SAR image ship target detection and recognition integrated method based on deep learning

A deep learning and target detection technology, applied in scene recognition, character and pattern recognition, instruments, etc., can solve problems such as unsatisfactory application, inability to achieve practicality in rough classification, and inability to achieve automatic recognition, and achieve correct ship recognition. The effect of improving the rate and improving the classification accuracy

Pending Publication Date: 2020-07-03
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF3 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, slice-based classification methods are not ideal for practical scenarios
At present, the ship target automatic recognition method based on SAR images can integrate ship target detection and rough classification. Rough classification refers to dividing ships into large ships and small ships according to the size of the ship. Obviously, rough classification cannot meet the practical requirements.
Therefore, the ship target in SAR image based on deep learning cannot realize automatic i...

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
  • SAR image ship target detection and recognition integrated method based on deep learning
  • SAR image ship target detection and recognition integrated method based on deep learning
  • SAR image ship target detection and recognition integrated method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The specific embodiments of the present invention will be described in further detail below in conjunction with the drawings and embodiments.

[0032] Such as figure 1 As shown, the integrated detection and recognition process of the present invention includes:

[0033] Step 1. Production of SAR image ship data set

[0034] The SAR image ship data comes from the Sentinel-1 data set of the OpenSARship platform. First, select the three types of ship pictures from the Sentinel-1 data set to make uniform SAR image slices. At the same time, the slice is marked, that is, the specific position of the ship target is drawn with a rectangular frame on the slice, and the category is marked. Then divide the data set into training set and test set according to the proportion.

[0035] Step 2: Build a deep neural network

[0036] The deep network proposed by the present invention uses the RetinaNet network as the network framework, and mainly includes three parts: ResNet50, Feature Pyramid...

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 radar remote sensing application, and particularly relates to an SAR image ship target detection and recognition integrated method based on deep learning. An important part of SAR image ship target interpretation is detection and identification, but all detection and identification are independent at present, and a detection and identification integrated method is an important research direction of SAR image interpretation in the future. Existing traditional methods and deep learning methods for target detection and recognition cannot achieve detection and recognition integration of SAR image ship targets. The invention provides an SAR image ship target detection and identification integrated method. Through an existing deep learning networkframework, a network used for SAR image ship target detection and identification integration is provided. Detection and classification tasks of ship targets are carried out at the same time by mainlyutilizing a detection and classification sub-network at the tail end of a network, so that a detection and identification integrated target is realized. Compared with a traditional ATR technology, thedetection and recognition process of the method is simpler and more efficient.

Description

Technical field [0001] The invention belongs to the technical field of radar remote sensing application, and specifically relates to an integrated method for detecting and identifying ship targets in SAR images based on deep learning. Background technique [0002] Synthetic Aperture Radar (Synthetic Aperture Radar, hereinafter referred to as SAR) is an active sensor that uses microwave imaging. Compared with optical and infrared sensors, SAR can work around the clock and has a certain penetration capability. Therefore, SAR technology has been widely used in civil and military fields. With the development of SAR imaging technology, the field of SAR image target detection and recognition faces many opportunities and challenges. The detection and recognition of ship targets in SAR images is an important part of modern maritime intelligent monitoring systems, so the interpretation of ship targets in SAR images is a current research hotspot. [0003] The traditional SAR image automati...

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): G06K9/00G06K9/32G06N3/04
CPCG06V20/13G06V10/25G06N3/045
Inventor 曹宗杰候泽生崔宗勇杨建宇
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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
Try Eureka
PatSnap group products