Title: Land cover blending: A new framework to generate high spatial and temporal resolution land cover maps from remotely sensed images
Abstract: The development of remote sensing has enabled the acquisition of land cover classes and their changes at different scales. The high-spatial-resolution images are usually acquired infrequently, whereas the low-spatial-resolution images which have high repetition rates cannot capture the land cover spatial detail information. A novel spatial-temporal land cover blending method (STLCB) is proposed to produce land cover maps at both high spatial and temporal resolutions, using a single or a series of low-spatial-resolution images and two high-spatial-resolution land cover maps which pre-date and post-date the low-spatial-resolution images as input. A spatial-temporal Markov-random-field based method, which integrates spatial and temporal links of pixels, is proposed in STLCB. The proposed STLCB method is validated based on synthetic and Landsat multi-spectral images. Results show that the overall accuracies of STLCB were higher than 90% in both experiments.
Publication Year: 2016
Publication Date: 2016-07-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot