Improved U-Net model for remote sensing image classification method based on distributed storage
- PDF / 2,487,467 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 45 Downloads / 197 Views
SPECIAL ISSUE PAPER
Improved U‑Net model for remote sensing image classification method based on distributed storage Weipeng Jing1,2 · Mingwei Zhang1,2 · Dongxue Tian1,2 Received: 18 March 2020 / Accepted: 3 October 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Aiming at the low efficiency of traditional methods for the management and classification of massive remote sensing image data, a mass remote sensing image classification method based on distributed storage is proposed. The aim is to obtain near real-time image classification in mobile devices or internet applications. In this paper, we designed two levels of an image processing structure. A distributed file system is taken as the underlying storage architecture to efficiently manage and query massive remote sensing images. The upper layer uses a GPU server to train the remote sensing image classification model to improve the classification accuracy. To improve the classification accuracy, we add two parameters to adjust the data of the current layer in U-Net. The experimental results show that the proposed method based on distributed storage has a high degree of scalability, and it has a short processing time while maintaining a high classification accuracy for remote sensing images. Keywords Distributed storage · Remote sensing image · Mobile device · Classification
1 Introduction Over the past few decades, remote sensing images have changed dramatically in terms of data quality, spatial resolution and coverage of available areas. Using remote sensing images can effectively monitor resources and obtain useful information [22]. Through an effective remote sensing classification method, accurate land cover information, number of forests, and agricultural planting conditions can be obtained. The result of image processing is usually visualized by augmented reality (AR) technology and applied to various fields, such as unmanned, urban planning, etc. In recent years, with the development of remote sensing technology, the number of remote sensing images has grown * Mingwei Zhang [email protected] Weipeng Jing [email protected] Dongxue Tian [email protected] 1
College of Information and Computer Engineering, Northeast Forestry University, Harbin, Heilongjiang, China
Key Laboratory of Forestry Data Science and Cloud Computing of State Forestry Administration, Harbin, Heilongjiang, China
2
rapidly. How to classify large quantities of remote sensing images efficiently has become an urgent problem [7]. And with the development of parallel technology and the popularity of mobile terminals, people have higher requirements for real-time access to information. Deep convolutional neural networks have shown excellent performance in various computer vision tasks, such as image recognition [3, 11], object detection [15] and classification [8, 14]. In remote sensing image classification research, traditional classification technology requires manually design features, and there are problems such as poor generalization and
Data Loading...
