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

An improved ship target detection method based on YOLO V2

A target detection and ship technology, applied in the fields of deep learning and computer vision, can solve the problem that the detection effect of small targets is not very good, and achieve the effect of satisfying real-time monitoring and solving poor results.

Inactive Publication Date: 2019-01-08
SHENYANG LIGONG UNIV
View PDF5 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The target detection method based on SSD (Single ShotMultibox Detector) also directly returns the position and category of the target, and the detection speed and detection accuracy are not satisfactory, but the detection effect on small targets is not very good

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
  • An improved ship target detection method based on YOLO V2
  • An improved ship target detection method based on YOLO V2
  • An improved ship target detection method based on YOLO V2

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Step 1: Make a dataset according to the VOC dataset format

[0038]Create the folder Boat-detection used to store the data set, and generate three folders under the Boat-detection folder, named Annotations, ImageSets, and JPEGImages folders. Adjust the ship image data format to .jpg format, and rename the image data from 000001.jpg according to the official naming method of PASCAL VOC, and store the image data in the JPEGImages folder. Annotate the image data, that is, mark the category and location information of the target, save the annotation information as a .xml file with the same name, and store it in the Annotations folder. Generate the training sample set and test sample set from the existing data in proportion, generate train.txt and test.txt files, which store the absolute path information of the training sample set and test sample set, and put the .txt file into the ImageSets folder In the Main folder, the structure diagram of the data production and storage...

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

An improved ship target detection method based on YOLO V2 comprises the following steps: (1) making a ship data set according to a VOC data set format; (2) taking an improved Darknet-19 Network as thebasic network to serve as a YOLO V2 feature extractor and a classification network, adding additional auxiliary network layer in the improved Darknet-19 to form a complete YOLO V2 target detection network; (3) training the model based on the network structure; (4) using the trained model to detect the ship target and evaluate the model. Effect 1: the method replaces the traditional non-imaging technology and imaging technology of ship detection and avoids the influence of external interference on ship detection, and the real-time and accuracy based on YOLO V2 can meet the requirements of real-time monitoring and accurate detection of marine monitoring system. Effect 2: aiming at the problem of small target detection, the network is improved to solve the problem of small target detection in the field of target detection to a large extent.

Description

technical field [0001] The invention relates to the fields of computer vision, deep learning technology and the like, and specifically designs an improved ship target detection method based on YOLO V2. Background technique [0002] The ocean is closely related to human life and is closely related, and it is of great strategic importance and significance in the military. In recent years, my country's maritime security situation has become increasingly complex. Other countries have violated my country's waters, illegal fishing is serious, and maritime security issues occur frequently. Marine supervision and real-time detection of intruding ships have become crucial. However, the detection accuracy and speed of traditional methods cannot meet the requirements, so a faster and more accurate detection method is urgently needed. [0003] There are two main existing traditional ship detection methods: non-imaging technology and imaging technology detection. Non-imaging technology...

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): G06K9/00G06K9/62
CPCG06V20/00G06V2201/07G06F18/24G06F18/214
Inventor 陈亮刘韵婷于洋宋建辉李世杰
Owner SHENYANG LIGONG UNIV
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