46results about How to "Improve target tracking efficiency" patented technology

The invention discloses a target tracking method and device, a storage medium and electronic equipment, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring a video stream acquired by a camera, and decoding the video stream to obtain multiple frames of images; for a to-be-processed current frame image, when the current frame image is afirst type image, performing target detection on the current frame image to obtain position information of a detected target in the current frame image; and when the current frame of image is a secondtype of image, predicting the position information of the detected target in the previous frame of image in the current frame of image based on the motion vector of the previous frame of image. According to the target tracking method provided by the invention, the information of all the decoded images is utilized, the problem of relatively low speed caused by performing target detection on the frame-by-frame images is avoided, the real-time tracking speed is ensured, and the accuracy of the obtained tracking result is also ensured.

The invention discloses a real-time tracking system and method of a moving target. The real-time tracking system comprises a video monitoring subsystem, a camera subsystem, a GIS (Geographical Information System) subsystem, an image recognition subsystem and a tracking control subsystem, wherein the GIS subsystem is used for providing a peripheral monitoring point list of the current position of a tracking target; the tracking control subsystem is used for extracting a target characteristic information list and a preset characteristic matching strategy; the image recognition subsystem is used for obtaining each path of video stream in the peripheral monitoring point list and carrying out real-time calculation on the monitoring camera of the current position of the tracking target according to the target characteristic information list and the preset characteristic matching strategy; the video monitoring subsystem is used for switching the main video into the monitoring camera of the current position of the tracking target. The real-time tracking system and method disclosed by the invention have the advantages that by integration of a GIS technology, a cloud computing technology and an image recognition technology, the real-time performance and the automation for tracking the moving target are realized, so that the labor cost and the personal errors are reduced to greatest extent and the tracking efficiency is improved.

The embodiment of the invention discloses an information generation method and apparatus. A specific implementation manner of the method comprises the steps of obtaining a to-be-detected image, and first area position information used for defining a human face area in a previous one frame of image of the to-be-detected image; based on the first area position information, performing a human face key point positioning operation on the to-be-detected image to obtain a coordinate of each human face key point included in the to-be-detected image; based on the coordinate of the each human face key point, determining second area position information used for defining the human face area in the to-be-detected image; and generating corresponding relationship information between the to-be-detected image and the second area position information. According to the implementation manner, the information generation rich in pertinence is implemented.

The invention provides an oil field dynamic intrusion target tracking system and method based on multiple unmanned aerial vehicles, and relates to the technical field of unmanned aerial vehicle inspection and dynamic prediction tracking. The system comprises a control module, a monitoring module, a dynamic target trajectory prediction module and a multi-machine cooperative target tracking trajectory optimization module. According to the system, the monitoring module mounted on the unmanned aerial vehicle is used for collecting data of a dynamic intrusion target in a ground oil field working environment, then the dynamic target trajectory prediction module is used for predicting trajectories of the target in multiple sampling moments in the future, and the multi-machine cooperative target tracking trajectory optimization module is used for optimizing the distance error between the unmanned aerial vehicle and the target in the tracking process according to the predicted trajectories, sothat the distance error is kept within an effective tracking threshold value, and cooperative target tracking of the multiple unmanned aerial vehicles is achieved.

The invention provides a target tracking method based on natural language information assistance. The method comprises the steps: carrying out the description of visual features and natural language features of an image, and extracting the visual features of a target and a candidate region and the natural language features for describing the target; constructing a visual tracking model based on the initial visual features, and constructing a natural language response model based on the initial natural language tracking model; and determining a response of the visual feature by using the imagefeature detector fused with the multi-layer features, and detecting the candidate image by using the natural language feature detector while acquiring a response of the candidate image to the naturallanguage feature. According to the method, the advantages of natural language information and visual information are combined, high-level semantic features provided by a natural language are used forassisting visual features, and target tracking is more stable.

The invention relates to a tracking control method and device, an object tracking component and a storage medium. The method is applied to an object tracking component of a photographing equipment stabilizer. The distance between the object tracking component and a camera of the shooting equipment is smaller than or equal to a preset distance threshold value and is consistent with the camera in orientation; the method comprises the following steps: acquiring detection information of a facing position to obtain first detection information; when it is determined that the first feature recognition information of the target object exists according to the first detection information, position information of the target object is obtained; and a tracking movement instruction is sent to a control component according to the position information, and the control component is triggered to perform position adjustment control according to the position information, so that a camera of the shooting equipment faces the target object. According to the scheme, the object tracking component can adjust the orientation of the camera through interaction with the control component, target tracking is achieved, cooperation with a specific APP is not needed, and the target tracking efficiency can be effectively improved.

