External Thread Measurement Based on ResUnet and HMM

Abstract

There are many methods of thread measurement. In addition, these thread measurement methods require manual segmentation of regions of interest (threaded area) and These methods are easily disturbed by the environment (e.g. dust, iron filings, oil stains, etc.), resulting in inaccurate measurement results. This paper proposes an external thread measurement method based on ResUnet and hidden Markov model (HMM). First, we propose a ResUnet-based thread edge recognition method that omits the process of calibrating the threaded area and identify the thread edge in a complex environment. Secondly, we use HMM to classify the thread edge points so that the threaded parts can be placed at any angle during the measurement, simplifying the measurement steps and calculating the thread parameters based on the classification results. Finally, we evaluated our method using our own dataset, and the results showed that the difference between the measured value and the standard value is within 0.01 mm.