Facial Image Verification With Stored Templates Against Catfishing

The verification system is segmented into distinct functional modules: a facial feature extractor module that processes captured images to extract identifying features, an image template database that stores reference templates, and a match processing module that compares features and determines verification results. This modular segmentation allows each component to specialize in a specific task, improving overall reliability while managing complexity through clear separation of concerns.

The patent introduces an intermediary verification system that acts as a mediator between the user's captured image and the authentication decision. The match processing module serves as an intermediary that objectively compares extracted features against stored templates using algorithmic matching, eliminating subjective human judgment and reducing the risk of fraud. This intermediary layer enhances reliability by providing consistent, automated verification.

The system performs preliminary classification by automatically routing verified images to a verified gallery and unverified images to an unverified gallery upon receipt. This preliminary action eliminates the need for manual sorting and status checking later, as images are pre-categorized based on verification results. The match processing module immediately determines verification status and directs images to appropriate storage locations, saving time in subsequent image management operations.

The verification system operates autonomously to manage image classification without requiring manual intervention. The match processing module automatically compares features, determines verification status, and routes images to the appropriate gallery (verified or unverified) based on the matching results. This self-service capability precision-tracking of image status while minimizing the time users need to spend on manual image management.

The system extracts only the essential identifying features from facial images rather than processing entire high-resolution images. The facial feature extractor module identifies and extracts key biometric characteristics (such as facial geometry, feature positions, and patterns) that are sufficient for verification. This extraction approach significantly reduces the data volume requiring comparison while maintaining verification accuracy, thereby reducing energy consumption during the matching process.

The verification process focuses computational resources on analyzing specific local features of the face (eyes, nose, mouth, facial contours) rather than uniformly processing the entire image. The facial feature extractor module applies localized analysis to identify distinguishing characteristics in key facial regions, allocating processing power efficiently to areas that provide the most verification value. This local quality approach improves accuracy by concentrating on discriminative features while minimizing overall energy expenditure.