Automatic Detection and Extraction Framework of Ground Control Points for Remotely Sensed Data

In remote sensing applications particularly, the Ground Control Points (GCPs) are not always available for the purpose of image processing. Indeed, in inaccessible regions, traditional methods of acquiring accurate GCPs, such as ground surveying or GPS measurements, may not be feasible. Identifying GCPs in difficult-to-reach areas can pose significant challenges due to limited ground access and logistical constraints. This study introduces an alternative method for automating the detection and extraction of GCPs from remote-sensing satellite imagery. The method consists of two main phases: Data pre-processing and GCPs identification and extraction. In the first phase, the study focuses on enhancing image quality by utilizing band selection and contrast enhancement techniques. In the second phase, the identification and extraction of GCPS is established by combining SURF and SIFT algorithms. The proposed framework is tested to ALSAT2 satellite images in challenging scenarios. These findings indicate the effectiveness of the proposed method in accurately identifying and extracting GCPs for remote sensing satellite imagery. The findings show a 36.5% rise in the quantity of identified GCPs using the proposed technique in contrast to the traditional SURF algorithm across all datasets.