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

Visual target tracking method based on dense convolutional network feature

A convolutional network and target tracking technology, applied in the field of target tracking, can solve problems such as limitations and restrictions on the wide application of tracking algorithms, and achieve the effects of reducing pollution, practicability, and good robustness

Active Publication Date: 2020-01-14
CHANGAN UNIV +1
View PDF8 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the limitation of the accuracy and speed of the current tracking algorithm, the wide application of the tracking algorithm in practical scenarios such as video surveillance, intelligent transportation, and human-computer interaction is restricted.

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
  • Visual target tracking method based on dense convolutional network feature

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The technical solutions of the present invention will be described in further detail below with reference to the accompanying drawings and embodiments.

[0036] Such as figure 1 As shown, a kind of visual object tracking method based on dense convolution network feature provided by the present invention comprises the following steps:

[0037] Step 1. Build the initial position filter:

[0038] 1. In the first frame of image that has been manually marked, the center position and size of the target have been clarified. According to the target location, select a suitable ROI region (Region of Interest, region of interest).

[0039] 2. In the dense convolutional network (DenseNet), five layers are selected to perform feature extraction on the ROI area of ​​the first frame image, and five different features are obtained:

[0040] The names of the five selected layers are: 'conv1|relu', 'conv2_block6_concat', 'conv3_block12_concat', 'conv4_block48_concat' and 'conv5_block3...

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 belongs to the technical field of target tracking, and particularly relates to a visual target tracking method based on dense convolutional network features. According to the technical scheme provided by the invention, the method comprises the following five steps: step 1, constructing an initial position filter; step 2, constructing an initial scale filter; step 3, performing targetpositioning; step 4, performing scale estimation; step 5, updating the model. Target features are extracted through different layers of a deep dense convolutional network, an optimal response graph is adaptively selected based on an APCE (Average Peak-to-correlation Energy) value, and then the center position of a target is determined. On the basis, the target scale is further estimated. The method can adapt to the change of the target scale, the size of the target is accurately determined, and the model is selectively updated on line. The method is high in algorithm precision and success rate, achieves the robust tracking of the target, and is easy to popularize and apply in an actual scene.

Description

technical field [0001] The invention belongs to the technical field of target tracking, and in particular relates to a visual target tracking method based on dense convolution network features. Background technique [0002] As an important research field of artificial intelligence, computer vision aims to make machines have vision similar to human beings. Computer vision uses images (videos) as input to study image information organization, object and scene recognition, etc., and then explain events, so as to realize the expression and understanding of the environment. The field of computer vision includes many different research directions, such as object detection, semantic segmentation, object tracking, 3D reconstruction, action recognition, etc. [0003] As one of the basic problems in the field of computer vision, object tracking has always been a research hotspot in this field. Target tracking refers to the detection, extraction and identification of moving targets i...

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
IPC IPC(8): G06T7/246
CPCG06T7/246G06T2207/20024G06T2207/20084
Inventor 马素刚侯志强惠飞赵祥模孙韩林王忠民
Owner CHANGAN UNIV