parallel; the fixed platform is a convex circular table with gradually decreased caliber; three groups of guiding rail units are uniformly distributed on the fixed platform; each guiding rail unit group consists of two guiding rails; each guiding rail is equipped with a driving sliding block; a telescopic unit is hinged on each driving sliding block in a one-to-one correspondence manner; all the telescopic units are hinged to the movable platform; each movement point can perform coordination action by following four movement units, so that special forms and positions can be effectively reduced; the convex circular table shape of the fixed platform is matched with the driving sliding blocks to realize overturn of the antenna; the multiple driving sliding blocks are matched with the telescopic units to perform a coordination operation, so that not only all-around tracking of the antenna, but also negative angle pitching of the antenna can be realized; and consequently, the stability, flexibility and rigidity of the mechanism are highly reinforced.

The invention relates to a production line object part motion detection and tracking method based on videos, and belongs to the field of automatic production lines. A single digital camera is adopted for acquiring conveying images of production line objects, modeling is conducted through background subtraction and object characteristics, and then image segmentation of moving objects and a production line background is achieved. In the normal operating process of a production line, through background subtraction, a background region of the production line and a moving object part region are determined, and modeling is conducted through surface characteristics and contour characteristics of the products. In the assembly process of the production line, a Gaussian model of background images is continuously updated, and object parts on the production line are better segmented and are not affected by external environment interference. Meanwhile, motion trails of the objects are matched in multiple images, and therefore the multiple objects can be continuously tracked. By the adoption of the method, the calculation reliability of the positions of the product parts in continuous motion can be improved, the objects can be accurately tracked, and the requirements for product assembly automation can be met.

The invention discloses a fast and accurate medical service method and system for patients. The method comprises steps of establishing a medical database, receiving requests, waking up medical treatment-needing person service, making an appointment, initial diagnosis and diagnosis result feedback. The system comprises a server storage module, a medical treatment-needing person client module, an appointment selection determination module, a medical staff client module, a communication module, a voice input processing module, a text input module, a query archiving keyword generation module, a screening retrieval result generation module and a feedback module. The method and the system are advantaged in that problems that the medical procedure is cumbersome and difficult to register in the prior art can be solved, medical service efficiency and portability can be effectively improved, the waiting time is reduced for medical treatment-needing persons, the medical care pressure is alleviated, the accurate and efficient self-service is achieved through the Internet technology, and properties of high system intelligence degree, advanced and scientific methods, relatively high practical value and broad application prospects are further achieved.

The invention discloses a cross-camera pedestrian re-identification and tracking method, which comprises the following steps of: 1) image preprocessing: selecting pedestrian images from a video sequence for pedestrian re-identification, and outputting a clear pedestrian image for pedestrian re-identification by inputting a plurality of low-resolution pedestrian images; 2) cross-camera cooperativeidentification: performing feature extraction of pedestrian images, and performing feature matching of pedestrian images between cameras; 3) blind area track restoration and tracking: a, detecting pedestrians by adopting a target detection method, and inputting images of the pedestrians acquired by a camera in a vision field into a GAN network for restoration so as to realize continuous tracking of a target track; and b, calibrating the image captured by the camera and the real coordinate through a geographic information system to obtain a target starting point and a key position, and integrating through an algorithm to obtain a possible track of the target. According to the invention, cross-camera re-identification and track recovery and tracking are realized, and the efficiency is improved, so that the accuracy is improved.

The embodiment of the invention relates to the field of data processing, and discloses a monitoring method, monitoring equipment and a computer readable storage medium. In some embodiments of the invention, the monitoring method comprises the steps of obtaining first image data shot by a first camera of monitoring equipment; determining indication information of the target object according to thefirst image data, wherein the indication information of the target object indicates first position information of the target object in the first image data; and according to the indication information, determining whether to monitor the target object based on second image data shot by at least one second camera of the monitoring equipment, the focal length of the at least one second camera being different from the focal length of the first camera. This increases the monitoring range of the monitoring equipment.

The invention discloses a people tracking method based on intelligent video analysis. The people tracking method comprises the following steps of analyzing the content of an image, and removing unusable background information; according to the sizes, moving speeds and moving rules of different moving targets, accurately identifying people, animals, vehicles or other objects, and extracting important information; comparing the moving rule of each moving target and a formulated safety rule, and verifying the moving safety of each moving target; under a complicated background, positioning and tracking each moving target in real time; predicting and filtering the state parameters of each target, such as mass center displacement, speed and acceleration; after each target is blocked by a small blocking object, continuing to track; after the target tracking is completed, automatically returning back to a current preset position. The people tracking method based on the intelligent video analysis has the advantages that the multi-target tracking is realized, the continuous tracking on each target with blocking time not exceeding three seconds is realized, and the accurate tracking before crossing and after separation of multiple targets is realized.

The invention relates to an image continuous matching method and system. The method comprises the following steps that: S1, a preset image size is calculated according to the size of a target image and the printing precision of a printing device, and an original image is processed according to the preset image size, so that an image to be matched can be obtained, wherein the original image is an image obtained through the shooting of the target image by a camera; S2: feature extraction is performed on the image to be matched, and all feature point information of the image to be matching is obtained, and feature matching is performed on the image to be matched and all reference images in a reference image library; S3, when matching fails, all sub-images of which the feature point number is smaller than a preset feature point number in the image to be matched are obtained; and S4, feature point adding processing is performed on the all the sub-images, so that a reference image is obtained, and the reference image is stored in an image matching reference image library. With the image continuous matching method and system of the invention, when detected feature points are not enough, feature points are added into the image to be matched, so that recognition degree can be improved, and therefore, the continuous matching and tracking of the target image can be realized.

The invention discloses an expressway driving early warning system and method. According to the invention, respective satellite positioning data of a key vehicle early warning device and a common vehicle early warning device is sent to a background data processing system; and the background data processing system determines whether the distances between key vehicles, and between key vehicles and common vehicles and the relative speed meet an early warning condition and sends corresponding early warning information to the key vehicle early warning device and the common vehicle early warning device, thereby achieving objectives that possibly occurring dangers can be warned earlier and avoided in advance and driving safety on the expressway is improved. According to the invention, by use of information technology of mobile Internet, real-time continuous tracking and monitoring of vehicle running states are achieved; the function that beyond visual range running early warning function based on early warning scene real-time identification is achieved; and occurrence of accidents of collisions and rear-end collisions on the expressway can be effectively reduced.

The invention discloses a host-oriented suspicious network connection identification method. The host-oriented suspicious network connection identification method comprises the steps of: (A), determining various network connection characteristics used for realizing network connection type clustering, and constructing a multi-dimensional space based on the network connection characteristics, so that each network connection is mapped into one point in the multi-dimensional space; and (B), collecting all in and out network flow of an appointed host within a unit time period, mapping all network connections in the flows to the points in the multi-dimensional space constructed in the step (A) one by one, clustering the mapped points to obtain multiple sub-types composed of the points corresponding to the network connections, and judging that all the network connections corresponding to the sub-types, the point number of which is less than the appointed threshold value, as suspicious network connections. By means of the host-oriented suspicious network connection identification method disclosed by the invention, the suspicious network connections are continuously tracked, so that detection and continuous tracking of all kinds of APT attacks can be realized.

The invention discloses a stepping type all-day area scanning searching device and method, and the device consists of two searching mechanisms which are respectively placed at the upper side and the lower side of a satellite and are responsible for 180-degree airspace searching tasks; Each searching mechanism is composed of a one-dimensional rotating mechanism used for stepping rotation and two cameras arranged on the one-dimensional rotating mechanism. A view field of each camera is 30 degrees * 30 degrees, and a splicing view field formed by the two cameras is 60 degrees * 30 degrees. And each searching mechanism realizes 60-degree * 180-degree full-view-field coverage by virtue of the rotation of the one-dimensional rotating mechanism. According to the method, the strategy problem of multi-target detection and tracking in an all-day area is solved, the rapidity of multi-target detection is improved, and the false alarm rate is reduced; the rapidity of multi-target tracking is improved; therefore, detection and continuous tracking of the spatial unnatural celestial body target are realized. And a feasible engineering foundation is provided for on-orbit satellite anti-collision avoidance.

The invention discloses a vision tracking mobile robot based on WIFI auxiliary positioning and a control method. The vision tracking mobile robot comprises a robot shell, a chassis, a steering engineholder, a visual module, WIFI-AP modules, a control module, drivers, a driving motor, driving wheels and universal wheels. The control module is electrically connected with the steering engine holder,the visual module, the WIFI-AP modules and the drivers. The steering engine holder is arranged at the top outside of the robot shell, and the visual module is installed on the steering engine holder.At least three WIFI-AP modules are arranged and installed at different positions outside of the robot shell. The control module receives signals collected by the visual module, generates driving instructions and realizes target tracking, and the control module receives signals collected by the WIFI-AP modules to locate the position of the robot. The vision tracking mobile robot based on the WIFIauxiliary positioning and the control method integrates vision tracking of the mobile robot and the WIFI positioning effectively, realizes continuous tracking of a specific target, and improves positioning accuracy and robustness.

