Inicio  /  Applied Sciences  /  Vol: 12 Par: 8 (2022)  /  Artículo
ARTÍCULO
TITULO

Remote Sensing Scene Data Generation Using Element Geometric Transformation and GAN-Based Texture Synthesis

Zhaoyang Liu    
Renxiang Guan    
Jingyu Hu    
Weitao Chen and Xianju Li    

Resumen

Classification of remote sensing scene image (RSSI) has been broadly applied and has attracted increasing attention. However, scene classification methods based on convolutional neural networks (CNNs) require a large number of manually labeled samples as training data, which is time-consuming and costly. Therefore, generating labeled data becomes a practical approach. However, conventional scene generation based on generative adversarial networks (GANs) involve some significant limitations, such as distortion and limited size. To solve the mentioned problems, herein, we propose a method of RSSI generation using element geometric transformation and GAN-based texture synthesis. Firstly, we segment the RSSI, extracting the element information in the RSSI. Then, we perform geometric transformations on the elements and extract the texture information in them. After that, we use the GAN-based method to model and generate the texture. Finally, we fuse the transformed elements with the generated texture to obtain the generated RSSI. The geometric transformation increases the complexity of the scene. The GAN-based texture synthesis ensures the generated scene image is not distorted. Experimental results demonstrate that the RSSI generated by our method achieved a better visual effect than a GAN model. In addition, the performance of CNN classifiers was reduced by 0.44?3.41% on the enhanced data set, which is partly attributed to the complexity of the generated samples. The proposed method was able to generate diverse scene data with sufficient fidelity under conditions of small sample size and solve the accuracy saturation issues of the public scene data sets.

Palabras claves

 Artículos similares

       
 
Zhou Fang, Xiaoyong Wang, Liang Zhang and Bo Jiang    
Currently, deep learning is extensively utilized for ship target detection; however, achieving accurate and real-time detection of multi-scale targets remains a significant challenge. Considering the diverse scenes, varied scales, and complex backgrounds... ver más

 
Juan M. Soria, Juan Víctor Molner, Rebeca Pérez-González, Bárbara Alvado, Lucía Vera-Herrera and Susana Romo    
The Albufera of Valencia, a Mediterranean coastal lagoon, has been in a turbid state since 1974, with only four episodes of temporary water transparency in spring. Despite its average depth of 1 m and oligohaline waters, excessive turbidity, fish grazing... ver más

 
Zhaoyue Ma, Yong Zhao, Wenjing Zhao, Jiajun Feng, Yingying Liu, Jin Yeu Tsou and Yuanzhi Zhang    
This study on total suspended matter (TSM) in the Pearl River Estuary established a regression analysis model using Landsat 8 reflectance and measured TSM data, crucial for environmental management and engineering projects. High coefficients of determina... ver más

 
Salvatore Savastano, Paula Gomes da Silva, Jara Martínez Sánchez, Arnau Garcia Tort, Andres Payo, Mark E. Pattle, Albert Garcia-Mondéjar, Yeray Castillo and Xavier Monteys    
Coasts are continually changing and remote sensing from satellites has the potential to both map and monitor coastal change at multiple scales. Unlike optical technology, synthetic aperture radar (SAR) is uninfluenced by darkness, clouds, and rain, poten... ver más

 
Xiaohui Yan, Tianqi Zhang, Wenying Du, Qingjia Meng, Xinghan Xu and Xiang Zhao    
Water quality prediction, a well-established field with broad implications across various sectors, is thoroughly examined in this comprehensive review. Through an exhaustive analysis of over 170 studies conducted in the last five years, we focus on the a... ver más