Image Processing Device for Efficient Best Shot Frame Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies fail to efficiently and accurately extract frame images corresponding to the best shots from moving images, particularly in scenarios with multiple persons or long shooting periods, leading to increased processing time and lower accuracy.
Innovation Solution
An image processing device and method that acquire moving and still images, extract characteristic information from the still images, adjust image analysis conditions based on relevance with the moving images, and selectively analyze and output frame images with evaluation values above a threshold, optimizing the extraction process for efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all persons in a moving image are tracked to extract frame images, then the extraction is comprehensive, but the processing time increases significantly
Solution Approach 1:
The patent segments the moving image into multiple regions based on still image analysis results. By dividing the image space into regions of interest and non-interest areas, the system only performs detailed tracking in relevant regions, significantly reducing processing time while maintaining extraction completeness for important subjects.
Solution Approach 2:
The patent applies different processing qualities to different regions of the moving image. High-quality detailed tracking is applied only to regions identified as containing persons of interest from still images, while other regions receive minimal or no processing. This local differentiation maintains reliability for important subjects while reducing overall processing time.
2Reliability
If tracking is applied to all persons in the moving image, then all subjects are captured, but the accuracy of extracting relevant frame images decreases
Solution Approach 1:
The patent segments persons into different categories based on their appearance in still images. By identifying and separating persons of interest from unrelated persons, the system can apply appropriate tracking strategies to each group, improving the accuracy of extracting relevant frame images while still capturing all subjects for comprehensive coverage.
Solution Approach 2:
The patent uses still images as an intermediary to guide the tracking process in the moving image. By analyzing still images first to identify persons of interest, the system creates a reference framework that improves the accuracy of subsequent frame image extraction, while still maintaining coverage of all subjects through comprehensive initial analysis.
3Reliability
If the shooting period is extended to capture more scenes, then more potential best shots are available, but the processing time increases
Solution Approach 1:
The patent performs preliminary analysis of still images before processing the entire moving image sequence. By pre-identifying regions and time periods of interest from still images, the system can limit detailed frame image extraction to only those relevant segments, maintaining scene coverage reliability while significantly reducing processing time for extended shooting periods.
4Measurement precision
If detailed image analysis is applied to all portions of the moving image, then extraction accuracy is maximized, but processing efficiency decreases
Solution Approach 1:
The patent applies detailed image analysis only to specific regions identified as containing persons of interest, while using simpler or no analysis for other regions. This local differentiation maintains high extraction accuracy for relevant frame images while significantly improving overall processing efficiency by avoiding unnecessary detailed analysis throughout the entire image.
Data Source
AI summary
In the image processing device, the method and the recording medium, the characteristic information extractor extracts the shooting time of the still image. The image analysis condition determiner sets the image analysis condition for the portion of the moving image which portion was shot within the shooting time range covering certain periods of time before and after the shooting time of the still image, to be rougher than that for others. The frame image analyzer extracts frame images from the moving image and perform image analysis on the extracted frame images in accordance with the image analysis condition. The frame image output section calculates the evaluation value of the frame image based on the result of the image analysis and output the frame image having the evaluation value not less than the threshold value.


