Phase Extraction of Electronic Speckle Interference Fringe Image based on Convolutional Neural Network

Electronic speckle pattern interferometry (ESPI) is a kind of full-field, non-contact nondestructive measurement technology, which is suitable for deformation measurement and nondestructive testing of optical rough surface. Fringe phase is an important information of interference fringe image, and the accuracy of phase estimation plays an important role in the extraction of fringe information. However, because of the discontinuity of fringe image and the influence of noise, phase extraction has always been a challenging problem. In this paper, we propose a fringe image phase extraction technique based on convolutional neural network, and experimental results show that this method has a good effect on phase estimation of the interference fringe.