Unlock instant, AI-driven research and patent intelligence for your innovation.

Target detection method based on combination of improved KCF and DSST for aerial photography

A target detection and combination technology, applied in the field of computer vision, can solve problems such as complex background and target scale changes, improve adaptability and solve poor tracking effects

Pending Publication Date: 2021-08-24
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Effectively solves the problem of target scale changes caused by complex background and distance changes between the target and the drone in the UAV-oriented algorithm

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
  • Target detection method based on combination of improved KCF and DSST for aerial photography
  • Target detection method based on combination of improved KCF and DSST for aerial photography
  • Target detection method based on combination of improved KCF and DSST for aerial photography

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] The technical solution of the present invention is described in detail in combination with the accompanying drawings.

[0042] An improved KCF and DSST-oriented target detection and tracking method for aerial photography of the present invention is as follows: figure 1 As shown, it specifically includes the following steps:

[0043] Step 1: Read the video and perform frame division processing to obtain the number of sequence frames and the true value of the marked frame;

[0044](1.1) Read the video sequence of the data set, perform frame division processing, and count the pictures according to the video sequence label;

[0045] (1.2) Read the annotation frame file of the video sequence to obtain the true value information of the annotation frame;

[0046] Step 2 Obtain the actual position of the tracking target in the picture frame in the first frame, extract the feature information, and construct training samples based on the form of a circulatory matrix, wherein th...

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 discloses a target detection method based on combination of improved KCF and DSST for aerial photography, and belongs to the technical field of computer vision. According to the method, in a pre-calculation stage, a video sequence is initialized, framing processing is carried out on a video, and a target is marked; in a target positioning stage, features of a target position candidate frame are extracted, a Gaussian kernel is calculated, response values of a new target template, a position predicted by a filter and positions of sampling points around the target template are calculated according to the target position, the position with the highest response value is taken as the position of a tracked target, and a tracker is updated; and finally the deviation of target motion is determined according to the peak value of the response diagram, and the target position is found for tracking. According to the invention, based on the KCF algorithm, the KCF algorithm is combined with the DSST algorithm, and multi-feature fusion is carried out. The problem of target scale change caused by distance change under the aerial photography view angle of the unmanned aerial vehicle is solved, and a multi-feature fusion and multi-scale correlation filter tracking method is realized.

Description

technical field [0001] The invention relates to a target detection method combining improved KCF and DSST for aerial photography, and belongs to the technical field of computer vision. [0002] technical background [0003] Unmanned Aerial Vehicle (UAV) has many advantages, such as simple operation, low energy consumption, and strong mobility. Therefore, it is widely used in various aspects of life: circuit inspection, map mapping, search and rescue , military reconnaissance, traffic control, civil aerial photography. In terms of search and rescue, when some natural disasters, such as earthquakes and forest fires, rescue is very difficult, drones can be allowed to enter the affected area, observe the surrounding environment, find the affected people, and avoid unnecessary injuries to rescuers; in daily traffic In the process, UAVs can also be used to observe road congestion and make reasonable road planning for citizens; In the wars between the United States and other count...

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): G06K9/00G06K9/32G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/13G06V10/25G06V10/44G06V2201/07G06N3/045G06F18/241G06F18/214
Inventor 张志宏王皓许敏捷王新华张帅
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS