Feature Extraction Method for Land Consolidation from High Resolution Imagery

Land consolidation is a tool for increasing the area of the arable land and improving the effectiveness of land cultivation. With the development of high resolution image, the progress of land consolidation project can be monitored by acquiring information from the image objectively. This paper presents a method to extract the wells and roads in land consolidation project from high resolution images. The well extraction method is based on the gray-level template matching algorithm. The road extraction method is based on mathematical morphology, which is a method for detecting image components that are useful for representation and description. The vector planning maps and high resolution images used to monitor the completion of land consolidation project are registered. The candidate areas are created using the functions of buffer and extraction by mask in GIS. The well template is selected manually from the image. The template is used to find the wells which match the template perfectly. In the road extraction section, Top-hat transform and gray dilation are used to filter the noise of the image. In this way the road feature in the image became wider and even more obvious to be recognized. Then image binarization and thinning algorithm are used to extract the one-pixel centerline of the road. At last, the thinning results are converted to the final vector detection results. Key-Words: Feature extraction, Mathematical morphology, Template matching, Land consolidation, Remote sensing