Traffic target detection and identification method based on unmanned aerial vehicle monitoring
A technology of target detection and recognition method, applied in character and pattern recognition, computer parts, instruments, etc., can solve the problem of limited monitoring range and achieve the effect of easy operation
Active Publication Date: 2021-08-24
XUZHOU UNIV OF TECH
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] However, when the existing video surveillance methods detect and identify traffic targets, the cameras are generally installed in fixed positions, the monitoring range is limited, and it is impossible to detect and identify traffic targets in a wide range.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0056] 1) Collect 1280*720 color sample images of various traffic targets (pedestrians, cars, etc.) in the traffic environment of interest through drones in advance, and generate the feature vector of traffic targets in each image through human-computer interaction, The generation process is as follows:
[0057] a. For each image collected, use a rectangle to frame the traffic target in the image manually, assuming that the length L=128 of the circumscribed rectangle of a certain video target, and the width W=72;
[0058] b. Inside the circumscribed rectangle, multiple smaller rectangles are continuously generated. Here we set up to continuously generate 3 rectangles, which are 90*50, 63*35, and 44*24;
[0059] c. Obtain 4 rectangles through steps 1) and 2), and calculate the color difference vectors of adjacent rectangles from the inside to the outside, such as figure 2 As shown, three 8 3 Dimensional vectors, denoted as L 1 , L 2 , L 3 , so as to finally form the featu...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a traffic target detection and identification method based on unmanned aerial vehicle monitoring. The method comprises the following steps: collecting images of traffic targets by using an unmanned aerial vehicle, generating feature vectors of the traffic targets in each image in a man-machine interaction mode, and constructing a feature matching tree by using the extracted feature vectors; then extracting features of a monitoring area image collected by an unmanned aerial vehicle falling into a sliding window for multiple times to form feature vectors, traversing a complete image by using the method, and matching all the formed feature vectors with tree nodes of a feature matching tree to obtain the position and category of a traffic target appearing in the image. According to the method, the specific traffic target is detected and identified in a large-range traffic area; the method is easy to operate, and information support and decision basis are provided for a traffic department to better manage and control the traffic target.
Description
technical field [0001] The invention relates to a traffic target detection and recognition method based on unmanned aerial vehicle monitoring, and belongs to the technical field of traffic detection and recognition. Background technique [0002] Traffic targets, including pedestrians, vehicles, traffic signs, etc., the detection and recognition of traffic targets is an important part of the intelligent transportation system, which can provide strong information support and decision support for the traffic department. At present, the method based on video surveillance is a relatively commonly used traffic target detection method. [0003] However, when detecting and identifying traffic targets in the existing video-based monitoring methods, the cameras are generally installed in fixed positions, and the monitoring range is limited, which makes it impossible to detect and identify traffic targets in a wide range. Contents of the invention [0004] Aiming at the problems exi...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06K9/00G06K9/46G06K9/62
CPCG06V20/52G06V20/54G06V10/443G06V10/50G06V10/56G06V2201/07G06F18/23213G06F18/24323
Inventor 李子龙鲍蓉郑凌霄王洪栋鲍文正厉丹党向盈王德江
Owner XUZHOU UNIV OF TECH