Depth-Based Exposure Computation for High Dynamic Range Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current digital camera exposure algorithms often yield unacceptable results due to the influence of background light levels on foreground exposure, leading to suboptimal photographs with unbalanced illumination.

Innovation Solution

Implementing a 3D camera system that segments a scene based on depth, allowing for accurate determination of the foreground subject of interest and adjusting exposure parameters like aperture, shutter speed, and illumination to match discrete compositional zones, thereby achieving correct illumination of both foreground and background.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a single exposure level is used for the entire scene, then the camera operation is simple and fast, but the foreground and background cannot both be correctly exposed when they have different light levels

Engineering Contradiction:
Improveexposure accuracyVSAvoidexposure control complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The scene is segmented into multiple depth-based zones using a depth map, where each zone corresponds to a different exposure level. This allows the camera to apply different exposure parameters to different spatial regions, enabling correct exposure of both foreground and background subjects simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different exposure parameters (shutter speed, aperture, ISO) are applied locally to different depth zones rather than uniformly across the entire scene. This ensures that each region receives the optimal exposure settings appropriate for its specific lighting conditions and distance from the camera.

Inventive Principle:
Principle #3Local quality

2Reliability

If multiple images at different exposures are captured and combined, then high dynamic range and natural illumination are achieved, but the processing time and computational complexity increase

Engineering Contradiction:
Improveimage qualityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

A depth map of the scene is generated in advance before exposure computation, allowing the system to pre-determine the optimal exposure parameters for each depth zone. This preliminary depth-based segmentation enables efficient exposure calculation without requiring iterative processing of multiple full images.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Instead of capturing and processing multiple full-resolution images at different exposures, the system uses a depth map (a simplified representation) to guide exposure parameter selection. This copying approach allows the system to work with lightweight depth information rather than heavy image data, reducing processing time.

Inventive Principle:
Principle #26Copying

3Ease of manufacture

If traditional zone-based metering is used to determine exposure, then the exposure calculation is simple, but background light levels incorrectly influence foreground exposure

Engineering Contradiction:
Improveexposure algorithm simplicityVSAvoidexposure measurement accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The exposure calculation transitions from two-dimensional zone-based metering in the image plane to three-dimensional depth-based zoning using a depth map. By incorporating the depth dimension, the system can distinguish between foreground and background regions and calculate exposure parameters that are independent of background lighting conditions.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3241155B1Exposure computation via depth-based computational photography
Publication Date: 2023.11.01 DELL PROD LP
  • EP3241155B1 patent drawingFigure 1
  • EP3241155B1 patent drawingFigure 2
  • EP3241155B1 patent drawingFigure 3

AI summary

A method and electronic information handling system provide recording a first image of a scene at a first exposure level using a three-dimensional (3D) camera, correlating distances from the 3D camera and exposure levels over a plurality of image elements of the first image, selecting an exposure parameter value for at least one of the plurality of image elements having a z-distance value falling within a range of z-distance values, recording a second image of the scene according to the exposure parameter value, and constructing a composite image based on at least a portion of the second image for the at least one of the plurality of image elements.