Camera Autofocus Using Face Detection and Open-Loop Lens Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autofocus methods in cameras, especially the contrast detection mode, are time-consuming and have varying accuracy due to the need for manual lens movement and reliance on phase difference or contrast filters, which are sensitive to design and lighting conditions.
Innovation Solution
An autofocus method and controller that utilize an open-loop actuator to adjust the lens position based on face detection, with a lookup table to determine initial lens positions and adaptive contrast filter coefficients for improved focus determination, reducing the search range and enhancing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If contrast detection mode is used to achieve accurate focus, then measurement precision is improved, but loss of time increases due to the need to move the lens until maximum contrast is detected
Solution Approach 1:
The patent applies preliminary action by using face detection results to pre-determine the lens start position before contrast detection begins. The controller calculates the expected lens position based on face size and characteristics, then uses this as the starting point for contrast-based focus adjustment. This preliminary positioning significantly reduces the search range and time required to achieve accurate focus while maintaining measurement precision.
2Measurement precision
If a high-priced range sensor is used to achieve high autofocus performance, then measurement precision is improved, but manufacturing cost increases
Solution Approach 1:
The patent replaces the mechanical/optical range sensor system with an image-processing-based focus determination system. Instead of using expensive active sensors (ultrasound or infrared transceivers), the system uses the image sensor to capture images and processes these images through contrast detection and face detection algorithms to determine focus status and lens position, achieving high autofocus performance without additional expensive hardware.
Solution Approach 2:
The patent makes the image sensor serve multiple functions: it is used for both capturing the image content (face detection) and for focus determination (contrast detection). This multi-functionality eliminates the need for a separate range sensor, reducing manufacturing cost while maintaining autofocus performance through intelligent processing of the captured image data.
3Device complexity
If contrast filter coefficients are fixed, then device complexity is reduced, but measurement precision deteriorates due to varying lighting conditions and face characteristics
Solution Approach 1:
The patent applies dynamics by making the contrast filter coefficients adaptive rather than fixed. The controller dynamically adjusts the coefficients of the contrast filter based on face detection results, including face size, position, and characteristics. This dynamic adaptation allows the system to optimize focus detection accuracy for different lighting conditions and face characteristics while maintaining manageable device complexity through algorithmic control.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach speeds up autofocus by reducing the lens movement range and increases accuracy by using face-specific contrast filters, especially in portrait imaging, by determining focus based on peak contrast values.
Implementation Method 1
a lens and an electronic controller. The electronic controller determines whether imagery of a face exists in a first image created by light received through the lens
Implementation Method 2
calculating a contrast value of the next image using a contrast filter, and determining whether the lens is in focus based on the contrast value
Data Source
AI summary
An autofocus method of a camera, having an actuator controlling the position of a lens in an open-loop mode, includes receiving and analyzing an image captured using the lens and generating a face detection signal indicating whether a face is detected based on the analysis result. The size of the face is calculated when the face is detected, and a start position of the lens is determined based on the size of the face. A focus position of the lens is sought based on the start position of the lens.