Image Recognition Apparatus Dynamic Exposure Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image recognition systems face challenges in achieving high accuracy for object recognition and speed estimation due to limitations in capturing conditions, leading to suboptimal performance in autonomous systems like autonomous driving vehicles and industrial robots.
Innovation Solution
An image recognition apparatus that utilizes multiple-exposure image capturing and adjusts capturing conditions based on recognition accuracy and error thresholds, employing a processing circuit to calculate recognition accuracy and a control circuit to modify exposure parameters such as exposure period, interval, sensitivity, and resolution to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple-exposure image capturing is used to improve recognition accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system calculates recognition accuracy from captured images and uses this accuracy information as feedback to dynamically adjust capturing conditions. When accuracy is insufficient, the system modifies exposure parameters (exposure period, interval, sensitivity, resolution) to improve subsequent recognition accuracy, creating a closed-loop control system that adapts to varying recognition requirements.
Solution Approach 2:
The patent implements dynamic adjustment of capturing conditions based on real-time recognition accuracy assessment. Instead of using fixed capturing parameters, the system continuously adapts exposure period, exposure interval, sensitivity, and resolution according to the calculated recognition accuracy, making the capturing process flexible and responsive to actual performance needs.
2Measurement precision
If capturing conditions are dynamically adjusted to improve recognition accuracy, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system adjusts capturing conditions selectively based on whether recognition accuracy meets a predetermined threshold. When accuracy is sufficient, no adjustment is made, avoiding unnecessary time loss. Only when accuracy falls below the threshold does the system modify capturing conditions, applying partial action only when needed rather than continuously adjusting parameters.
3Measurement precision
If multiple-exposure image capturing with adjusted parameters is used, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The system modifies specific capturing parameters (exposure period, exposure interval, sensitivity, resolution) based on recognition accuracy requirements. By changing these parameters dynamically rather than using fixed multiple-exposure settings, the system optimizes the balance between measurement precision and processing efficiency, adjusting the degree of multiple-exposure capturing according to actual performance needs.
Data Source
AI summary
An image recognition apparatus includes a processing circuit and a control circuit. The processing circuit receives a first image obtained by performing multiple-exposure image capturing of a first subject. The processing circuit uses the first image to calculate a recognition accuracy of the first subject. The control circuit changes a condition of the multiple-exposure image capturing in a case where the recognition accuracy is lower than a recognition accuracy threshold.