Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Semi-artificial initial tracking frame determining and tracking method and system combined with target detection

A target detection and tracking system technology, applied in neural learning methods, character and pattern recognition, image data processing, etc., can solve the problems of tracking frame and actual position deviation, deviating from the target, etc., to achieve reduced interference, strong adaptability, Effect of improving tracking quality

Pending Publication Date: 2022-01-18
上海鹰觉科技有限公司
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Tracking drift is also a common phenomenon. During the tracking process, it is easy to gradually deviate between the tracking frame and the actual position, and finally completely deviate from the target.

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Semi-artificial initial tracking frame determining and tracking method and system combined with target detection
  • Semi-artificial initial tracking frame determining and tracking method and system combined with target detection
  • Semi-artificial initial tracking frame determining and tracking method and system combined with target detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] According to a semi-manual initial tracking frame determination and tracking method combined with target detection provided by the present invention, it includes:

[0044] Step S1: Manually mark the target with a tracking frame of a preset size;

[0045] Step S2: According to the size of the target detection result, judge whether the current target is a large ship that meets the preset requirements. If it is a large ship that meets the preset requirements, continue to use the manually marked tracking frame of the preset size for marking; if not When a large ship meets the preset requirements, the target detector is used to correct the tracking frame to obtain the final tracking frame;

[0046] Step S3: use the final tracking frame to track through the Mosse algorithm;

[0047]The target detector and the Mosse algorithm are alternately performed according to a preset ratio, and the latest detected position is used as a new tracking position. Mosse and yolov5 detection ...

Embodiment 2

[0066] Embodiment 2 is a preferred example of embodiment 1

[0067] The present invention provides a semi-manual initial tracking frame determination and tracking method combined with target detection, such as figure 2 As shown; the application scenario of the present invention is the initial position of the target, which is given artificially. The specific operation method is: manually use the mouse to click on the screen, and then draw a square initial frame with this as the center point, and the side length of the square is set in advance. The default side length is 80 pixels to accommodate ships of various sizes. However, when faced with many small boats (small sampans, small fishing boats, and pedestrians in the distance), a square with a side length of 80 pixels will include a large number of background areas. When the proportion of the background area exceeds the target pixel, the main focus of the tracker becomes the background, but ignores the target, resulting in ...

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

No PUM Login to View More

Abstract

The invention provides a semi-artificial initial tracking frame determining and tracking method and system combined with target detection. The method comprises the following steps: S1, manually marking a target by using a tracking frame with a preset size; s2, judging whether the current target is a large ship meeting the preset requirement or not according to the size of the target detection result, and when the current target is the large ship meeting the preset requirement, continuously using the manually-labeled tracking frame with the preset size for labeling; when the ship is not the large ship meeting the preset requirement, using the target detector for correcting the tracking frame, and obtaining a final tracking frame; s3, performing tracking through a Mose algorithm by using the final tracking frame; the target detector and the Mosse algorithm are alternately carried out according to a preset proportion, and the newest detection position is used as a new tracking position.

Description

technical field [0001] The invention relates to the technical field of ship target detection, in particular to a semi-manual initial tracking frame determination and tracking method combined with target detection. Background technique [0002] In the prior art, target detection refers to obtaining the position of the target (indicated in the form of a rectangular box) from a given image through an automated method. The closer the rectangular frame corresponding to the target is to the outer frame of the real target, the higher the degree of overlap means the better the detection quality. Such as figure 1 shown. Yolov5 is currently a popular target detection method. This method is based on a deep neural network, using a large number of sample pictures, and after multiple iterations of training, it achieves the effect of target detection and positioning. [0003] Object tracking is a common application of computer vision, and it is a common form of application by artificia...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/269G06V20/52G06N3/04G06N3/08
CPCG06T7/269G06N3/08G06N3/047G06N3/045
Inventor 林德银邓宏平
Owner 上海鹰觉科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com