Two-stream feature aggregation deep neural network for scene classification of remote sensing images

2020 ◽  
Vol 539 ◽  
pp. 250-268
Author(s):  
Kejie Xu ◽  
Hong Huang ◽  
Peifang Deng ◽  
Guangyao Shi
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Deep Neural Network ◽  
Scene Classification ◽  
Remote Sensing Images ◽  
Feature Aggregation

scholarly journals Lifting Scheme-Based Deep Neural Network for Remote Sensing Scene Classification

2019 ◽  
Vol 11 (22) ◽  
pp. 2648 ◽  
Author(s):  
Chu He ◽  
Zishan Shi ◽  
Tao Qu ◽  
Dingwen Wang ◽  
Mingsheng Liao
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Filter Bank ◽  
Deep Neural Network ◽  
Hardware Implementation ◽  
Fundamental Problem ◽  
Lifting Scheme ◽  
Scene Classification ◽  
Remote Sensing Images ◽  
Network Applications

Recently, convolutional neural networks (CNNs) achieve impressive results on remote sensing scene classification, which is a fundamental problem for scene semantic understanding. However, convolution, the most essential operation in CNNs, restricts the development of CNN-based methods for scene classification. Convolution is not efficient enough for high-resolution remote sensing images and limited in extracting discriminative features due to its linearity. Thus, there has been growing interest in improving the convolutional layer. The hardware implementation of the JPEG2000 standard relies on the lifting scheme to perform wavelet transform (WT). Compared with the convolution-based two-channel filter bank method of WT, the lifting scheme is faster, taking up less storage and having the ability of nonlinear transformation. Therefore, the lifting scheme can be regarded as a better alternative implementation for convolution in vanilla CNNs. This paper introduces the lifting scheme into deep learning and addresses the problems that only fixed and finite wavelet bases can be replaced by the lifting scheme, and the parameters cannot be updated through backpropagation. This paper proves that any convolutional layer in vanilla CNNs can be substituted by an equivalent lifting scheme. A lifting scheme-based deep neural network (LSNet) is presented to promote network applications on computational-limited platforms and utilize the nonlinearity of the lifting scheme to enhance performance. LSNet is validated on the CIFAR-100 dataset and the overall accuracies increase by 2.48% and 1.38% in the 1D and 2D experiments respectively. Experimental results on the AID which is one of the newest remote sensing scene dataset demonstrate that 1D LSNet and 2D LSNet achieve 2.05% and 0.45% accuracy improvement compared with the vanilla CNNs respectively.

scholarly journals An Analysis of State-of-the-art Activation Functions For Supervised Deep Neural Network

2021 ◽  
Author(s):  
Anh Nguyen ◽  
Khoa Pham ◽  
Dat Ngo ◽  
Thanh Ngo ◽  
Lam Pham
Keyword(s):  
Neural Network ◽  
Supervised Classification ◽  
Deep Neural Network ◽  
State Of The Art ◽  
Network Architectures ◽  
Activation Functions ◽  
Scene Classification ◽  
Learning Network ◽  
Deep Learning Network

This paper provides an analysis of state-of-the-art activation functions with respect to supervised classification of deep neural network. These activation functions comprise of Rectified Linear Units (ReLU), Exponential Linear Unit (ELU), Scaled Exponential Linear Unit (SELU), Gaussian Error Linear Unit (GELU), and the Inverse Square Root Linear Unit (ISRLU). To evaluate, experiments over two deep learning network architectures integrating these activation functions are conducted. The first model, basing on Multilayer Perceptron (MLP), is evaluated with MNIST dataset to perform these activation functions.Meanwhile, the second model, likely VGGish-based architecture, is applied for Acoustic Scene Classification (ASC) Task 1A in DCASE 2018 challenge, thus evaluate whether these activation functions work well in different datasets as well as different network architectures.

scholarly journals Scene classification of remote sensing images based on hierarchical sparse coding

2018 ◽  
Vol 2018 (16) ◽  
pp. 1650-1657
Author(s):  
Xu Jiaqing ◽  
Lv Qi ◽  
Liu Hongjun ◽  
He Jie
Keyword(s):  
Remote Sensing ◽  
Sparse Coding ◽  
Scene Classification ◽  
Remote Sensing Images

scholarly journals Learning Multi-Modality Features for Scene Classification of High-Resolution Remote Sensing Images

2018 ◽  
Vol 06 (11) ◽  
pp. 185-193
Author(s):  
Feng’an Zhao ◽  
Xiongmei Zhang ◽  
Xiaodong Mu ◽  
Zhaoxiang Yi ◽  
Zhou Yang
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Scene Classification ◽  
Remote Sensing Images

scholarly journals Automatic Detection of Flood Using Remote Sensing Images

2020 ◽  
Vol 02 (01) ◽  
pp. 11-26 ◽  
Author(s):  
Amith Chandrakant Chawan ◽  
Vaibhav K Kakade ◽  
Jagannath K Jadhav
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Convolutional Neural Network ◽  
Detection System ◽  
Deep Convolutional Neural Network ◽  
Remote Sensing Images ◽  
Hazardous Area ◽  
High Resolution Images ◽  
Flood Detection

Remote sensing imaging (RSI) technology has recently been identified as an effective photogrammetric data acquisition platform to rapidly provide high resolution images due to its profitability, its ability to fly at low altitude and the ability to analysis in dangerous areas. The various kinds of classification techniques are have been used for flood extent mapping for finding the flood affected region, but based on the color region based analysis the classified hazardous area has very complex. Due to over the above issues in this work there significant enhancements have appeared in the classification of remote sensing images using Contiguous Deep Convolutional Neural Network (CDCNN).In the flood detection system the four different kinds of process like preprocessing, segmentation, feature extraction and the Contiguous Deep Convolutional Neural Network (CDCNN) has been executed for identifying the flood defected region. This works also investigates and compare with the possible methods with the proposed CDCNN for accurately identified by the Classification details of the RSI

Sign in / Sign up

Export Citation Format

Share Document