Imaging Device Exposure Control for Face Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional face recognition systems in digital cameras face challenges in maintaining stable recognition and tracking due to fluctuations in the depth of field and insufficient lighting, especially when faces move or are small, leading to degraded resolution and recognition accuracy.

Innovation Solution

An imaging device that adjusts exposure time and aperture based on the area and moving speed of the face region, extending exposure time for larger areas to increase light and depth of field, and shortening it for smaller areas to reduce motion blur, while also controlling aperture values to optimize light and noise levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the diaphragm is narrowed to increase depth of field for small faces, then the recognition range is extended, but the light amount becomes insufficient leading to degraded recognition accuracy

Engineering Contradiction:
Improverecognition rangeVSAvoidlight amount
Core Design Contradiction:
Adaptability or versatilityVSIllumination intensity

Solution Approach 1:

The patent dynamically adjusts the diaphragm aperture value based on the detected face size. For small faces, the diaphragm is narrowed to extend depth of field and recognition range. For large faces, the diaphragm is opened to increase light amount. This dynamic adaptation resolves the contradiction between recognition range and light amount.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the diaphragm aperture parameter according to face size conditions. By adjusting this physical parameter, the system optimizes both the recognition range (through depth of field control) and light amount (through aperture size), resolving the technical contradiction.

Inventive Principle:
Principle #35Parameter changes

2Illumination intensity

If the exposure time is extended to increase light amount for large faces, then the light amount is sufficient, but motion blur increases leading to degraded resolution

Engineering Contradiction:
Improvelight amountVSAvoidresolution
Core Design Contradiction:
Illumination intensityVSManufacturing precision

Solution Approach 1:

The patent dynamically adjusts exposure time based on face size. For large faces where motion blur is less critical, exposure time is extended to increase light amount. For small faces where resolution is critical, exposure time is kept short to minimize motion blur. This dynamic adjustment resolves the contradiction between light amount and resolution.

Inventive Principle:
Principle #15Dynamics

3Manufacturing precision

If the focusing position is adjusted based on immediately preceding image data, then the focusing is optimized for previous frame, but the face escapes out of depth of field when face moves

Engineering Contradiction:
Improvefocusing accuracyVSAvoidtracking stability
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent performs preliminary action by detecting face size in advance and pre-adjusting the diaphragm aperture value before capturing the image. This ensures that when the face is captured, the depth of field is already optimized for that face size, preventing the face from escaping the depth of field during movement.

Inventive Principle:
Principle #10Preliminary action

4Illumination intensity

If the lighting function is activated to improve light amount, then the illumination is sufficient, but power consumption increases and flash usage may be prohibited

Engineering Contradiction:
Improvelight amountVSAvoidpower consumption
Core Design Contradiction:
Illumination intensityVSUse of energy by moving object

Solution Approach 1:

The patent replaces the mechanical/electrical lighting system (flash) with an optical control mechanism (diaphragm adjustment). By adjusting the diaphragm aperture, the system controls light amount optically without consuming additional power for active illumination, thus avoiding the drawbacks of flash usage.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8860875B2Imaging device for recognition and method of controlling the same
Publication Date: 2014.10.14 PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA
  • US8860875B2 patent drawing
  • US8860875B2 patent drawing
  • US8860875B2 patent drawing

AI summary

An imaging device for recognition (100) includes an image pickup unit (101) that sequentially generates a plurality of images by shooting an object, a face recognition unit (105) that recognizes, in each of the plurality of images, an object region in which the object is located in the image, a face tracking unit (106) that tracks the object region across the plurality of images, an area calculation unit (109) that calculates an area of the object region, and an exposure control unit (102) that sets a first exposure time for the image pickup unit (101) in the case where the area is smaller than a first threshold, and sets a second exposure time longer than the first exposure time in the case where the area is equal to or larger than the first threshold.