The invention provides a road position intelligent recognition and interference source elimination method. The method comprises the steps of 1, data preprocessing, including the steps of performing background denoising on original data of a ground penetrating radar, setting a zero line and performing gain and sliding averaging; 2, interference source recognition and elimination, wherein an interference source sample library is established, and automatic recognition and effective elimination of an interference source are realized through machine learning; and 3, position tracking and information extraction, wherein a reference road seed point is used as a center to establish a three-level window operator, and tracking and judgment are sequentially performed till acquired position coordinateinformation meets a requirement. The core of the method is a "three-level window operator" position tracking algorithm, and precise positioning and recognition on a tracking road position can be realized.

The invention discloses a parcel tracking method and a related device. The parcel tracking method comprises the following steps: acquiring video data obtained by scanning a parcel transmission area bya scanning device; taking the image of the target package detected in the video data as a target image, and taking the next frame of image of the target image in the video data as a to-be-detected image; based on the position of the target package in the target image, determining a tracking area corresponding to the target package in the target image and the to-be-detected image; training based on the image data of the tracking area of the target image to obtain a detection model; and detecting a tracking area of the to-be-detected image by using the detection model to obtain a target area corresponding to the target package in the to-be-detected image. According to the scheme, the package detection efficiency can be improved.

The invention discloses a furnace gas analysis model-based decarburization control method of vacuum oxygen decarburization refining. The furnace gas analysis model-based decarburization control method comprises following steps: after vacuum pumping, pre-treatment, and oxygen lance oxygen blowing, exhaust gas parameter information, including furnace gas generation amount and furnace gas composition, of an oxygen blowing period is calculated; oxygen blowing decarburization rate is calculated based on the furnace exhaust gas parameter information; tracking calculation on molten steel carbon content is carried out; determination on oxygen blowing decarburization rate correction factors and molten steel carbon content correction computation is carried out; determination on finish of top lance oxygen blowing is carried out; exhaust gas related parameter information, including suction air amount and exhaust gas flow amount calculation, furnace gas generation amount calculation, and furnace gas composition calculation, at a vacuum period is calculated; vacuum treatment decarburization rate is calculated; molten steel carbon content is calculated; stop determination on deep vacuum free decarburization treatment is carried out; and subsequent processing is carried out until VOD (vacuum oxygen decarburization) processing is finished. The furnace gas analysis model-based decarburization control method is capable of reducing interference of environment change on furnace gas detection information, and can be used for model calculation and process monitoring more effectively, and for accurate control on process oxygen blowing and vacuum treatment time.

The invention discloses a medical whole-process intelligent service method and system. The method comprises the steps of establishing a medical database, receiving the request, awakening the medical personnel service, confirming the appointment, preliminary diagnosis and diagnosis result feedback and the like. The system comprises a server storage module, a medical personnel client module, an appointment selection determination module, a medical personnel client module, a communication module, a voice input processing module, a text input module, a query and archiving keyword generation module, a screening and retrieval result generation module and a feedback module. The problems that the present medical procedure is tedious and registration is difficult can be solved, the efficiency and the portability of the medical service can be effectively improved, waiting time of the medical personnel can be reduced, the medical pressure can be alleviated, the accurate and efficient self-servicecan be realized through the Internet technology, the system has high intelligent degree, the method is advanced and scientific, and the practical value is high and the application prospect is wide.

The invention provides a target tracking method and device. The target tracking method and device are applied to a server storing feature models corresponding to target objects. The method comprises the steps of performing target detection on a current video frame image, and extracting corresponding CNN features according to position information, obtained through detection, of all to-be-detected objects; calculating to obtain a corresponding similarity matrix according to the position information and the CNN features of each to-be-detected object and the position information and the feature model of each target object in the previous video frame image; performing data association on each to-be-detected object and each target object based on the similarity matrix to obtain an optimal matching result; and if a to-be-detected object successfully matched with the corresponding target object exists in the optimal matching result, updating the corresponding feature model according to the CNNfeatures of the to-be-detected object, and obtaining a corresponding tracking result based on the to-be-detected object. The method has the advantages of strong anti-interference capability and hightracking success rate, and can continuously track the target object.