Confidential Text Redaction in Bitmap Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing screen-content sharing applications fail to adequately address the security needs of enterprise users by inadvertently leaking or maliciously exfiltrating confidential information, as traditional techniques are not well-suited for detecting and classifying confidential information in bitmap images or video, especially when text is encoded as pixel intensity values rather than explicit text formats.
Innovation Solution
Implementing a process that uses optical character recognition (OCR) to extract text from bitmap images, analyzes the text using pattern matching and natural language processing, applies a redaction filter to detected confidential information, and calculates a risk score, while also processing video frames by breaking them into segments, filtering out non-relevant frames, and applying the analysis and redaction algorithms to each unique frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional screen-content sharing applications are used, then ease of operation is improved, but security and confidentiality deteriorate due to leakage of confidential information
Solution Approach 1:
The system performs preliminary actions by detecting, classifying, and redacting confidential information in screen captures before they are transmitted or stored. The process includes: (1) capturing screen content, (2) detecting text regions using OCR, (3) classifying text as confidential or non-confidential, and (4) redacting confidential information before sharing. This preliminary processing prevents information leakage while maintaining the ease of screen sharing functionality.
2Reliability
If confidential information is detected and redacted in bitmap images, then security is improved, but device complexity increases due to multiple processing steps
Solution Approach 1:
The system segments the complex security processing task into distinct modular steps: (1) image capture, (2) optical character recognition to extract text, (3) pattern matching to identify confidential information, (4) classification of confidential vs. non-confidential text, and (5) redaction of identified information. Each step can be independently implemented and optimized, reducing overall system complexity while maintaining high security standards.
3Measurement precision
If optical character recognition and pattern matching are applied to detect confidential text, then detection precision is improved, but loss of time increases due to processing requirements
Solution Approach 1:
The system applies local quality by focusing computational resources only on regions and characteristics that are most likely to contain confidential information. Instead of processing every pixel uniformly, the system: (1) uses OCR to identify text regions, (2) applies pattern matching only to extracted text, and (3) classifies only potentially confidential text. This targeted approach improves detection precision for confidential information while reducing overall processing time compared to uniform analysis of entire images.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Effectively detects and redacts confidential information in both images and video, reducing the risk of information leakage and providing a risk scoring system to monitor and manage sharing behavior, thus enhancing security and compliance in enterprise environments.
Implementation Method 1
causing optical character recognition (OCRing) of text in the bitmap image and obtaining, as a result of the OCRing, an OCR record with text appearing in the bitmap image
Data Source
AI summary
Provided is a process, including: receiving a screen capture event from an operating system of a first client computing device of a first user, the screen capture event including, or being associated with, a bitmap image of at least part of a display of the first computing device; causing optical character recognition (OCRing) of text in the bitmap image; classifying each of the n-grams into two or more categories, the two or more categories including a category for confidential information; and for each of the n-grams classified in the category for confidential information, obfuscating the respective n-gram in the bitmap image to form a modified version of the bitmap image.