scholarly journals Integration of Convolutional Neural Networks and Object-Based Post-Classification Refinement for Land Use and Land Cover Mapping with Optical and SAR Data

2019 ◽  
Vol 11 (6) ◽  
pp. 690 ◽  
Author(s):  
Shengjie Liu ◽  
Zhixin Qi ◽  
Xia Li ◽  
Anthony Yeh
Keyword(s):  
Neural Networks ◽  
Land Use ◽  
Land Cover ◽  
Convolutional Neural Networks ◽  
Spatial Resolution ◽  
Spatial Information ◽  
Object Based ◽  
Sar Data ◽  
The University ◽  
Image Object

Object-based image analysis (OBIA) has been widely used for land use and land cover (LULC) mapping using optical and synthetic aperture radar (SAR) images because it can utilize spatial information, reduce the effect of salt and pepper, and delineate LULC boundaries. With recent advances in machine learning, convolutional neural networks (CNNs) have become state-of-the-art algorithms. However, CNNs cannot be easily integrated with OBIA because the processing unit of CNNs is a rectangular image, whereas that of OBIA is an irregular image object. To obtain object-based thematic maps, this study developed a new method that integrates object-based post-classification refinement (OBPR) and CNNs for LULC mapping using Sentinel optical and SAR data. After producing the classification map by CNN, each image object was labeled with the most frequent land cover category of its pixels. The proposed method was tested on the optical-SAR Sentinel Guangzhou dataset with 10 m spatial resolution, the optical-SAR Zhuhai-Macau local climate zones (LCZ) dataset with 100 m spatial resolution, and a hyperspectral benchmark the University of Pavia with 1.3 m spatial resolution. It outperformed OBIA support vector machine (SVM) and random forest (RF). SVM and RF could benefit more from the combined use of optical and SAR data compared with CNN, whereas spatial information learned by CNN was very effective for classification. With the ability to extract spatial features and maintain object boundaries, the proposed method considerably improved the classification accuracy of urban ground targets. It achieved overall accuracy (OA) of 95.33% for the Sentinel Guangzhou dataset, OA of 77.64% for the Zhuhai-Macau LCZ dataset, and OA of 95.70% for the University of Pavia dataset with only 10 labeled samples per class.

scholarly journals Dual and Single Polarized SAR Image Classification Using Compact Convolutional Neural Networks

2019 ◽  
Vol 11 (11) ◽  
pp. 1340 ◽  
Author(s):  
Mete Ahishali ◽  
Serkan Kiranyaz ◽  
Turker Ince ◽  
Moncef Gabbouj
Keyword(s):  
Neural Networks ◽  
Land Use ◽  
Land Cover ◽  
Convolutional Neural Networks ◽  
Systematic Approach ◽  
Land Cover Classification ◽  
Human Interaction ◽  
Land Use Land Cover ◽  
Intensity Data ◽  
Sar Data

Accurate land use/land cover classification of synthetic aperture radar (SAR) images plays an important role in environmental, economic, and nature related research areas and applications. When fully polarimetric SAR data is not available, single- or dual-polarization SAR data can also be used whilst posing certain difficulties. For instance, traditional Machine Learning (ML) methods generally focus on finding more discriminative features to overcome the lack of information due to single- or dual-polarimetry. Beside conventional ML approaches, studies proposing deep convolutional neural networks (CNNs) come with limitations and drawbacks such as requirements of massive amounts of data for training and special hardware for implementing complex deep networks. In this study, we propose a systematic approach based on sliding-window classification with compact and adaptive CNNs that can overcome such drawbacks whilst achieving state-of-the-art performance levels for land use/land cover classification. The proposed approach voids the need for feature extraction and selection processes entirely, and perform classification directly over SAR intensity data. Furthermore, unlike deep CNNs, the proposed approach requires neither a dedicated hardware nor a large amount of data with ground-truth labels. The proposed systematic approach is designed to achieve maximum classification accuracy on single and dual-polarized intensity data with minimum human interaction. Moreover, due to its compact configuration, the proposed approach can process such small patches which is not possible with deep learning solutions. This ability significantly improves the details in segmentation masks. An extensive set of experiments over two benchmark SAR datasets confirms the superior classification performance and efficient computational complexity of the proposed approach compared to the competing methods.

