Image Selection for Moving Object Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Object recognition processing in systems involving moving objects is inefficient, leading to bottlenecks and reduced convenience, as it requires objects to be stationary during recognition, slowing down system operations and increasing complexity.
Innovation Solution
An information processing apparatus with a detection unit to identify objects in captured images and a determination unit to select the most suitable images for feature extraction, reducing the number of images needed for recognition and enhancing processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If object recognition processing is performed on multiple captured images to ensure accurate identification, then recognition precision is improved, but processing time increases and system productivity deteriorates
Solution Approach 1:
The patent extracts only the most suitable captured image for feature extraction based on predetermined evaluation criteria (image quality, object position, lighting conditions). By selecting one optimal image from multiple captured images instead of processing all of them, the system maintains high recognition precision while significantly reducing processing time and computational resources.
Solution Approach 2:
The patent changes the parameter of image selection from processing all captured images to processing only the most suitable image based on evaluated parameters such as image quality, object position, and lighting conditions. This parameter-based selection approach optimizes the balance between recognition precision and processing efficiency.
2Reliability
If feature extraction is performed on all captured images to ensure comprehensive object analysis, then recognition reliability is improved, but computer resource consumption increases
Solution Approach 1:
The patent extracts and processes only the most suitable captured image for feature extraction based on predetermined evaluation criteria. By eliminating the need to perform feature extraction on all captured images and processing only one optimal image, the system maintains high recognition reliability while significantly reducing computer resource consumption including CPU usage, memory requirements, and processing time.
3Measurement precision
If multiple images are processed for moving objects to maintain recognition accuracy, then measurement precision is improved, but the complexity of controlling object movement increases
Solution Approach 1:
The patent enables the system to automatically evaluate and select the most suitable captured image for moving objects based on predetermined criteria without requiring external control or intervention. The system autonomously handles the selection process by evaluating image quality, object position, and other parameters, thereby maintaining recognition accuracy while reducing operational complexity.
Data Source
AI summary
An information processing apparatus (2000) detects an object (30) from each of a plurality of captured images (20) generated by an image capture apparatus (10). The information processing apparatus (2000) further determines, for each detected object (30), an appropriate image being a captured image (20) suitable for feature extraction of the object (30) from among a plurality of captured images (20) in which the object (30) is detected.