Eye Size-Based Autofocus Region Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional camera imaging devices with face-priority AF require manual focus adjustment to target the eyes, lacking an automatic solution that effectively focuses on eyes regardless of size or position, especially when multiple people are photographed.
Innovation Solution
An imaging device and AF control method that includes a facial organ detection section, eye selection, size comparative determination, and AF region setting to automatically focus on the eyes, determining the appropriate AF region based on eye size and position, ensuring accurate focus regardless of the number or size of individuals in the frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If face-priority AF is used to automatically focus on the face, then automatic focusing is achieved, but the ability to automatically focus on the eyes is lost
Solution Approach 1:
The patent segments the face detection and focusing process into multiple stages: first detecting the face, then detecting eyes within the face, and finally selecting the appropriate eye for focusing. This segmentation allows the system to maintain automatic focusing while specifically targeting the eyes, resolving the contradiction between automatic face focusing and manual eye adjustment requirements.
Solution Approach 2:
The patent applies local quality by shifting the focusing target from the entire face to specific local regions (eyes). The system detects eye positions and sizes within the face and uses this localized information to determine the AF region, enabling automatic eye focusing while maintaining the overall automatic focusing capability.
2Device complexity
If the entire face is used for focus adjustment, then automatic focusing is simplified, but precision for eye focusing is reduced
Solution Approach 1:
The patent divides the face into functional regions, specifically identifying eye locations within the face. By segmenting the detection process into face detection followed by eye detection, the system achieves both simplified automatic focusing and precise eye targeting without increasing overall system complexity.
Solution Approach 2:
The patent transitions from two-dimensional face detection to three-dimensional spatial reasoning by calculating eye positions, sizes, and relative distances. This dimensional expansion enables precise eye focusing while maintaining the simplicity of the overall automatic focusing process through algorithmic rather than hardware complexity.
3Extent of automation
If eye size is used as a selection criterion, then automatic eye selection is achieved, but accuracy for small eyes is reduced
Solution Approach 1:
The patent changes the selection parameters from solely eye size to a composite evaluation including eye position, eye size, and distance from the camera. This parameter expansion allows the system to automatically select eyes while maintaining accuracy for small eyes by considering their positional and depth information rather than relying on size alone.
Solution Approach 2:
The patent implements feedback by continuously evaluating multiple eye characteristics (size, position, distance) and adjusting the AF region selection based on this feedback. The system uses the detected eye properties to iteratively refine the focusing target, ensuring accurate selection even for small eyes that might be missed by size-only criteria.
4Measurement precision
If AF region is set based on eye position and size, then focusing accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary actions by detecting face and eye positions, calculating eye sizes, and determining relative distances before final AF region selection. By completing these detection and calculation tasks in advance, the system reduces real-time processing requirements and achieves high focusing accuracy without significant time penalty.
Solution Approach 2:
The patent applies partial action by selecting only the necessary eye parameters (position, size, distance) required for accurate focusing, rather than analyzing all possible eye characteristics. This selective approach maintains focusing accuracy while minimizing processing time by avoiding unnecessary detection and calculation steps.
Data Source
AI summary
An imaging device of the present invention comprises a facial organ detection section for detecting facial organs within a face from an image that has been taken by an imaging section, an eye selection section for comparing sizes of eyes and selecting the largest eye, an eye size comparative determination section for comparatively determining the size of the eye with respect to a specified value, and an AF region setting section for setting an AF region in accordance with size and position of the selected eye when it has been determined that the size of the eye is larger than a specified value, and setting an AF region in accordance with position and size of a face to which the selected eye belongs when it has been determined that the size of the eye is smaller than a specified value.