Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A ship target tracking method based on depth learning

A target tracking and deep learning technology, applied in image data processing, instruments, climate change adaptation, etc., can solve problems such as tracking drift, high algorithm complexity, and poor adaptive ability

Active Publication Date: 2019-03-22
HANGZHOU DIANZI UNIV
View PDF8 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] 1. Features need to be manually designed, the generalization ability of the scene is insufficient, and the background information is easily ignored, resulting in tracking drift
[0005] 2. Relying on the existing detection results, the algorithm complexity is high, and the target tracking cannot be performed online
[0006] 3. The non-linear characteristics of motion are not considered, and the adaptive ability is poor

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
  • A ship target tracking method based on depth learning
  • A ship target tracking method based on depth learning
  • A ship target tracking method based on depth learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0072] The technical solutions provided by the present invention will be further described below in conjunction with the accompanying drawings.

[0073] see Figure 1-4 , the present invention provides a ship target tracking method based on deep learning, figure 1 Shown is the architecture diagram of the ship target tracking method based on deep learning in the present invention. Overall, the present invention includes 2 major steps. Step S1: use self-built ship classification and detection data sets to train and compress the deep network to obtain a Lightweight ship detection network; step S2: use the trained lightweight model to detect and track ship targets in the video;

[0074] Step S1 is based on the self-built ship classification and detection data set, and uses the method of channel pruning to iteratively train the initial feature extraction network model. Among the many parameters of the deep network, some parameters have very low weights, resulting in redundancy, wh...

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 ship target tracking method based on depth learning, belonging to the field of image-based target tracking. The invention can be applied in the fields of border and coastal defense, intelligent ocean monitoring system, ship situation estimation and the like. Step S1) training and compressing that depth network by use a self-built ship classification and detection data set to obtain a lightweight ship detection network; Step S2: Using the trained lightweight detection network model to detect the ship target in the video in real time and track the target. By adopting the technical scheme of the invention, the self-built ship data set is trained to obtain a lightweight detection network, and the ship target tracking is carried out by combining the improved trackingalgorithm framework, thereby realizing the real-time tracking of the ship target. The overall scheme has the characteristics of low equipment dependence, high tracking stability, strong real-time performance, and the like.

Description

technical field [0001] The invention relates to the field of computer vision target tracking, in particular to a ship target tracking method based on deep learning. Background technique [0002] my country's coastline is extremely long, and the types of waters are complex. All kinds of ships in trade, fishery, transportation and other industries are active in my country's waters. Under the new situation, a large amount of ship image and video data can be obtained through multi-platform, multi-sensor and other channels. These data have the characteristics of "5V+1C", that is, large capacity (Volume), diversity (Variety), timeliness (Velocity) , Accuracy (Veracity), Value (Value) and Complexity (Complexity), have brought huge challenges to the monitoring department. Therefore, there is an urgent need for an efficient automatic target tracking and recognition method in the field of ship monitoring, which can automatically track and identify ship targets in massive video data, ...

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): G06T7/277
CPCG06T2207/20081G06T2207/30236G06T7/277Y02A10/40
Inventor 刘俊王立林李亚辉孙乔林贤早
Owner HANGZHOU DIANZI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
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