Image Augmentation via Graphical Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies require complete replacement of older images with new ones for updates, even if the new images have lower quality or resolution, and often necessitate purchasing from third parties, failing to provide up-to-date representations of landscapes and features in maps or scenes.
Innovation Solution
A system and method that analyze and amend images by extracting graphical features from a first image and applying them to a second image, using machine learning algorithms to detect differences and generate a third set of features for stylizing the first image, allowing for updating of object features without rebuilding the entire image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a new image is purchased from third parties to update the image, then the image currency is improved, but the image quality and resolution deteriorate
Solution Approach 1:
The patent segments the image update process by separating the base image (high quality, older) from the update layer (new features extracted from new image). Only the differential features are extracted and applied, rather than replacing the entire image. This allows maintaining the high quality of the base image while incorporating current features.
Solution Approach 2:
The patent merges the high-quality base image with the extracted graphical features from the new image. The machine learning algorithm combines the structural quality of the older image with the updated feature information, creating an augmented image that has both current features and high resolution.
2Loss of time
If the old image is completely replaced with a new image to reflect current conditions, then the image currency is improved, but the cost increases due to purchasing from third parties
Solution Approach 1:
Instead of purchasing and using the entire new image, the patent creates a copy of only the necessary update information by extracting graphical features from the new image. These extracted features are then applied to the existing base image, avoiding the need to purchase the complete new image from third parties.
Solution Approach 2:
The patent extracts only the essential graphical features (such as building contours, road layouts, and landmark positions) from the new image using machine learning algorithms. This extraction process isolates the update information needed for currency improvement while discarding the redundant high-resolution data that would incur additional costs.
3Measurement precision
If machine learning algorithms are trained with human-provided examples, then the object detection accuracy is improved, but the time and cost for training increases
Solution Approach 1:
The patent performs preliminary action by pre-training the machine learning algorithms with a comprehensive set of human-provided examples during the system setup phase. This initial training creates a robust model that can automatically extract features with high accuracy without requiring continuous human intervention for each update scenario.
Solution Approach 2:
Once trained, the machine learning algorithm becomes self-sufficient in extracting graphical features from new images without requiring ongoing human moderation or example provision. The algorithm autonomously performs feature extraction, comparison, and augmentation, eliminating the need for continuous human time investment.
Data Source
AI summary
A system and method is provided for augmenting an image with stylized features. An exemplary method includes identifying, in a first image, a first version of an object having a first set of graphical features and identifying, in a second image, a second version of the object having a second set of graphical features. Moreover, the method includes extracting the first and second sets of graphical features from the first and second images, respectively, and generating a third set of graphical features by calculating differences between the first and second sets of graphical features. Finally, using the third set of graphical features, the method includes augmenting the first version of the object in the first image.