Automated Subject Data Extraction for Machine Learning Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information processing systems for machine learning require manual selection of subject images from captured images, which is inefficient and labor-intensive, especially when dealing with large datasets or dynamic imaging scenarios.
Innovation Solution
An information processing apparatus equipped with a processor that automatically identifies and extracts specific subject data from images captured by an image sensor, using features like focus operations and similarity evaluation values to distinguish and isolate subject regions, thereby facilitating the collection of training data for machine learning algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual selection of subject images is used, then accuracy of subject identification can be maintained, but labor intensity and time consumption increase significantly
Solution Approach 1:
The imaging apparatus performs self-service by automatically identifying and extracting subject images using focus operation information and similarity evaluation values, eliminating the need for manual selection while maintaining identification accuracy through automated processing
Solution Approach 2:
The patent replaces manual mechanical selection with automated computational processing, using algorithms that evaluate image similarity and focus characteristics to automatically identify subject images, thereby reducing time consumption while preserving accuracy
2Loss of time
If automated extraction of subject images is implemented, then time consumption is reduced, but complexity of the processing system increases
Solution Approach 1:
The system performs preliminary actions by capturing focus operation information and pre-calculating similarity evaluation values during the imaging process, so that subject image extraction can be performed efficiently without requiring complex real-time processing, thus reducing overall system complexity
Solution Approach 2:
The patent segments the complex processing into distinct functional components: focus operation information capture, similarity evaluation value calculation, and subject image extraction, making each component simpler and more manageable while achieving automated processing
3Measurement precision
If focus operation information is used for subject identification, then accuracy in dynamic scenarios improves, but requirements for imaging conditions increase
Solution Approach 1:
The system adapts to varying imaging conditions by dynamically adjusting the evaluation parameters based on focus operation information, allowing accurate subject identification across different scenarios without requiring fixed, restrictive imaging conditions, thus improving versatility
Data Source
AI summary
An information processing apparatus includes a processor and a memory connected to or built in the processor. In a case in which imaging accompanied by a focus operation in which a specific subject is used as a focus target region is performed by an image sensor, the processor outputs specific subject data related to a specific subject image indicating the specific subject in a captured image obtained by the imaging as data used in machine learning.