Neural Network Exposure Adjustment for Digital Image Luminosity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Adjusting exposure attributes in digital images to achieve optimal brightness levels is subjective and varies among human observers, leading to inconsistencies in image aesthetics, especially when capturing images under different light conditions.
Innovation Solution
A computer system processes curated and modified digital images to learn exposure adjustments using a machine learning system, specifically a neural network, which extracts features from images and generates correction values to automatically adjust luminosity characteristics, ensuring well-exposed images are produced.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual exposure adjustment is used based on human observer preference, then subjective aesthetic appeal can be achieved, but consistency and reliability of exposure quality deteriorates due to variation between observers
Solution Approach 1:
The system uses machine learning models that automatically learn optimal exposure adjustments from training data consisting of curated images and their corresponding adjustment values. The model self-improves by processing training images, extracting features, and learning the relationship between image characteristics and optimal exposure corrections, eliminating the need for manual observer-based adjustment while maintaining consistent quality
Solution Approach 2:
The system transforms the subjective exposure adjustment problem into an objective parameter-based solution by using machine learning models that analyze image features (such as luminance histograms, color distributions, and scene characteristics) and predict optimal exposure correction values. This converts subjective aesthetic judgment into measurable parameter optimization
2Manufacturing precision
If traditional automatic exposure algorithms are used in digital cameras, then processing speed is maintained, but exposure accuracy and aesthetic quality deteriorates due to subjective variation
Solution Approach 1:
The system performs exposure adjustment in advance by training machine learning models on large datasets of curated images with known optimal exposure values. The trained model is then deployed in the camera system, allowing rapid inference-based exposure correction without real-time processing delays. The heavy computational work is done beforehand during the training phase
Solution Approach 2:
The patent replaces traditional mechanical/optical exposure control systems with an intelligent software-based machine learning system. The ML model analyzes image features and predicts optimal exposure adjustments, substituting complex algorithmic processing for traditional exposure metering and control mechanisms
3Manufacturing precision
If machine learning systems are trained with curated and modified images, then exposure adjustment accuracy is improved, but device complexity and computational resources increase
Solution Approach 1:
The system segments the exposure adjustment task into distinct components: image feature extraction, training data preparation (curated and modified images), model training, and inference. By dividing the complex problem into manageable modules, the system achieves high accuracy while organizing complexity into separate, well-defined functional blocks
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a computing system that receives images that each have a predefined exposure attribute. For each image, a first set of features of the image are extracted. The first set of features are associated with a label indicating no modification of the image is required. A luminosity characteristic of the image is adjusted to form an adjusted image. A second set of features of the adjusted image are extracted. A neural network is trained to adjust luminosity characteristics of images using the first set of features and the second set of features of the adjusted image. An exposure adjustment model adjusts luminosity characteristics of images based on correction values determined using the trained neural network.