GUI-Enabled Boundary Regularization System for Urban Buildings Using the Tkinter

Building detection contributes to monitoring changes in land use and land cover, providing insights into urbanization trends and environmental impacts. This study focuses on the development of a user-friendly graphical user interface (GUI) using the Tkinter library in Python. It begins with the preprocessing of the dataset, involving resizing, noise removal, and enhancement techniques. Subsequently, the processed dataset undergoes building detection using the Mask R-CNN technique, enabling real-time identification of buildings within satellite images. To refine the precision of building boundaries, a Guided Filter is applied to enhance the accuracy of the output masks. The research further employs two images captured at distinct timelines to conduct change detection, revealing alterations in building structures over time. The results encompass the presentation of predicted buildings within the GUI, achieved by providing the test image to the model. A subsequent step involves change detection, wherein buildings are classified into three distinct categories: demolished, unchanged, and changed. The GUI, which facilitates the display and exploration of images generated by a proposed model, presents an effective approach for studying changes in land use and cover.