Video Masking Apparatus for Privacy Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video masking techniques fail to securely and efficiently protect personal information in recorded videos, as they require original image data for unmasking, leading to potential leaks of personal information when unmasked in normal video players.
Innovation Solution
A video masking processing method and apparatus that detects and masks areas containing personal information, such as faces and vehicle number plates, within video frames, inserts masking information into the frames, and stores the masked frames, allowing secure reproduction with a generated key for unmasking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a H.264 codec-based dynamic privacy masking/unmasking technique is used to enable original image restoration according to authority, then the ability to restore original images is improved, but personal information may be leaked when the codec allowing unmasking is used in normal video players
Solution Approach 1:
The patent extracts and separates the masking information from the original video data by embedding it as metadata in the video file. This allows the masking data to be independently managed and protected, while the original video remains intact. The masking information includes coordinates and identification data that can be used to selectively restore or protect specific regions without exposing the entire original image in unauthorized contexts.
Solution Approach 2:
The patent applies masking to specific regions of the video before storage or transmission. By pre-identifying sensitive areas (faces, license plates) and applying masking algorithms to these regions, the system ensures that personal information is protected from the outset. The masking information is embedded in advance, allowing authorized systems to later restore these regions without requiring access to the original unmasked video data.
2Reliability
If automatic masking of vehicle number plates and persons' faces is implemented using pattern recognition and image processing, then personal information protection is improved, but the complexity of the video processing system increases
Solution Approach 1:
The patent implements an automatic masking system that autonomously identifies sensitive regions in video frames using pattern recognition algorithms. The system automatically detects faces, license plates, and other personally identifiable information, applies appropriate masking treatments, and embeds the masking information without requiring manual intervention. This self-service capability reduces operational complexity while maintaining high protection standards.
Solution Approach 2:
The patent employs image processing techniques that analyze video frame parameters (pixel patterns, color distributions, geometric features) to automatically identify and mask sensitive regions. By changing and analyzing multiple parameters simultaneously, the system achieves accurate detection and masking without requiring complex manual configuration or processing pipelines.
3Reliability
If masking information is embedded in the video frame to store area masking information, then the ability to restore masked areas is improved, but the storage requirements and data processing load increase
Solution Approach 1:
The patent embeds masking information within the video file structure as metadata, nesting this auxiliary data within the existing video container format. This approach allows the masking information (coordinates, region identifiers, masking type) to be stored alongside the video data without requiring separate storage systems. The nested structure enables efficient access and processing while minimizing additional storage overhead.
Data Source
AI summary
A video masking processing method and a video masking processing apparatus using the same are provided. The video masking processing method photographs frames constituting a video, detects at least one masking area to be masked in the photographed frames, masks an image in the detected at least one masking area, inserts masking information including information on the masking into the photographed frame, and stores the frame which has undergone the masking and has the masking information inserted thereinto. Therefore, since images related to personal information in a photographed video are masked, a user does not infringe others' privacy in a video normally photographed by a CCTV or a dashboard camera.