Camera Subject Detection Using Multi-Space Binarization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image-capturing devices cannot accurately specify the position, size, or shape of a photographic subject based on the user-selected AF area, limiting their ability to focus on the intended subject effectively.
Innovation Solution
A method involving binarization of subject images using color and luminance information, followed by evaluation value calculations to determine the position, size, and shape of the subject, incorporating techniques such as hue classification, inertial moment analysis, and evaluation value-based filtering to identify and refine potential subject areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple binarized images and evaluation values are used to specify subject position, size, and shape, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The subject image is segmented into multiple binarized images based on different color spaces (RGB, YCbCr, HSV) and luminance information. Each binarized image highlights different aspects of the subject, allowing comprehensive analysis from multiple perspectives simultaneously, which improves measurement precision without requiring a single complex processing method
Solution Approach 2:
Multiple evaluation values are calculated by changing parameters such as area ratios, aspect ratios, and inertial moments of the binarized images. These parameter variations provide diverse metrics for subject specification, enabling accurate determination of position, size, and shape through comparative analysis of multiple parameters rather than relying on a single complex measurement
2Manufacturing precision
If binarization based on color difference information and luminance information is performed, then manufacturing precision is improved, but loss of information increases
Solution Approach 1:
The patent transforms color and luminance information from a single-dimensional representation into multiple binarized images across different color spaces (RGB, YCbCr, HSV) and luminance channels. This dimensional transformation preserves the essential characteristics of the subject by distributing information across multiple binary representations, preventing information loss while improving extraction accuracy
Solution Approach 2:
Multiple binarized images derived from different color spaces and luminance information are combined to form a composite subject representation. This composite approach integrates information from various sources (color differences, luminance variations) to create a more accurate and complete subject extraction, compensating for information loss in individual binarized images
Data Source
AI summary
A photographic subject determination method includes: a binarization step of creating a plurality of binarized images of a subject image, based upon color information or luminance information in the subject image; an evaluation value calculation step of, for each of the plurality of binarized images, calculating an evaluation value that is used for specifying at least one of a position, a size, and a shape of a photographic subject within the subject image; and a photographic subject specification step of specifying at least one of the position, the size, and the shape of a photographic subject within the subject image, based upon the evaluation value.