This study addresses the deficiency in change detection methods for large structures within medium-resolution (MR) satellite imagery, focusing on the specific case of detecting building alterations. Adapting the successful CSA-CDGAN [6] framework originally designed for high-resolution imagery to the unique challenges of 5m resolution, we explore its efficacy in detecting changes across diverse landcovers and seasonal variations. Utilizing datasets created from VENμS satellite MR imagery, we conduct experiments demonstrating model performance improvements through transfer learning and image scaling strategies. The results showcase adaptability not only to seasonal changes but also to 10m Sentinel-2 imagery, emphasizing the approach’s robustness in detecting subtle alterations across varying medium spatial resolutions. Our findings highlight the potential of advanced deep learning techniques in enhancing change detection capabilities for MR satellite imagery, particularly in scenarios with extended temporal spans and complex environmental transformations.

We explore the concept of static-dynamic coupling--the degree to which changes in a program's static modular structure imply changes to its dynamic structure. This paper investigates the impact of static-dynamic coupling in a programming language on the effort required to evolve the coarse modular structure of programs written in that language. We performed a series of remodularization case studies in both Java and SubjectJ. SubjectJ is designed to be similar to Java, but have strictly less static-dynamic coupling. Our results include quantitative measures-time taken and number of bugs introduced--as well as a more subjective qualitative analysis of the remodularization process. All results point in the same direction and suggest that static-dynamic coupling causes substantial accidental complexity for the remodularization of Java programs.