Abstract
Over the past decade, many excellent data sharing efforts have enriched the remote sensing scene classification (SC) methods. These datasets have achieved great success in complex high-level semantic information interpretation. However, most existing datasets are collected from standard and ungeoreferenced image patches for algorithm training and evaluation. These datasets do not fit for practical applications and cannot be directly applied in further geographical study. Accordingly, we provide a large range high-resolution SC dataset with multiple time phases, called “ <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">W</b> u <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">h</b> an <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</b> ulti <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a</b> pplication <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V</b> HR <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b> cene classification dataset (WH-MAVS).” It facilitates the study of SC and scene change detection (SCD) algorithms. Moreover, it can also be directly employed to perform a variety of real-life land use application tasks. To the best of our knowledge, this is the first free, publicly available, georeferenced, and annotated dataset to cover almost an entire megacity. The WH-MAVS was collected and annotated from Google Earth imagery with the same spatial resolution and uniform nonoverlapping patch size, covering the central area of Wuhan, China. The total number of scene samples is 47 137, which belong to 14 classes with 23 567 labeled patches for each time phase in 2014 and 2016, respectively. The geographic coordinates of all samples in both time phases exhibit one-to-one correspondence with 23 202 unchanged image patches of scene categories and 365 changed ones. The distribution of the number of samples in each class is highly imbalanced; moreover, there are large intraclass differences and indistinguishable interclass variances. These characteristics are closer to the real land use/land cover application tasks and introduce further challenges to the related algorithm research. In addition, we conducted benchmark experiments on SC and SCD based on the WH-MAVS dataset with widely used deep learning models. DenseNet169 was found to achieve the best performance. The overall accuracies are 91.07% and 92.09%, respectively, in the 2014 and 2016 validation sets of WH-MAVS. Furthermore, SCD obtained by DenseNet169 has a binary change detection accuracy of 89.56% and a multiple (from–to) change detection accuracy of 86.70%. Over and above the research value of the algorithm, it is also proven to have practical applications in fields such as urban planning, landscape pattern analysis, and urban dynamic monitoring and analysis.
Highlights
I N recent years, given the dramatic development of remote sensing (RS) satellites, image spatial resolution has been increased to the sub-meter level, enabling RS images to obtain detailed textures and plentiful structural information [1,2,3,4,5,6]
We present the first large-range dataset capable of bridging the gap between algorithms and applications and pushing down the barriers of datasets between different tasks, such as Scene classification (SC) and Scene change detection (SCD)
A set of application-oriented classification criteria was developed in accordance with the relevant classification reference for urban planning
Summary
I N recent years, given the dramatic development of remote sensing (RS) satellites, image spatial resolution has been increased to the sub-meter level, enabling RS images to obtain detailed textures and plentiful structural information [1,2,3,4,5,6]. Recent RSSC datasets focus primarily on land-use mapping at a single point in time; there is no wide-ranged multi-temporal RS scene dataset designed for studying SCD, SC with temporal correlation, or updating scene maps These problems mentioned above, indicating that it is difficult to directly apply existing datasets from the algorithmic level of RS scene interpretation to the level of real LULC-related applications, is known as the “application gap” of RS scene interpretation. The present work makes the following major contributions: 1) We construct the first novel multi-application large-range VHR SC dataset labeled, with 14 categories based on high-level semantics for urban functions, and bridge the application gap between the algorithms and real-life applications. The source code and our dataset will be made open-source for future academic research
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
