Image Processing Apparatus Region Detection Using Previous Frame Seeds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional region detection methods for image processing are inefficient, particularly in high-resolution images, as they require numerous iterations and significant memory usage, leading to increased complexity and potential inaccuracies in real-time video processing.
Innovation Solution
An image processing apparatus and method that utilize the region detection result of a previous frame to perform accurate region detection in subsequent frames with fewer operations and reduced memory usage, by identifying adjacent pixel regions and performing region growing processing based on similarity thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional region detection method is used with multiple iterations, then region detection accuracy is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent applies preliminary action by using region detection results from previous frames as initialization for current frame detection. The seed region and detection results from frame n-1 are utilized to accelerate region detection in frame n, avoiding redundant iterations and reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent uses copying by replicating the region detection approach across frames. The detection results from one frame are copied and adapted for use in subsequent frames, where the seed region and detection mask from frame n-1 serve as starting points for frame n, reducing the need for repeated full iterations.
2Measurement precision
If conventional region detection method is used with multiple iterations, then region detection accuracy is improved, but processing speed decreases
Solution Approach 1:
The patent applies preliminary action by using region detection results from previous frames as initialization for current frame detection. The seed region and detection results from frame n-1 are utilized to accelerate region detection in frame n, avoiding redundant iterations and reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent ensures continuity of useful action by maintaining region detection results across frames. Instead of performing complete region detection from scratch for each frame, the system continuously builds upon previous detection results, making the processing efficient and suitable for real-time video applications.
3Measurement precision
If image size increases in high-resolution displays, then image quality is improved, but number of operations and memory usage increase rapidly
Solution Approach 1:
The patent applies preliminary action by using region detection results from previous frames as initialization for current frame detection. The seed region and detection results from frame n-1 are utilized to accelerate region detection in frame n, avoiding redundant iterations and reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent applies discarding and recovering by selectively retaining only the necessary region detection results from previous frames (seed region and detection mask) while discarding redundant information. This allows the system to maintain high-resolution image processing capability with reduced memory usage by keeping only the essential detection data needed for acceleration.
4Productivity
If number of iterations is limited for real-time processing, then processing speed is improved, but region detection accuracy deteriorates
Solution Approach 1:
The patent applies preliminary action by using region detection results from previous frames as initialization for current frame detection. The seed region and detection results from frame n-1 are utilized to accelerate region detection in frame n, avoiding redundant iterations and reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent uses copying by replicating the region detection approach across frames. The detection results from one frame are copied and adapted for use in subsequent frames, where the seed region and detection mask from frame n-1 serve as starting points for frame n, reducing the need for repeated full iterations.
Data Source
Figure 1~2b
Figure 2c~3b
Figure 4a
AI summary
An image processing apparatus includes a memory configured to store first region detection information corresponding to at least one object of a first frame, and a processor configured to perform region growing processing based on a seed region identified from a second frame and the first region detection information, obtain second region detection information corresponding to at least one object of the second frame, and perform image processing on the second frame based on the second region detection information.