Image Processing Apparatus for Sensitive Area Concealment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems for monitoring cameras fail to accurately conceal sensitive areas, such as human bodies, due to errors in detection and illumination variations, leading to potential exposure of protected regions.
Innovation Solution
An image processing apparatus that includes modules for moving object detection, human body detection, background updating, and image comparison, which generates a background image by managing stable background time and using detection information from multiple units to ensure accurate concealment of protected areas, even under varying illumination conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single detection unit is used to generate background image, then device complexity is reduced, but detection reliability deteriorates due to detection failures
Solution Approach 1:
The detection system is segmented into multiple independent detection units: a moving object detection unit that detects moving objects by comparing current images with background images, and a human body detection unit that detects human bodies using pattern recognition. These segmented detection units work together to compensate for individual detection failures, with the human body detection unit specifically addressing cases where moving object detection fails.
Solution Approach 2:
The patent merges the results from multiple detection units by integrating the moving object detection results with human body detection results. The background image generation unit combines information from both detection units to generate accurate background images, ensuring that protected areas are correctly identified even when one detection unit fails.
2Adaptability or versatility
If background image is updated frequently to adapt to illumination changes, then adaptability to illumination variations is improved, but detection precision deteriorates due to incorrect protection area identification
Solution Approach 1:
The background image update mechanism is made dynamic through the stable background time management. The system dynamically adjusts the timing of background image updates based on detection stability, using a determination unit to assess whether updates should proceed. This dynamic approach allows the system to adapt to illumination changes while preventing updates during transient variations that would compromise detection precision.
Solution Approach 2:
The system implements feedback through the determination unit that evaluates detection results before triggering background image updates. The feedback mechanism monitors detection stability and illumination variation patterns, only allowing background image updates when conditions indicate high confidence in protection area identification, thus maintaining precision while adapting to genuine illumination changes.
3Productivity
If moving object detection is used alone to identify protection areas, then processing speed is improved, but manufacturing precision of protection area identification deteriorates due to detection failures
Solution Approach 1:
The human body detection unit performs preliminary detection of human bodies using pattern recognition before the moving object detection unit processes the images. This preliminary action ensures that human bodies are identified even before motion detection occurs, allowing the system to prepare accurate protection area identification in advance and compensate for potential moving object detection failures.
Solution Approach 2:
The system applies beforehand cushioning by using human body detection results as a backup and correction mechanism for moving object detection. When moving object detection fails to identify a protection area, the human body detection results provide a cushioning effect that ensures the protection area is still correctly identified, preventing detection failures from compromising overall system accuracy.
Data Source
Figure 1
Figure 2A~2F
Figure 3A~3F
AI summary
An image processing apparatus includes, an input unit configured to input images, a first detection unit configured to detect a first area based on features in the images input by the input unit, a second detection unit configured to detect a second area based on variations between the images input by the input unit, a generation unit configured to generate a background image by using a result of the first area detection by the first detection unit and a result of the second area detection by the second detection unit, and a processing unit configured to perform image processing for reducing a visibility of a specific region identified through a comparison between a processing target image acquired after a generation of the background image by the generation unit and the background image.