Image Processing Apparatus for Selective Background Averaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
When performing average processing on time-series images, shadows of individuals often remain due to insufficient removal of movement areas, leading to incomplete erasure of persons from images.
Innovation Solution
An image processing technique that acquires multiple images taken at different times, selects regions with significant differences using a criterion, and performs average processing only on regions where the difference meets the specified criteria, effectively eliminating areas with substantial changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If average processing is performed on all regions of time-series images, then processing is simplified, but persons and shadows remain in the processed image
Solution Approach 1:
The image processing is divided into two distinct stages: first, a difference image is generated by comparing adjacent time-series images to identify regions with significant changes (where persons are located); second, average processing is applied only to regions where the difference falls within a predetermined range (background regions). This segmentation allows precise person removal while maintaining processing efficiency.
Solution Approach 2:
Different processing methods are applied to different regions of the image based on local characteristics. Regions with large differences (containing persons) are excluded from average processing, while regions with small differences (background) undergo average processing. This local differentiation ensures that persons are completely removed while background regions maintain their quality.
2Manufacturing precision
If difference threshold is set low to remove all person regions, then person removal is thorough, but background regions with noise are also affected
Solution Approach 1:
The processing uses a feedback mechanism where the difference image serves as a guide for subsequent average processing. By calculating the difference between adjacent frames and using this information to determine which regions should be excluded from averaging, the system dynamically adjusts processing based on actual image content, ensuring persons are removed while preserving background integrity.
Solution Approach 2:
The method introduces a difference threshold parameter that separates person regions from background regions. By changing the processing parameter (applying average processing only when difference is within the threshold), the system achieves both thorough person removal and background preservation without affecting regions with significant changes.
3Measurement precision
If multiple images are compared to improve person detection accuracy, then detection precision increases, but processing time increases
Solution Approach 1:
The method performs preliminary difference calculation between adjacent time-series images before conducting average processing. This preliminary action identifies person regions in advance, allowing the subsequent average processing to be applied efficiently only to relevant background regions, thereby reducing overall processing time while maintaining high detection accuracy.
Data Source
AI summary
An image processing apparatus (100) includes: an acquisition unit (102) that acquires a plurality of images acquired by photographing a same location at different timing; a selection unit (104) that compares at least two of the plurality of images, and selects a target region being a region where a difference between the two images satisfies a criterion; and a processing unit (106) that performs average processing of averaging the target region included in each of the at least two images.