Optical Information Training Apparatus for Camera Color Tone Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge is to generate two-dimensional images of the same color tone regardless of the differences in color tones between individual cameras when capturing a three-dimensional scene from arbitrary viewpoints.
Innovation Solution
An optical information training and generation apparatus is developed, which includes an optical information training and inference part, a sensor identifier input part, a spectral characteristic conversion part, and a spectral characteristic setting part. This apparatus sets a training spectral characteristic corresponding to the sensor identifier in the spectral characteristic conversion part, allowing it to output color values based on the optical information according to the training spectral characteristic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If two-dimensional images are acquired using different cameras, then the coverage of three-dimensional scene imaging is improved, but the color tone consistency deteriorates
Solution Approach 1:
The patent applies parameter changes by transforming images from different camera color spaces to a standardized reference color space. Each camera's color characteristics are converted using predetermined conversion parameters, allowing images from diverse cameras to be unified in color tone while maintaining the ability to use multiple camera types for broader scene coverage.
2Manufacturing precision
If individual models are trained for each camera, then the color tone precision is improved, but the system complexity increases
Solution Approach 1:
The patent implements universality by creating a single unified model that can process images from multiple camera types. Instead of maintaining separate models for each camera, the system uses one model with predetermined color conversion parameters that adapt to different camera characteristics, reducing model management complexity while maintaining precision.
Solution Approach 2:
The patent applies preliminary action by pre-calculating and storing color conversion parameters for different camera types before actual image processing. These predetermined parameters are prepared in advance and applied during image generation, eliminating the need for real-time calibration and reducing system complexity.
3Device complexity
If a unified model is used for multiple cameras, then the system complexity is reduced, but the color tone precision deteriorates
Solution Approach 1:
The patent resolves this contradiction by incorporating camera-specific color conversion parameters into the unified model. The model remains single and simple to manage, but it internally applies different parameter sets depending on the input camera type, thereby maintaining high color tone accuracy without increasing complexity.
Data Source
AI summary
An optical information training and generation apparatus includes an optical information training and inference part which includes an optical training and inference section for training for information on optics with a two-dimensional image of a three-dimensional scene acquired by a sensor as teacher data, a sensor identifier input part for receiving a sensor identifier to identify the sensor, a spectral characteristic conversion part, and a spectral characteristic setting part for setting a spectral characteristic in the spectral characteristic conversion part. The spectral characteristic setting part sets a training spectral characteristic corresponding to the sensor identifier in the spectral characteristic conversion part. The optical training and inference section outputs optical information to a latter part and the spectral characteristic conversion part outputs color values based on the optical information according to the training spectral characteristic.