Subject Extraction Device with Dynamic Calculation Resource Assignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing subject extraction technologies face challenges in efficiently processing high-definition images in real-time, particularly when dealing with multiple subjects or complex shapes, leading to increased processing time and resource demands.
Innovation Solution
A subject extraction device that includes a resolution reducing unit, a subject possibility extraction unit, a resolution increasing unit, and a calculation resource assignment unit, which reduces image resolution, extracts subject possibilities, increases resolution of boundary portions, and dynamically assigns calculation resources based on the number and complexity of subjects to optimize processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subject extraction is performed on high-definition images in real-time, then extraction accuracy is improved, but processing time exceeds the limited time available
Solution Approach 1:
The image processing is segmented into two stages: first processing the entire image at low resolution to identify candidate regions, then processing only the boundary portions of these candidates at high resolution. This segmentation allows the system to maintain high extraction accuracy while reducing overall processing time by avoiding full high-resolution processing of the entire image.
Solution Approach 2:
Different resolution qualities are applied to different regions of the image. The low-resolution processing is applied to the entire image for initial subject candidate identification, while high-resolution processing is applied locally only to the boundary portions of these candidates where precise classification is most critical. This local quality approach optimizes the balance between accuracy and processing speed.
2Measurement precision
If calculation resources are increased for subject possibility extraction, then extraction accuracy is improved, but resource consumption exceeds the upper limit
Solution Approach 1:
Instead of applying high computational resources to the entire image, the system performs low-resolution processing on the full image and reserves high computational resources only for processing the boundary portions of subject candidates. This partial application of excessive computation (high-resolution processing) only where necessary maintains extraction accuracy while keeping total resource consumption within the upper limit.
3Reliability
If the unclassified region becomes large due to multiple subjects or complex shapes, then subject extraction completeness is improved, but processing amount increases beyond capacity
Solution Approach 1:
The system extracts and isolates the boundary portions of subject candidates from the rest of the image for high-resolution processing. By taking out only these critical boundary regions where classification uncertainty exists, the system can handle large unclassified regions resulting from multiple or complex subjects without proportionally increasing the overall processing amount, thus maintaining both completeness and efficiency.
Data Source
AI summary
A subject extraction device (10) according to the present disclosure includes: a resolution reducing unit (111) that reduces resolution of an image to generate a low resolution image; a subject possibility extraction unit (112) that extracts possibilities of a subject portion from the low resolution image; a resolution increasing unit (113) that increases resolution of a boundary portion between the subject portion and a non-subject portion, judges whether the boundary portion is the non-subject portion pixel by pixel, and decides the subject portion and the non-subject portion on the basis of a judgement result; and a calculation resource assignment unit (114) that determines a value of the calculation resource to be assigned to the subject possibility extraction unit (112) as a first value and a value of the calculation resource to be assigned to the resolution increasing unit (113) as a second value such that the values are equal to or less than an upper limit value of the calculation resource that can be used by the subject possibility extraction unit (112) and the resolution increasing unit (113), and assigns the calculation resource to the subject possibility extraction unit (112) and the resolution increasing unit (113) on the basis of the first value and the second value that have been determined.