PROJECT TITLE :

A Novel Automatic Change Detection Method for Urban High-Resolution Remotely Sensed Imagery Based on Multiindex Scene Representation

ABSTRACT:

The new generation of Earth observation sensors with high spatial resolution can offer detailed info for modification detection. The widely used strategies for prime-resolution image amendment detection rely on textural/structural options. But, these spatial options forever produce high-dimensional information area since they're connected to a series of parameters, e.g., window sizes and directions. Machine learning ways also are commonly employed, however their performances are subject to the number and quality of the coaching samples, and hence, a lot of effort should be created to collect the high-quality samples. To handle these issues, during this study, a completely unique multiindex automatic change detection technique is proposed for the high-resolution imagery. The notable benefits of the proposed model embody the following: 1) Complicated urban scenes are represented by a group of low dimensional but semantic info indexes, replacing the high-dimensional however low-level features (e.g., textural and structural features), and a couple of) the amendment detection model is administrated automatically without using training samples since the data indexes will directly indicate the primitive urban classes. The multiindex representation refers to the enhanced vegetation index, the water index, and also the recently developed morphological building index. Experiments were conducted on the multitemporal WorldView-two images over Shenzhen City (south of China) and Kuala Lumpur (the capital of Malaysia), where promising results were achieved by the proposed technique. Moreover, the ancient ways primarily based on the state-of-the-art textural/morphological features were also implemented for the purpose of comparison, that further validates the advantages of our proposed model.