Progressive Data Augmentation Method for Remote Sensing Ship Image Classification Based on Imaging Simulation System and Neural Style Transfer

نویسندگان

چکیده

AbstractDeep learning has shown great power in processing remote sensing data, especially for fine-grained ship image classification. However, the lack of a large amount effective training data greatly limits performance neural networks. Based on current augmentation methods, images ships sea generated have problem distortion, blurring, and poor diversity. To tackle this problem, we propose novel progressive method that combines simulation samples style transfer (NST) based network to generate transferred images. Our consists two stages. The first stage uses visible light imaging system (VLISS) through 3D models real This can significantly increase diversity dataset. For second stage, eliminate domain gap between samples, few newly designed NST-based called Sim2RealNet are employed realize from proposed was applied variety targets verify its effectiveness compared other methods classification tasks. experimental results demonstrate method. Index TermsDomain gap, classification, (NST), sensing, samples.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Comparison Studies on Classification for Remote Sensing Image Based on Data Mining Method

Data mining methods have been widely applied on the area of remote sensing classification in recent years. In these methods, neural network, rough sets and support vector machine (SVM) have received more and more attentions. Although all of them have great advantages on dealing with imprecise and incomplete data, there exists essential difference among them. Until now, researches of these three...

متن کامل

Remote Sensing Image Classification Based on Gray System Theory

The intelligence and automation of image processing and analysis is a bottle problem for photogrammetry and remote sensing. Artificial neural networks is a new solver which imitates brain and gray system theory is a new tool which handles undetermined problem. This paper describe how to combine artificial neural networks with gray system theory to realize classification of remote sensing image ...

متن کامل

An Unsupervised Classification Method for Hyperspectral Remote Sensing Image Based on Spectral Data Mining

© 2012 Wen and Yang, licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. An Unsupervised Classification Method for Hyperspectral Remote Sensing Image Based on ...

متن کامل

Novel Artificial Neural Networks For Remote-Sensing Data Classification

This paper discusses two novel artificial neural network architectures applied to multi-class classification problems of remote-sensing data. These approaches are 1) a spiking-neural-network model for the partitioning of data into clusters, and 2) a neuron model based on complex-valued weights (CVN). In the former model, the learning process is based on the Spike Timing-Dependent Plasticity rul...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

سال: 2021

ISSN: ['2151-1535', '1939-1404']

DOI: https://doi.org/10.1109/jstars.2021.3109600