Deep convolutional neural network based SAR image sea ice sorting method

A deep convolution, neural network technology, applied in the field of sea ice monitoring, to achieve the effect of strong operability, meeting real-time requirements, and short processing time

Active Publication Date: 2017-12-26
SHANGHAI OCEAN UNIV
View PDF7 Cites 40 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0013] Therefore, there is an urgent need for a classification method for SAR image sea ice classification that uses convolutional neural net

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
  • Deep convolutional neural network based SAR image sea ice sorting method
  • Deep convolutional neural network based SAR image sea ice sorting method
  • Deep convolutional neural network based SAR image sea ice sorting method

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0099] Example 1

[0100] refer to figure 1 , figure 1 It is a schematic diagram of the sea ice classification method based on deep convolutional neural network in SAR images. A kind of SAR image sea ice classification method based on depth convolutional neural network of the present invention comprises the following steps:

[0101] S01: Segment existing sea ice SAR images

[0102] S011: Segment existing sea ice SAR images into regional maps of different sea ice types;

[0103] S012: Cut the area map into multiple sample sizes of different sizes, and use them as original training samples;

[0104] S02: Perform data preprocessing

[0105] S021: Labeling the original training samples;

[0106] S022: Convert data storage format;

[0107] S023: data averaging processing;

[0108] S03: Conduct model training and build a model

[0109] S031: Construct a convolutional neural network model (CNN model);

[0110] S032: Divide the sample data into a training set and a test set;...

Example Embodiment

[0122] Example 2

[0123] A kind of SAR image sea ice classification method based on deep convolutional neural network of the present invention, its specific working procedure is:

[0124] S01: Segment existing sea ice SAR images

[0125] S011: Segment the existing sea ice SAR images into regional maps of different sea ice types according to the interpretation map (Ice Chart) marked by experts;

[0126] S012: Cut the area map into multiple sample sizes of different sizes, and use them as original training samples;

[0127] S02: Perform data preprocessing

[0128] S021: Labeling the original training samples;

[0129] S022: Convert data storage format;

[0130] S023: data averaging processing;

[0131] S03: Conduct model training and build a model

[0132] S031: Construct a convolutional neural network model;

[0133] S032: Divide the sample data into a training set and a test set according to different sizes;

[0134] S033: Send the training set and the test set to the...

Example Embodiment

[0174] Example 3

[0175] refer to figure 2 , figure 2 is the process of convolutional neural network. A specific implementation of the convolutional neural network model of the SAR image sea ice classification method based on the deep convolutional neural network of the present invention is as follows:

[0176] A01: Construct a convolutional neural network model, including 3 convolutional layers, 3 pooling layers, 1 fully connected layer, and Softmax loss layer;

[0177] Its parameters are as follows:

[0178] The convolution kernel size of the convolution layer is 5×5, the convolution kernel quantity of the first convolution layer is 32, the convolution kernel quantity of the second convolution layer is 32, and the convolution kernel quantity of the third convolution layer is 32. The number of convolution kernels in the product layer is 64;

[0179] The pooling layer filter is 3×3, and the step size is 2;

[0180] The activation function is a ReLu function;

[0181]...

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 relates to a deep convolutional neural network based SAR image sea ice sorting method. The method includes steps of S01, segmenting a prior sea ice SAR image; S02, performing data pre-treatment; S03, performing model training and establishing a model; S04, performing treatment on a to-be-sorted sea ice SAR image; S05, combining sorting results. The method is advantaged in that the deep convolutional neural network model constructed in the method can realize characteristic automatic extraction based on images and excessive manual intervention is not needed; the method is an end-to-end SAR image sea ice sorting method and can reach sea ice monitoring service level and meet requirements on real time performance for offshore workers; image characteristics can be extracted automatically depending on a large amount of mark samples by the model and reliance on specialist knowledge is not needed; convergence is accelerated by utilizing a random gradient descending method and model training condition can be judged according to a loss function and accuracy; problems of gradient vanish or gradient diffusion in network parameter optimization counter-propagation are solved by utilizing an normative approach.

Description

technical field [0001] The invention relates to the technical field of sea ice monitoring, in particular to a SAR image sea ice classification method based on a deep convolutional neural network. Background technique [0002] About 3-4% of the world's oceans are covered by sea ice. On the one hand, sea ice can have an important impact on global climate, heat balance, and water balance; Serious obstacles and even catastrophic disasters, so all countries have carried out close monitoring of sea ice. [0003] At present, the main task of sea ice monitoring is to use sea ice images to create bitmaps that can indicate the geographical distribution of different types of sea ice, that is, sea ice interpretation, including inversion of sea ice types, sea ice area, sea ice maximum edge line, A series of information such as the density of sea ice. The classification of sea ice images is the process of marking all image pixels as different sea ice types according to their characteris...

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): G06T7/11G06N3/04G06K9/62
CPCG06T7/11G06T2207/30181G06T2207/20081G06T2207/10044G06N3/045G06F18/24G06F18/214
Inventor 黄冬梅宋巍李明慧杜艳玲贺琪郑小罗李瑶
Owner SHANGHAI OCEAN 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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap