Foreground Extraction Using Segmented Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing foreground extraction methods, such as background differencing and machine learning-based techniques, fail to accurately extract motionless objects as foreground areas, leading to increased processing loads and inefficiencies, particularly in scenarios like soccer games where both moving and stationary objects are present.
Innovation Solution
An image processing apparatus that differentiates between two areas in an inputted image, using a machine learning-based method for one area and a background differencing technique for the other, allowing for the extraction of both moving and motionless objects as foreground areas while minimizing processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If background differencing technique is used to extract foreground areas, then processing load is reduced, but motionless objects cannot be extracted as foreground areas
Solution Approach 1:
The image is divided into multiple regions, and different extraction methods are applied to different regions. The background differencing technique is applied to regions containing moving objects, while machine learning-based method is applied to regions containing motionless objects, allowing both types of objects to be extracted effectively while maintaining low processing load
Solution Approach 2:
Different extraction methods are selectively applied to different regions of the image based on the characteristics of objects present in each region. This allows the system to use the most appropriate method for each local area, optimizing both processing efficiency and extraction accuracy
2Measurement precision
If machine learning-based foreground extraction method is applied to the entire image, then foreground extraction accuracy is improved, but processing load increases significantly
Solution Approach 1:
The image is segmented into multiple regions, and the machine learning-based extraction method is applied only to regions containing motionless objects, while the background differencing technique is used for regions with moving objects. This selective application significantly reduces the overall processing load while maintaining high extraction accuracy where needed
Solution Approach 2:
Instead of applying the machine learning-based method to the entire image (excessive action), the system applies it only to the necessary regions containing motionless objects (partial action), thereby reducing processing load while maintaining sufficient accuracy
Data Source
AI summary
In order to extract a foreground area more appropriately, an image processing apparatus for extracting a foreground area from an inputted image includes an image input unit that sets a first area and a second area different from the first area in the inputted image, a first extraction unit that extracts a foreground area from the first area, and a second extraction unit that extracts a foreground area from the second area by using an extraction method different from an extraction method used by the first extraction unit.