Object Vertex Recognition Model for Accurate Edge Detection
Solution Overview
Problem
Existing digital image processing methods struggle to efficiently extract object edges from images, leading to inaccurate object recognition and processing due to redundant environmental information.
Innovation Solution
A method and system utilizing an object vertex recognition model, which employs machine learning to detect object vertices and their relative positions within an image, allowing for accurate determination of object edge lines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional edge detection methods are used, then the processing can be performed, but the edge detection accuracy is low due to redundant environmental information
Solution Approach 1:
The patent extracts only the essential object vertex information from the complete image data. By using an object vertex recognition model to identify and extract vertex positions and relative positions, the system separates the useful object information from the redundant environmental information, enabling accurate edge detection without being distracted by unnecessary background data.
Solution Approach 2:
The patent segments the image processing task into distinct stages: first recognizing object vertices through a specialized model, then determining relative positions, and finally calculating actual positions and generating edge lines. This segmentation allows each stage to focus on specific aspects, improving overall accuracy while managing computational efficiency.
2Measurement precision
If the entire image is processed to locate object edges, then complete edge information can be obtained, but the processing time increases due to large amount of data
Solution Approach 1:
The patent extracts only the essential object vertex information from the complete image data. By using an object vertex recognition model to identify and extract vertex positions and relative positions, the system separates the useful object information from the redundant environmental information, enabling accurate edge detection without being distracted by unnecessary background data.
Solution Approach 2:
The patent performs preliminary recognition of object vertices and their relative positions before final edge line generation. This preliminary action filters and pre-processes the data, identifying only the relevant vertex information needed for edge detection, thereby reducing the amount of data that requires full processing and decreasing overall processing time.
3Measurement precision
If machine learning model is used for vertex recognition, then the vertex detection accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent segments the image processing task into distinct stages: first recognizing object vertices through a specialized model, then determining relative positions, and finally calculating actual positions and generating edge lines. This segmentation allows each stage to focus on specific aspects, improving overall accuracy while managing computational efficiency.
Solution Approach 2:
The patent applies the object vertex recognition model specifically to detect object vertices and their relative positions, rather than processing the entire image with high computational complexity. By localizing the machine learning application to only the essential vertex recognition task, the system achieves high accuracy for critical parameters while reducing overall computational burden.
Data Source
AI summary
A method and system of recognizing and processing object edges and a computer-readable storage medium are provided. The method includes: obtaining an input image, where the input image includes an object with edges, and the edges of the object include a plurality of object vertices; recognizing, through an object vertex recognition model, the input image and obtaining a relative position of each object vertex and a corresponding image vertex thereof; determining an actual position of each object vertex in the input image according to the relative position of each object vertex and the corresponding image vertex thereof; and sequentially connecting adjacent object vertices to form edge lines to obtain the edges of the object with edges in the input image according to the actual position of each object vertex in the input image.