Abstract
Abstract. Understanding building area in agricultural land is important since arable land area in Taiwan is limited. One of the practical ways is manual digitization on high resolution satellite imagery, which can avoid field investigation and achieve satisfying results. However, such practice is tedious and labor intensive. Past researches have shown that deep learning methods are useful to segment buildings in different cities using satellite imagery. In this study, ENVINet5 model was trained and used to segment buildings from high resolution Pleiades pansharpened imagery. The training images (with the size of 2500 pixels × 2500 pixels) were randomly selected from 9 counties/cities to increase diversity since each county/city has different building patterns. The performance of ENVINet5 model reached 0.977, 0.814, 0.847, and 0.829 respectively on accuracy, precision, recall, and F1 score. Since evaluation by pixels can be difficult to show geometry of buildings, we evaluated the model by counting the number of inference building segments, which was post-processed from inference result of ENVINet5 trained model. Further analysis by counting the inference building segments is discussed in this study.
Highlights
Building segmentation in agricultural land is an important issue in Taiwan since arable land area can be influenced by building area
omission error rate (OER) and commission error rate (CER) varied with the different settings of probability threshold from inference results
We demonstrated the feasibility of deep learning approach to segment buildings automatically in agricultural land using high resolution Pleiades pansharpened imagery
Summary
Building segmentation in agricultural land is an important issue in Taiwan since arable land area can be influenced by building area. One of the practical ways to estimate building area in agricultural land is to digitize them manually on high resolution satellite imagery, which is rich in spatial information and helpful for building visual interpretation. Deep learning approaches have been applied to segment buildings automatically on satellite imagery in many studies (Boonpook et al, 2018; Maltezos et al, 2017; Vakalopoulou et al, 2015). Results on building segmentation can be different depending on various building patterns, and deep learning models (Zhang et al, 2020). Blur Distance helps the model to learn building borders by blurring the edges and decreasing the blur during training
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
