Redirigiendo al acceso original de articulo en 20 segundos...
Inicio  /  Applied Sciences  /  Vol: 13 Par: 3 (2023)  /  Artículo
ARTÍCULO
TITULO

Tree Species Classification Based on Self-Supervised Learning with Multisource Remote Sensing Images

Xueliang Wang    
Nan Yang    
Enjun Liu    
Wencheng Gu    
Jinglin Zhang    
Shuo Zhao    
Guijiang Sun and Jian Wang    

Resumen

In order to solve the problem of manual labeling in semi-supervised tree species classification, this paper proposes a pixel-level self-supervised learning model named M-SSL (multisource self-supervised learning), which takes the advantage of the information of plenty multisource remote sensing images and self-supervised learning methods. Based on hyperspectral images (HSI) and multispectral images (MSI), the features were extracted by combining generative learning methods with contrastive learning methods. Two kinds of multisource encoders named MAAE (multisource AAE encoder) and MVAE (multisource VAE encoder) were proposed, respectively, which set up pretext tasks to extract multisource features as data augmentation. Then the features were discriminated by the depth-wise cross attention module (DCAM) to enhance effective ones. At last, joint self-supervised methods output the tress species classification map to find the trade-off between providing negative samples and reducing the amount of computation. The M-SSL model can learn more representative features in downstream tasks. By employing the feature cross-fusion process, the low-dimensional information of the data is simultaneously learned in a unified network. Through the validation of three tree species datasets, the classification accuracy reached 78%. The proposed method can obtain high-quality features and is more suitable for label-less tree species classification.

 Artículos similares

       
 
Jeonghoon Han, Han-Jun Kim, Kyun-Woo Lee and Young-Ung Choi    
In this study, we document the first recorded range extension of the genus Synaphobranchus from the East Mariana Basin in the Western Pacific Ocean. We sequenced the complete mitochondrial (mt) genome of two deep-sea eels (Synaphobranchus brevidorsalis a... ver más

 
Huihui Sun, Hao Yan, Mostafa Hassanalian, Junguo Zhang and Abdessattar Abdelkefi    
Unmanned air vehicle (UAV) systems for performing forestry applications have expanded in recent decades and have great economic benefits. They are validated to be more appealing than traditional platforms in various aspects, such as repeat rate, spatial ... ver más
Revista: Aerospace

 
Jeonghoon Han, Jong Guk Kim, O-Nam Kwon, Jordan Jun Chul Park, Kyun-Woo Lee and Young-Ung Choi    
The geoduck clam is a high-value species in South Korea. However, the molecular and morphological characteristics of geoduck clams from the southern coast of South Korea remain unexamined. The accurate identification of native geoduck clams is crucial fo... ver más

 
Yaniv Shmuel, Yaron Ziv and Baruch Rinkevich    
The three-dimensional structural complexities generated by living sessile organisms, such as trees and branching corals, embrace distinct communities of dwelling organisms, many of which are adapted to specific niches within the structure. Thus, characte... ver más

 
Dimitris Poursanidis, Kostas Kougioumoutzis, Vasileios Minasidis, Niki Chartosia, Demetris Kletou and Stefanos Kalogirou    
Biological invasions are considered among the largest threats to native biodiversity. The Mediterranean Sea, connecting the Indo-Pacific and Atlantic oceans, is characterized as a global marine invasion hotspot, due to a multitude of human pathways and v... ver más