An automated approach for updating land cover maps based on integrated change detection and classification methods

Abstract Updating land cover maps from remotely sensed data in a timely manner is important for many areas of scientific research. Unfortunately, traditional classification procedures are very labor intensive and subjective because of the required human interaction. Based on the strategy of updating land cover data only for the changed area, we proposed an integrated, automated approach to update land cover maps without human interaction. The proposed method consists primarily of the following three parts: a change detection technique, a Markov Random Fields (MRFs) model, and an iterated training sample selecting procedure. In the proposed approach, remotely sensed data acquired in different seasons or from different remote sensors can be used. Meanwhile, the approach is completely unsupervised. Therefore, the methodology has a wide scope of application. A case study of Landsat data was conducted to test the performance of this method. The experimental results show that several sub-modules in this method work effectively and that reasonable classification accuracy can be achieved.