Hand Motion Analysis for Recognition of Qualified and Unqualified Welders using 9-DOF IMU Sensors and Support Vector Machine (SVM) Approach

This research aimed to find out how identify qualified and unqualified welders of shielded metal arc welding (SMAW) in the shipyard industry. A cost-effective system that can welder skills real time is needed reduce cost inspection maintain weldment quality. In this study, 9-degree freedom (DOF) sensors inertial measurement unit (IMU) were applied measure record typical hand motions welders. These consisted an accelerometer, a gyroscope, magnetometer installed microcontroller board, known as wearable device. The device was fitted on welder's monitor wrist-hand both data measurements sent through Bluetooth connection then saved memory card smartphone. Some properties, such root mean square (RMS), correlation index, spectral peaks, power, extracted from time-series characterize motions. support vector machine (SVM) method, part artificial intelligence (AI) technique, classify recognize two types using supervised learning approach. validation results showed proposed able