Monitoring the green evolution of vernacular buildings based on deep learning and multi-temporal remote sensing images

The increasingly mature computer vision (CV) technology represented by convolutional neural networks (CNN) and available high-resolution remote sensing images (HR-RSIs) provide opportunities to accurately measure the evolution of natural artificial environments on Earth at a large scale. Based advanced CNN method net (HRNet) multi-temporal HR-RSIs, framework is proposed for monitoring green courtyard buildings characterized their courtyards being roofed (CBR). consists an expert module focusing scenes analysis, CV automatic detection, evaluation containing thresholds, output data analysis. this, changes in adoption different CBR technologies (CBRTs), including light-translucent CBRTs (LT-CBRTs) non-light-translucent (NLT-CBRTs), 24 villages southern Hebei were identified from 2007 2021. was featured as inverse S-curve, differences found stage, ratio, development speed villages. LT-CBRTs are dominant type but replaced surpassed NLT-CBRTs some villages, characterizing preferences research provides reference vernacular buildings, laws enable trace predict particular village. This work lays foundation future exploration occurrence mechanism phenomenon important optimization promotion CBRTs.