Object Detection Using Color Segmentation Bounding Polygons
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for localizing objects within multi-dimensional image data, such as RGB-D data, are imprecise due to the combination of color and depth information, which hinders accurate object identification and characterization.
Innovation Solution
Performing color segmentation to define bounding polygons that minimize free space, allowing for cropping of image data, followed by image processing techniques like thresholding, edge detection, and binary inversion to enhance object localization and identification within the cropped image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If color segmentation is performed to define bounding polygons, then object localization precision is improved, but processing complexity increases
Solution Approach 1:
The patent applies color segmentation to divide the image into distinct color regions, creating bounding polygons that isolate objects based on their color characteristics. This segmentation process separates objects from the background and from each other, enabling precise localization while managing complexity through color-based differentiation
Solution Approach 2:
The patent extracts color information from the image data and uses it to define bounding polygons. By taking out color characteristics and using them as the basis for object isolation, the system achieves precise localization without requiring complex multi-dimensional analysis of the entire image space
2Adaptability or versatility
If multi-dimensional image data is used for object identification, then object characterization capability is improved, but localization accuracy deteriorates
Solution Approach 1:
The patent segments the multi-dimensional image data by color, creating simplified bounding polygons that isolate objects. This segmentation reduces the complexity of multi-dimensional data while maintaining localization accuracy, as each color segment corresponds to a distinct object or region
Solution Approach 2:
The patent extracts color information from multi-dimensional image data and uses it to define bounding polygons. By taking out color characteristics, the system achieves accurate localization while still maintaining the ability to characterize objects through their color properties and spatial relationships
3Measurement precision
If bounding polygons are used to crop image data, then object isolation is improved, but data volume increases
Solution Approach 1:
The patent applies local quality by creating bounding polygons that precisely enclose objects based on their color characteristics. Each bounding polygon is optimized for its specific object, isolating it from the background while maintaining only the necessary data volume for accurate representation
Solution Approach 2:
The patent uses partial action by creating bounding polygons that encompass objects with minimal necessary space. The polygons are designed to be just large enough to contain the objects of interest, avoiding excessive data inclusion while ensuring complete object isolation
Data Source
AI summary
Image color data for a field of view is received. Thereafter, color segmentation can be performed on the image color data to define at least one bounding polygon that minimizes an amount of free space within each bounding polygon. The at least one bounding polygon is then used to crop the image color data to result in cropped image color data. Image processing can then be applied to the cropped image color data to identify at least one object therein. Related apparatus, systems, techniques and articles are also described.