Hybrid object-based approach for land use/land cover mapping using high spatial resolution imagery

2011 ◽  
Vol 25 (6) ◽  
pp. 1025-1043 ◽  
Author(s):  
Eva Savina Malinverni ◽  
Anna Nora Tassetti ◽  
Adriano Mancini ◽  
Primo Zingaretti ◽  
Emanuele Frontoni ◽  
...  
Keyword(s):  
Land Use ◽  
Land Cover ◽  
Spatial Resolution ◽  
High Spatial Resolution ◽  
Land Cover Mapping ◽  
Land Use Land Cover ◽  
Object Based ◽  
High Spatial Resolution Imagery

Exploring the synergistic use of multi-scale image object metrics for land-use/land-cover mapping using an object-based approach

2015 ◽  
Vol 36 (13) ◽  
pp. 3544-3562 ◽  
Author(s):  
Ning Han ◽  
Huaqiang Du ◽  
Guomo Zhou ◽  
Xiaojun Xu ◽  
Hongli Ge ◽  
...  
Keyword(s):  
Land Use ◽  
Land Cover ◽  
Land Cover Mapping ◽  
Land Use Land Cover ◽  
Multi Scale ◽  
Object Based ◽  
Image Object

scholarly journals TOWARDS BETTER CLASSIFICATION OF LAND COVER AND LAND USE BASED ON CONVOLUTIONAL NEURAL NETWORKS

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 139-146
Author(s):  
C. Yang ◽  
F. Rottensteiner ◽  
C. Heipke
Keyword(s):  
Neural Networks ◽  
Land Use ◽  
Land Cover ◽  
Convolutional Neural Networks ◽  
Land Cover Classification ◽  
Aerial Images ◽  
Infrared Band ◽  
Rgb Images ◽  
Geospatial Databases

<p><strong>Abstract.</strong> Land use and land cover are two important variables in remote sensing. Commonly, the information of land use is stored in geospatial databases. In order to update such databases, we present a new approach to determine the land cover and to classify land use objects using convolutional neural networks (CNN). High-resolution aerial images and derived data such as digital surface models serve as input. An encoder-decoder based CNN is used for land cover classification. We found a composite including the infrared band and height data to outperform RGB images in land cover classification. We also propose a CNN-based methodology for the prediction of land use label from the geospatial databases, where we use masks representing object shape, the RGB images and the pixel-wise class scores of land cover as input. For this task, we developed a two-branch network where the first branch considers the whole area of an image, while the second branch focuses on a smaller relevant area. We evaluated our methods using two sites and achieved an overall accuracy of up to 89.6% and 81.7% for land cover and land use, respectively. We also tested our methods for land cover classification using the Vaihingen dataset of the ISPRS 2D semantic labelling challenge and achieved an overall accuracy of 90.7%.</p>

scholarly journals Land use land cover classification using local multiple pattern from very high resolution satellite imagery

2014 ◽  
Vol XL-8 ◽  
pp. 971-976 ◽  
Author(s):  
R. Suresh Kumar ◽  
A. R. Mahesh Balaji
Keyword(s):  
Land Use ◽  
Land Cover ◽  
Spatial Resolution ◽  
Spatial Information ◽  
High Spatial Resolution ◽  
Spectral Variation ◽  
Land Use Land Cover ◽  
Texture Information ◽  
Very High Spatial Resolution ◽  
Very High

The recent development in satellite sensors provide images with very high spatial resolution that aids detailed mapping of Land Use Land Cover (LULC). But the heterogeneity in the landscapes often results in spectral variation within the same and spectral confusion among different LU/LC classes at finer spatial resolution. This leads to poor classification performances based on traditional spectral-based classification. Many studies have been addressed to improve this classification by incorporating texture information with multispectral images. Although different methods are available to extract textures from the satellite images, only a limited number of studies compared their performance in classification. The major problem with the existing texture measures is either scale/orientation/illumination variant (Haralick textures) or computationally difficult (Gabor textures) or less informative (Local binary pattern). This paper explores the use of texture information captured by Local Multiple Patterns (LMP) for LULC classification in a spectral-spatial classifier framework. LMP preserve more structural information and involves less computational efforts. Thus LMP is expected to be more promising for capturing spatial information from very high spatial resolution images. The proposed method is implemented with spectral bands and LMP derived from WorldView-2 multispectral imagery acquired for Madurai, India study area. The Multi-Layer-Perceptron neural network is used as a classifier. The proposed classification method is compared with LBP and conventional Maximum Likelihood Classification (MLC) separately. The classification results with 89.5% clarify the improvement offered by the LMP for LULC classification in comparison with the conventional approaches.

scholarly journals A Framework for Evaluating Land Use and Land Cover Classification Using Convolutional Neural Networks

2019 ◽  
Vol 11 (3) ◽  
pp. 274 ◽  
Author(s):  
Manuel Carranza-García ◽  
Jorge García-Gutiérrez ◽  
José Riquelme
Keyword(s):  
Neural Networks ◽  
Land Use ◽  
Land Cover ◽  
Convolutional Neural Networks ◽  
Experimental Comparison ◽  
Support Vector ◽  
K Nearest Neighbors ◽  
Digital Pattern ◽  
High Level ◽  
Social Applications

