Human-Body Exposure Weighting for Backlit Moving Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capturing systems struggle to appropriately expose human bodies, especially in complex scenes like backlight conditions or low-light environments, leading to difficulties in accurately detecting and analyzing multiple human faces.
Innovation Solution
An information processing apparatus that detects human bodies in an image, discriminates their parts and orientations, and calculates exposure amounts based on luminance measurements and weightings for each photometric area, ensuring proper exposure and enhanced visibility of human faces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single exposure amount is used for the entire image, then the overall image can be captured, but human faces in specific areas (such as backlight or low-light conditions) cannot be properly exposed
Solution Approach 1:
The patent divides the image into multiple photometric areas corresponding to different human bodies detected in the scene. Each photometric area is processed independently to calculate appropriate exposure amounts, allowing different regions to have optimized exposure settings tailored to their specific lighting conditions and content requirements.
Solution Approach 2:
The patent applies different weighting values to different photometric areas based on the detected human bodies' characteristics. By assigning higher weights to areas where human faces are detected and lower weights to other areas, the system achieves local optimization of exposure quality without requiring complete redesign of the entire image capture process.
2Measurement precision
If the exposure amount is adjusted to properly expose human faces, then face visibility improves, but the overall image balance may be compromised
Solution Approach 1:
The patent dynamically calculates exposure amounts for each photometric area based on real-time detection results and weighting values. The exposure adjustment is applied selectively only to areas where human faces are detected, while maintaining original exposure settings for other areas, thus achieving dynamic adaptation without compromising overall image balance.
Solution Approach 2:
The patent changes the exposure parameter locally by calculating different exposure amounts for different photometric areas. The weighting value assigned to each photometric area determines the degree of exposure adjustment, allowing precise control over which areas receive exposure optimization and by how much, thereby preventing information loss in any single region.
3Productivity
If multiple human bodies are detected and weighted, then exposure optimization for specific areas improves, but the complexity of detection and discrimination increases
Solution Approach 1:
The patent performs human body detection and photometric area identification as preliminary steps before exposure calculation. By pre-segmenting the image into distinct photometric areas and pre-determining the weighting values based on detected human characteristics, the system simplifies the subsequent exposure calculation process and improves overall efficiency.
Solution Approach 2:
The patent creates a simplified representation model where each detected human body is represented as a photometric area with assigned weighting values. This model allows the complex task of optimizing exposure for multiple human bodies to be reduced to a systematic calculation based on pre-determined weights and luminance measurements, thereby reducing computational complexity.
Data Source
AI summary
An information processing apparatus includes a detection unit configured to detect a human body in an image, a discrimination unit configured to discriminate a part and an orientation of a detected human body in a case where a plurality of human bodies is detected by the detection unit, a photometry unit configured to, in a case where a plurality of human bodies is detected by the detection unit, measure a luminance of each of a plurality of photometric areas corresponding to the human bodies, a determination unit configured to determine a weighting for each photometric area in accordance with the part and orientation discriminated by the discrimination unit, and a calculation unit configured to calculate an exposure amount for capturing a moving image, from the luminance of each photometric area measured by the photometry unit and the weighting for each photometric area determined by the determination unit.