Weld Seam Type Recognition System Based on Structured Light Vision and Ensemble Learning

Abstract

In this paper, we propose a weld seam recognition system based on structured light vision and ensemble learning. The proposed system consists of an industrial robot, a structured light vision sensor and a computer. The recognition procedures of proposed system include weld seam feature extraction and weld seam classification. In feature extraction part, the input images are processed by the following steps: noise filtration, laser stripe pattern extraction, main line extraction, edge points detection and feature computation. In classification part, ensemble learning models including BP-Adaboost and KNN-Adaboost are established to classify the images by the feature extracted. The experiment results validate the effectiveness and robustness of the proposed recognition system.