Analyzing land use and land cover (LULC) using remote sensing (RS) imagery is essential for many environmental and social applications. The increase in availability of RS data has led to the development of new techniques for digital pattern classification. Very recently, deep learning (DL) models have emerged as a powerful solution to approach many machine learning (ML) problems. In particular, convolutional neural networks (CNNs) are currently the state of the art for many image classification tasks. While there exist several promising proposals on the application of CNNs to LULC classification, the validation framework proposed for the comparison of different methods could be improved with the use of a standard validation procedure for ML based on cross-validation and its subsequent statistical analysis. In this paper, we propose a general CNN, with a fixed architecture and parametrization, to achieve high accuracy on LULC classification over RS data from different sources such as radar and hyperspectral. We also present a methodology to perform a rigorous experimental comparison between our proposed DL method and other ML algorithms such as support vector machines, random forests, and k-nearest-neighbors. The analysis carried out demonstrates that the CNN outperforms the rest of techniques, achieving a high level of performance for all the datasets studied, regardless of their different characteristics.

scholarly journals Comparing Deep Neural Networks, Ensemble Classifiers, and Support Vector Machine Algorithms for Object-Based Urban Land Use/Land Cover Classification

2019 ◽  
Vol 11 (14) ◽  
pp. 1713 ◽  
Author(s):  
Shahab Eddin Jozdani ◽  
Brian Alan Johnson ◽  
Dongmei Chen
Keyword(s):  
Neural Networks ◽  
Land Use ◽  
Land Cover ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Urban Land Use ◽  
Gradient Boosting ◽  
Support Vector ◽  
Land Use Land Cover ◽  
Object Based

With the advent of high-spatial resolution (HSR) satellite imagery, urban land use/land cover (LULC) mapping has become one of the most popular applications in remote sensing. Due to the importance of context information (e.g., size/shape/texture) for classifying urban LULC features, Geographic Object-Based Image Analysis (GEOBIA) techniques are commonly employed for mapping urban areas. Regardless of adopting a pixel- or object-based framework, the selection of a suitable classifier is of critical importance for urban mapping. The popularity of deep learning (DL) (or deep neural networks (DNNs)) for image classification has recently skyrocketed, but it is still arguable if, or to what extent, DL methods can outperform other state-of-the art ensemble and/or Support Vector Machines (SVM) algorithms in the context of urban LULC classification using GEOBIA. In this study, we carried out an experimental comparison among different architectures of DNNs (i.e., regular deep multilayer perceptron (MLP), regular autoencoder (RAE), sparse, autoencoder (SAE), variational autoencoder (AE), convolutional neural networks (CNN)), common ensemble algorithms (Random Forests (RF), Bagging Trees (BT), Gradient Boosting Trees (GB), and Extreme Gradient Boosting (XGB)), and SVM to investigate their potential for urban mapping using a GEOBIA approach. We tested the classifiers on two RS images (with spatial resolutions of 30 cm and 50 cm). Based on our experiments, we drew three main conclusions: First, we found that the MLP model was the most accurate classifier. Second, unsupervised pretraining with the use of autoencoders led to no improvement in the classification result. In addition, the small difference in the classification accuracies of MLP from those of other models like SVM, GB, and XGB classifiers demonstrated that other state-of-the-art machine learning classifiers are still versatile enough to handle mapping of complex landscapes. Finally, the experiments showed that the integration of CNN and GEOBIA could not lead to more accurate results than the other classifiers applied.

scholarly journals Land Use Classification of the Deep Convolutional Neural Network Method Reducing the Loss of Spatial Features

Sensors ◽  
2019 ◽  
Vol 19 (12) ◽  
pp. 2792 ◽  
Author(s):  
Xuedong Yao ◽  
Hui Yang ◽  
Yanlan Wu ◽  
Penghai Wu ◽  
Biao Wang ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Neural Networks ◽  
Land Use ◽  
Convolutional Neural Networks ◽  
Spatial Information ◽  
Semantic Segmentation ◽  
Land Use Classification ◽  
Deep Convolutional Neural Networks ◽  
Remote Sensing Imagery ◽  
Spatial Features

Land use classification is a fundamental task of information extraction from remote sensing imagery. Semantic segmentation based on deep convolutional neural networks (DCNNs) has shown outstanding performance in this task. However, these methods are still affected by the loss of spatial features. In this study, we proposed a new network, called the dense-coordconv network (DCCN), to reduce the loss of spatial features and strengthen object boundaries. In this network, the coordconv module is introduced into the improved DenseNet architecture to improve spatial information by putting coordinate information into feature maps. The proposed DCCN achieved an obvious performance in terms of the public ISPRS (International Society for Photogrammetry and Remote Sensing) 2D semantic labeling benchmark dataset. Compared with the results of other deep convolutional neural networks (U-net, SegNet, Deeplab-V3), the results of the DCCN method improved a lot and the OA (overall accuracy) and mean F1 score reached 89.48% and 86.89%, respectively. This indicates that the DCCN method can effectively reduce the loss of spatial features and improve the accuracy of semantic segmentation in high resolution remote sensing imagery.

Sign in / Sign up

Export Citation Format

Share Document