Using local features to measure land development in urban regions

8 Monitoring urban development in a given region provides valuable in9 formation to researchers. Currently available, very high resolution satellite 10 images can be used for this purpose. However, manually monitoring land 11 development using these large and complex images is time consuming and 12 prone to errors. To handle this problem, an automated system is needed to 13 measure development in urban regions. Therefore, in this study we propose 14 such an automated method to measure land development in a given urban 15 region imaged in different times. We benefit from novel land development 16 measures for this purpose. They are based on local features obtained from 17 sequential images. As a novel contribution, we represent these local features 18 in a spatial voting matrix. Then, we propose five different land development 19 measures on the formed voting matrix. We test our method on 19 sets of 20 sequential panchromatic Ikonos images. Our test results indicate the possible 21 use of our method in measuring land development automatically. 22