Imaging Apparatus Exposure Control Using Face and Subject Region Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging apparatuses face fluctuations in exposure when face detection fails, leading to unstable image capture due to sudden changes in exposure settings.
Innovation Solution
An imaging apparatus equipped with a face recognizer and a head recognizer that continues exposure compensation based on metering in the detected face region even when face detection is lost, using the head recognition information to maintain consistent exposure settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face detection is used to control exposure, then exposure accuracy for the subject is improved, but exposure fluctuation occurs when face detection fails
Solution Approach 1:
The system performs preliminary face detection and stores face region information when the face is detected. When face detection subsequently fails, the stored preliminary detection results are reused to maintain exposure compensation, preventing exposure fluctuations while preserving subject-focused exposure accuracy.
Solution Approach 2:
The patent introduces an intermediate mechanism where the system detects both face regions and broader subject regions. When face detection fails, the subject region detection serves as an intermediary that maintains the exposure compensation chain by identifying the subject through alternative means, thus bridging the gap between face detection failure and exposure control.
2Manufacturing precision
If exposure compensation is continuously performed based on face region, then subject exposure quality is improved, but exposure instability occurs when face region is lost
Solution Approach 1:
The system dynamically switches between face-based exposure compensation and subject-based exposure compensation. When the face region is detected, face-based compensation is applied for high quality. When face detection fails, the system transitions to subject-based compensation to maintain stability, creating a dynamic adaptation mechanism that optimizes both quality and consistency.
Solution Approach 2:
The system prepares alternative exposure control mechanisms in advance. Subject region detection and exposure calculation are performed as a backup mechanism that activates when face detection fails, cushioning against exposure instability before it can manifest as visible fluctuation in the captured images.
3Measurement precision
If face detection algorithm is used, then exposure control accuracy is improved, but detection reliability deteriorates when subject looks aside
Solution Approach 1:
The system implements multi-functionality by incorporating both face detection and subject region detection capabilities. The face detection algorithm provides accurate exposure control when the face is visible, while the subject region detection provides universal detection coverage that works regardless of face orientation, allowing the system to maintain reliability across different subject poses.
Solution Approach 2:
The system creates an alternative detection representation by implementing subject region detection as a backup to face detection. When face detection fails (e.g., subject looks aside), the subject region detection copy provides the necessary detection information to maintain exposure control, ensuring detection reliability without sacrificing the accuracy benefits of face detection when available.
Data Source
AI summary
An imaging apparatus comprises: an imager that captures an image of a subject to generate image data; a first detector that detects a face region corresponding to a face of the subject in the image data; a second detector that detects a subject region corresponding to at least a part of the subject in the image data; and a controller that controls exposure based on metering for the image data, wherein the controller performs, when the face region is detected by the first detector, exposure compensation according to metering in the detected face region, and continues the exposure compensation according to the metering in the detected face region, when the detected face region is lost by the first detector and the subject region corresponding to the subject of the detected face region is detected by the second detector.