Deep Feature-Based Matching of High-Resolution Multitemporal Images Using VGG16 and VGG19 Algorithms
摘要
This research focuses on feature-based, high-resolution multi-temporal image matching. The objective is to develop an efficient method for accurately matching features in images captured at different times. The proposed approach relies on the Visual Geometry Group-16 (VGG16) and VGG19 models for feature extraction and calculates Euclidean distances to enable accurate matching. A comparative analysis is carried out with Scale Invariant Feature Transform (SIFT) to evaluate the performance of both models. Experimental results demonstrate the effectiveness of the proposed approach in processing high-resolution multi-temporal imagery, highlighting the advantages of using the VGG19 and VGG16 models in feature-based matching techniques.