Camera Control for AI Object Detection Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image analysis systems using artificial intelligence struggle with detection and identification accuracy when the resolution and camera angle of teacher data images differ from those of the actual images taken, leading to poor object recognition.

Innovation Solution

A system and method that analyze the imaging conditions of teacher data, including resolution and camera angle, to control the camera to take images under matching conditions, thereby improving object detection and identification accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If the number of learning images is increased by processing reference images through imitation (defocusing or blurring), then the quantity of training data is improved, but the detection and identification accuracy deteriorates because the resolution and camera angle differ from actual images

Engineering Contradiction:

Improvenumber of learning imagesVSAvoiddetection and identification accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent changes the imaging parameters (resolution, camera angle, magnification) by analyzing teacher data to extract optimal parameters, then controlling the camera to capture images under these specific parameters. This ensures the captured images match the teacher data characteristics, resolving the accuracy issue while maintaining efficient training data generation

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

Instead of merely imitating reference images through software processing (which loses quality), the patent captures actual images under controlled conditions that replicate the teacher data characteristics. This physical copying approach preserves image quality and ensures parameter consistency

Inventive Principle:

Principle #26Copying

2Ease of operation

If images are taken without controlling imaging conditions to match teacher data, then the ease of operation is improved, but the object detection and identification accuracy deteriorates

Engineering Contradiction:

Improvecamera operation simplicityVSAvoidobject detection and identification accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system automatically analyzes teacher data to extract imaging parameters and self-adjusts the camera settings without requiring manual intervention. The camera control unit autonomously configures resolution, angle, and magnification based on the analyzed parameters, maintaining ease of operation while ensuring accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The imaging condition analysis unit continuously analyzes teacher data and provides feedback to the camera control unit, which adjusts the camera parameters accordingly. This closed-loop feedback mechanism ensures the captured images consistently match the desired characteristics without complex manual configuration

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10846326B2System and method for controlling camera and program

Publication Date: 2020.11.24 OPTIM

AI summary

The present invention is to improve the detection and the identification accuracy of an object in image analysis in a camera control system that controls a camera used to take an image to be analyzed by artificial intelligence. The image analysis system that performs machine learning by using a plurality of teacher data associating a label that indicates what the object is with image data to which the label is attached, includes an imaging condition analysis module 211 that analyzes the imaging condition of teacher data; and a camera control module 212 that controls a camera to take an image under the analyzed imaging condition, to analyze the image taken under the imaging condition.