Image Sensitive Information Protection via Geometric Template Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for protecting sensitive information in images, such as ID cards and bank cards, face challenges in accurately identifying and masking sensitive content due to the difficulty in semantic identification and the variability of image content, especially when images are defective.
Innovation Solution
A method and system that receive an image, identify its boundaries, match the image with a predetermined template to determine the specific information region, and perform processing on that region to protect the sensitive information, without requiring detailed semantic content analysis, using techniques like image machine learning and boundary detection to improve matching accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed semantic content analysis is performed to identify sensitive information in images, then the accuracy of content identification is improved, but the complexity of the processing system and the time required increase significantly
Solution Approach 1:
The patent extracts only the essential geometric features (boundaries, corners, edges) from the image instead of performing complete semantic analysis. By taking out only the necessary structural elements needed for template matching, the system achieves effective sensitive information identification without the complexity of full content understanding.
Solution Approach 2:
The patent segments the image processing task into distinct stages: boundary detection, corner point identification, edge extraction, and template matching. This segmentation allows each stage to focus on specific geometric features rather than attempting comprehensive semantic analysis, reducing overall system complexity while maintaining effectiveness.
2Reliability
If comprehensive semantic analysis is performed on image content to identify sensitive information, then the reliability of information protection is improved, but the processing time increases
Solution Approach 1:
The patent performs preliminary action by pre-defining templates that represent sensitive information patterns. During actual processing, the system only needs to match these pre-established templates against extracted geometric features rather than performing comprehensive semantic analysis, significantly reducing processing time while maintaining protection reliability.
Solution Approach 2:
The patent replaces the complex mechanical process of semantic content analysis with a more efficient geometric matching system. By substituting detailed content understanding with boundary and corner point matching against templates, the system achieves fast processing without sacrificing the reliability of sensitive information identification.
3Adaptability or versatility
If the system attempts to identify and match various types of sensitive content patterns in images, then the adaptability to different information types is improved, but the difficulty of detecting and measuring increases
Solution Approach 1:
The patent creates a universal template matching system that can handle multiple types of sensitive information (ID cards, bank cards, certificates, etc.) through a single geometric feature extraction and matching approach. This universal method adapts to different content types without requiring specialized detection techniques for each, reducing detection difficulty while maintaining versatility.
Data Source
AI summary
The present invention discloses a method for protecting specific information, comprising: receiving an image containing specific information; identifying boundaries of the image; matching the image with a predetermined image template based on the boundaries so as to determine a specific information region of the image; and performing predetermined processing on the specific information region of the image to be presented in response to determining the specific information region of the image, so as to protect the specific information. The method of the invention for protecting specific information can efficiently protect relevant specific information without identifying detailed semantic content in an image.