Neural Image Segmentation for Precise Material Joining Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer vision methods for automated welding struggle with accurately identifying joining points due to issues like changes in illumination, inhomogeneous backgrounds, and blurriness, which can lead to inaccuracies in material joining.
Innovation Solution
A method and system that utilize a trained neural network to process digital images of materials, converting them into tensors and generating segmentation masks to determine precise joining points, allowing for improved segmentation and accurate material joining through techniques like welding, brazing, or soldering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional computer vision methods are used for seam tracking and quality inspection, then automation is achieved, but measurement precision deteriorates due to illumination changes, inhomogeneous backgrounds, and blurriness
Solution Approach 1:
The patent transforms the input image into a tensor representation and applies neural network processing to change the parameter space, converting raw pixel data into segmented regions with clear boundaries. This parameter transformation enables accurate joining point identification even under varying illumination and background conditions.
Solution Approach 2:
The patent introduces segmentation masks as an intermediary representation between the raw image and the joining point detection. These masks serve as a mediator that isolates relevant features and eliminates the impact of harmful factors like illumination changes and background inhomogeneity.
2Measurement precision
If advanced neural network segmentation is implemented, then joining point identification accuracy is improved, but device complexity increases
Solution Approach 1:
The patent divides the image processing task into distinct segmentation steps using neural networks, where the image is split into meaningful regions based on material boundaries. This segmentation approach simplifies the subsequent joining point detection by focusing only on relevant regions rather than processing the entire image.
Solution Approach 2:
The patent replaces traditional mechanical image processing methods with neural network-based computational approaches. This substitution enables automated adaptation to varying conditions without requiring manual tuning of processing parameters, reducing operational complexity despite the advanced algorithms used.
3Manufacturing precision
If precise segmentation and joining point detection are achieved, then manufacturing precision is improved, but processing time increases
Solution Approach 1:
The patent applies segmentation only to the necessary regions of the image rather than processing the entire image in detail. By focusing computational resources on areas containing potential joining points and material boundaries, the system achieves high precision while reducing overall processing time.
Solution Approach 2:
The patent performs preliminary segmentation of the image into distinct material regions before conducting detailed joining point analysis. This preliminary action organizes the data structure and identifies regions of interest, enabling faster and more efficient subsequent processing steps.
Data Source
AI summary
A method for joining materials, comprising: providing two materials; placing a first portion of a first material adjacent to a second portion of a second material; taking a digital image of the first and second portions by an imaging sensor; converting the digital image into a tensor, the tensor comprising first, second, and third dimensions, wherein the first dimension comprises a height of the digital image, the second dimension comprises a width of the digital image, and the third dimension comprises a number of digital channels of the imaging sensor, entering the tensor into a trained neural network (NN); outputting a segmentation mask by the NN, determining a joining point using the segmentation mask; and joining the first and second material at the joining point.