Introducing Eurosat: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification

Lately, with deep learning outpacing the other machine learning techniques in classifying images, we have witnessed a growing interest of the remote sensing community in employing these techniques for the land use and land cover classification based on multispectral and hyperspectral images; the number of related publications almost doubling each year since 2015 is an attest to that. The advances in remote sensing technologies, hence the fast-growing volume of timely data available at the global scale, offer new opportunities for a variety of applications. Deep learning being significantly successful in dealing with Big Data, seems to be a great candidate for exploiting the potentials of such complex massive data. However, there are some challenges related to the ground-truth, resolution, and the nature of data that strongly impact the performance of classification. In this paper, we review the use of deep learning in land use and land cover classification based on multispectral and hyperspectral images and we introduce the available data sources and datasets used by literature studies; we provide the readers with a framework to interpret the-state-of-the-art of deep learning in this context and offer a platform to approach methodologies, data, and challenges of the field.

Download Full-text

Land use Land Cover Classifıcation using Deep Learning Classifiers for Remote Sensed Images

SSRN Electronic Journal ◽

10.2139/ssrn.3769768 ◽

2021 ◽

Author(s):

Natya S ◽

Seema Singh

Keyword(s):

Land Use ◽

Deep Learning ◽

Land Cover ◽

Land Cover Classification ◽

Land Use Land Cover ◽

Learning Classifiers

Download Full-text

Urban Land Use and Land Cover Classification Using Novel Deep Learning Models Based on High Spatial Resolution Satellite Imagery

Sensors ◽

10.3390/s18113717 ◽

2018 ◽

Vol 18 (11) ◽

pp. 3717 ◽

Cited By ~ 27

Author(s):

Pengbin Zhang ◽

Yinghai Ke ◽

Zhenxin Zhang ◽

Mingli Wang ◽

Peng Li ◽

...

Keyword(s):

Land Use ◽

Deep Learning ◽

High Resolution ◽

Land Cover ◽

Satellite Imagery ◽

Land Cover Classification ◽

Urban Land ◽

Learning Models ◽

Multi Scale ◽

Urban Land Cover

Urban land cover and land use mapping plays an important role in urban planning and management. In this paper, novel multi-scale deep learning models, namely ASPP-Unet and ResASPP-Unet are proposed for urban land cover classification based on very high resolution (VHR) satellite imagery. The proposed ASPP-Unet model consists of a contracting path which extracts the high-level features, and an expansive path, which up-samples the features to create a high-resolution output. The atrous spatial pyramid pooling (ASPP) technique is utilized in the bottom layer in order to incorporate multi-scale deep features into a discriminative feature. The ResASPP-Unet model further improves the architecture by replacing each layer with residual unit. The models were trained and tested based on WorldView-2 (WV2) and WorldView-3 (WV3) imageries over the city of Beijing. Model parameters including layer depth and the number of initial feature maps (IFMs) as well as the input image bands were evaluated in terms of their impact on the model performances. It is shown that the ResASPP-Unet model with 11 layers and 64 IFMs based on 8-band WV2 imagery produced the highest classification accuracy (87.1% for WV2 imagery and 84.0% for WV3 imagery). The ASPP-Unet model with the same parameter setting produced slightly lower accuracy, with overall accuracy of 85.2% for WV2 imagery and 83.2% for WV3 imagery. Overall, the proposed models outperformed the state-of-the-art models, e.g., U-Net, convolutional neural network (CNN) and Support Vector Machine (SVM) model over both WV2 and WV3 images, and yielded robust and efficient urban land cover classification results.

Download Full-text

EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing ◽

10.1109/jstars.2019.2918242 ◽

2019 ◽

Vol 12 (7) ◽

pp. 2217-2226 ◽

Cited By ~ 53

Author(s):

Patrick Helber ◽

Benjamin Bischke ◽

Andreas Dengel ◽

Damian Borth

Keyword(s):

Land Use ◽

Deep Learning ◽

Land Cover ◽

Land Cover Classification

Download Full-text

A Comparative Analysis of Spatial Data and Land Use/Land Cover Classification in Urbanized Areas and Areas Subjected to Anthropogenic Pressure for the Example of Poland

Sustainability ◽

10.3390/su13063070 ◽

2021 ◽

Vol 13 (6) ◽

pp. 3070

Author(s):

Patrycja Szarek-Iwaniuk

Keyword(s):

Land Use ◽

Land Cover ◽

Land Cover Classification ◽

Classification Systems ◽

Anthropogenic Pressure ◽

Land Use Land Cover ◽

Land Use Classification ◽

Land Cover Data ◽

Spatial Changes ◽

Land Use Types

Urbanization processes are some of the key drivers of spatial changes which shape and influence land use and land cover. The aim of sustainable land use policies is to preserve and manage existing resources for present and future generations. Increasing access to information about land use and land cover has led to the emergence of new sources of data and various classification systems for evaluating land use and spatial changes. A single globally recognized land use classification system has not been developed to date, and various sources of land-use/land-cover data exist around the world. As a result, data from different systems may be difficult to interpret and evaluate in comparative analyses. The aims of this study were to compare land-use/land-cover data and selected land use classification systems, and to determine the influence of selected classification systems and spatial datasets on analyses of land-use structure in the examined area. The results of the study provide information about the existing land-use/land-cover databases, revealing that spatial databases and land use and land cover classification systems contain many equivalent land-use types, but also differ in various respects, such as the level of detail, data validity, availability, number of land-use types, and the applied nomenclature.

Download Full-text