Video Person Detection With Permission-List Privacy Blurring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video surveillance systems lack the ability to selectively focus on specific individuals while preserving privacy, particularly for those who do not require close monitoring.

Innovation Solution

An image processing apparatus that detects individuals in video frames, determines if they are on a permission list, and applies treatment processing to make them harder to identify, while maintaining visibility of others.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If video surveillance monitors all persons in a facility, then surveillance coverage is comprehensive, but privacy of individuals not of interest is compromised

Engineering Contradiction:

Improvesurveillance coverageVSAvoidprivacy infringement

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent applies different processing treatments to different persons in the video based on their identification status. Persons matching the permission list receive face blurring treatment, while others receive no treatment. This local differentiation resolves the contradiction by protecting privacy only for those who need protection (permission list persons) while maintaining full surveillance capability for others.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The surveillance system segments persons into two categories: those on the permission list and those not on the list. By segmenting the treatment applied to different persons, the system can comprehensively monitor all individuals while selectively protecting the privacy of specific groups, thus resolving the contradiction between comprehensive surveillance and privacy protection.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If face recognition is applied to all persons, then identification accuracy is high, but processing time and computational load increase

Engineering Contradiction:

Improveidentification accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs face recognition and permission list matching in advance for persons detected in the video. Once a person is identified and matched against the permission list, the decision to blur their face is made immediately. This preliminary identification action allows the system to achieve high identification accuracy for those who need it while avoiding unnecessary processing time for persons who will be blurred anyway.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent extracts only the necessary identification information (face features) from persons in the video to check against the permission list, rather than performing full detailed analysis on all persons. This extraction approach maintains identification accuracy for permission list verification while reducing overall processing time and computational load.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS20250252782A1Image processing apparatus, control method, and non-transitory storage medium

Publication Date: 2025.08.07 NEC CORP

AI summary

An image processing apparatus (2000) acquires a video frame (14), and detects a person from the acquired video frame (14). The image processing apparatus (2000) determines whether the detected person (30) satisfies a predetermined condition based on a permission list (40). The image processing apparatus (2000) performs, on an image region in the video frame (14) representing the person (30) determined to satisfy the predetermined condition, first treatment processing of setting the person (30) in a hardly identified state. The predetermined condition includes a condition that the person (30) is included in the permission list (40).