Welding Image Segmentation for Multi-Phenomenon Behavior Measurement

The patent applies universality by creating a measurement system that can simultaneously measure multiple welding behaviors (spatter, fumes, molten pool shapes) using a single integrated image processing framework. The system divides the captured welding image into multiple regions of interest and processes them concurrently, allowing one system to perform multiple measurement functions that were previously required separate specialized methods.

The patent applies segmentation by dividing the captured welding image into multiple distinct regions of interest (spatter region, fume region, molten pool region) and processing each region separately through dedicated image processing steps. This segmentation approach allows the system to handle multiple measurement tasks in an organized manner, reducing the complexity that would otherwise arise from attempting to measure all behaviors simultaneously in a single undifferentiated process.

The patent applies merging by combining multiple measurement capabilities (spatter detection, fume detection, molten pool shape analysis) into a single integrated system that processes one captured image simultaneously. By merging these functions and processing them in parallel from the same image source, the system achieves comprehensive real-time quality determination without requiring multiple separate measurement systems, thereby maintaining cost-effectiveness while improving productivity.