Accurate image matching, especially for high-resolution images, remains a significant challenge in computer vision. This paper presents a novel graph-based approach to improve high-resolution image matching by leveraging Scale-Invariant Feature Transform (SIFT) features and machine learning (ML)-based distance metrics. We construct graphs where nodes represent SIFT nodes and edges encode spatial relationships, capturing richer contextual information than traditional feature-based methods, we employ a K-Nearest Neighbors (KNN) algorithm with a learned distance metric. To enhance feature comparison, we introduce Machine Learning (ML) based distance metrics specifically tailored for graph-structured data. Experimental results demonstrate the superiority of our proposed method in handling complex scenes, occlusions, and various image transformations. By evaluating the precision of keypoint matches and comparing our graph-based approach with the Oriented FAST and Rotated BRIEF (ORB) algorithm, we observed significant improvements in matching accuracy.