Abstract
Local climate zones (LCZ) have become a generic criterion for climate analysis among global cities, as they can describe not only the urban climate but also the morphology inside the city. LCZ mapping based on the remote sensing classification method is a fundamental task, and the protocol proposed by the World Urban Database and Access Portal Tools (WUDAPT) project, which consists of random forest classification and filter-based spatial smoothing, is the most common approach. However, the classification and spatial smoothing lack a unified framework, which causes the appearance of small, isolated areas in the LCZ maps. In this paper, a spatial-contextual information-based self-training classification framework (SCSF) is proposed to solve this LCZ classification problem. In SCSF, conditional random field (CRF) is used to integrate the classification and spatial smoothing processing into one model and a self-training method is adopted, considering that the lack of sufficient expert-labeled training samples is always a big issue, especially for the complex LCZ scheme. Moreover, in the unary potentials of CRF modeling, pseudo-label selection using a self-training process is used to train the classifier, which fuses the regional spatial information through segmentation and the local neighborhood information through moving windows to provide a more reliable probabilistic classification map. In the pairwise potential function, SCSF can effectively improve the classification accuracy by integrating the spatial-contextual information through CRF. The experimental results prove that the proposed framework is efficient when compared to the traditional mapping product of WUDAPT in LCZ classification.
Highlights
The local climate zone (LCZ) scheme is a novel climate-based classification scheme [1], which skillfully relates the urban climate represented by physical traits with urban morphology depicted through landscape cover
To test the performance of the proposed self-training classification framework (SCSF) method, three experiments were conducted, each using two Landsat 8 images and one ground truth provided as part of the 2017 Geoscience and Remote Sensing Society (GRSS) Data Fusion Contest (2017DFC)
The random forest (RF) classification was used as a baseline, the widely accepted LCZ mapping workflow proposed by World Urban Database and Access Portal Tools (WUDAPT) (comprising RF classification and majority-filter(MJ)-based spatial smoothing) was used as another reference framework, denoted as RF+MJ(WUDAPT), and the conditional random fields (CRF) for LCZ classification was denoted as RF+CRF
Summary
The local climate zone (LCZ) scheme is a novel climate-based classification scheme [1], which skillfully relates the urban climate represented by physical traits with urban morphology depicted through landscape cover. 2019, 11, 2828 in unreliable inputs for modeling the potential function in CRF Considering this fact, we propose an improved self-training method with spatial information-based pseudo-label selection to provide a more reliable probabilistic classification map. The spatial-contextual information based self-training classification framework for LCZs. Considering that the traditional remote-sensing-based LCZ mapping approaches separate the classification and the spatial smoothing into two independent components, which causes the small, isolated areas in LCZ maps, the proposed SCSF method adopts CRF to directly integrate the spatial-contextual information into the classification with a unified theoretical basis. The filter-based spatial smoothing uses spatial constraints to mitigate this problem, the small, isolated areas still remain due to the lack of a unified framework between the classification and spatial smoothing steps Given this fact, a spatial-contextual information-based method—CRF—is applied to describe the relationship between the samples and labels through the unary potentials and it simultaneously models the spatial correlation between the labeled and observed data by the pairwise potentials. The LCZ classification framework integrating classification and spatial-contextual information is effective even with limited data
Sign-in/Register to view highlights & summary 
Published Version (
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