The invention discloses a path planning method suitable for dynamic target tracking and combining a proportional guidance method and an artificial potential field method, and relates to the technical field of mobile robot path planning. The method is used for solving the problem that the tracking efficiency is not high due to the fact that the obstacle avoidance requirement in the dynamic target tracking process is not considered in an existing path planning method. The method comprises the steps: according to the positions of a mobile robot, acquiring a to-be-tracked target and an obstacle, the first movement speed expected direction of each step of the mobile robot through calculation based on a proportional guidance method; calculating and obtaining a second movement speed expected direction of each step of the mobile robot based on an artificial potential field method; performing fusing computing on the first movement speed expected direction and the second movement speed expected direction to obtain the actual movement direction of each step of the mobile robot; and dynamically updating the planned path according to the path planning step length and the actual motion direction. Therefore, the target tracking efficiency is improved, meanwhile, the good obstacle avoidance performance is guaranteed, and the method has the high engineering application value.

The invention discloses an infrared image generation method. The method comprises the steps of obtaining processed pixel data output by an infrared focal plane detector; determining a first mean valuematrix of the processed pixel data in the row direction and a second mean value matrix of the processed pixel data in the column direction; obtaining processed Dummy pixel data output by the infraredfocal plane detector; determining a third mean value matrix of the processed Dummy pixel data in the row direction and a fourth mean value matrix of the processed Dummy pixel data in the column direction; and generating an infrared image according to first array data determined by the processed pixel data, the first mean value matrix and the third mean value matrix and second array data determined by the processed pixel data, the second mean value matrix and the fourth mean value matrix. The method is convenient to implement, occupies few logic resources, is low in power consumption, and further improves the infrared image quality. Transverse and column line noises with relatively large amplitudes can be removed; and the data delay is small, and the pixel data output by the infrared detector is directly processed without depending on the effect of non-uniformity correction.

The embodiment of the invention discloses an image processing method, device and server, which are used to improve the coincidence degree of a target frame and a foreground target. In the embodiment of the present invention, first obtain the input frame and the background frame, then determine the row or column to be cut in the target frame, calculate the gradient difference matrix of the input frame and the background frame in the row or column to be cut, and calculate the gradient difference matrix The average value of the gradient difference, judge whether the average value is less than the gradient comparison value, if it is less than, cut off the position to be cut from the target frame. In the embodiment of the present invention, the similarity between the input frame and the background frame is used as the basis for cutting the target frame, and the average value of the gradient difference value of the row or column in the target frame of the input frame and the background frame is respectively calculated as a comparison feature, and compared with the gradient comparison value , so as to judge whether the current row or column is a real foreground target, if not, cut off the current row or column from the target frame. In this way, the target frame can be effectively corrected, and the coincidence between the target frame and the foreground target can be improved.

A computer vision-based vehicle-mounted target tracker convenient to install disclosed by the present invention comprises an installation adjustment platform and a target tracking device, the target tracking device is installed on a vehicle through the installation adjustment platform, and the installation adjustment platform comprises a first installation platform, a second installation platform and an adjusting mechanism. The adjusting mechanism is used for driving the first installation platform and the second installation platform to perform position adjustment, the target tracking device comprises a camera module, an information processing module and a display alarm module, the information processing module is used for analyzing and processing information sent by the camera module and then sending the information to the display alarm module for display, and the display alarm module is controlled to give an alarm. According to the invention, the target tracking device is installed through the installation adjustment platform, so that the target tracking device is convenient to disassemble and install, and meanwhile, the target tracking device tracks the target in real time according to the video information of a to-be-tracked target, so that the target tracking efficiency can be improved.

The invention relates to the technical field of video monitoring, and provides a target motion trail analysis method relating to two or more monitoring videos which can be invoked by a control center. The target motion trail analysis method comprises the following steps: confirming the first monitoring video in which a target appears; obtaining first time-space information and first target characteristics of the target according to the first monitoring video; confirming one or more monitoring videos taken around a first position; obtaining various object characteristics in the confirmed one or more monitoring videos, and confirming the second monitoring video from which the object characteristics obtained are matched with the first target characteristics with a similarity exceeding a preset threshold value; confirming the motion trail of the target according to second time-space information carried by the second monitoring video and the obtained first time-space information. The target motion trail analysis method has the advantages that the big data analysis function of cloud computing can be utilized for confirming the motion rail of the target in a plurality of monitoring videos, so that the target tracking efficiency is improved.

The present application relates to a tracking control method, device, object tracking component and storage medium. The method is applied to the object tracking component of the stabilizer of the shooting device; the distance between the object tracking component and the camera of the shooting device is less than or equal to a preset distance threshold, and is consistent with the orientation of the camera; the method includes: acquiring detection information of the facing position, Obtain the first detection information; when it is determined according to the first detection information that there is the first characteristic identification information of the target object, obtain the position information of the target object; send a tracking movement instruction to the control component according to the position information, and trigger the control component to perform position detection according to the position information. Adjust the controls so that the capture device's camera is facing the subject of interest. The object tracking part of the above solution can adjust the orientation of the camera by interacting with the control part to realize target tracking, and does not need to cooperate with a specific APP, which can effectively improve the efficiency of target tracking.