Journal Published Online: 11 December 2019
Volume 3, Issue 1

A Standardized PMML Format for Representing Convolutional Neural Networks with Application to Defect Detection

CODEN: SSMSCY

Abstract

Convolutional neural networks are becoming a popular tool for image processing in the engineering and manufacturing sectors. However, managing the storage and distribution of trained models is still a difficult task that is partially due to the lack of standardized methods for deep neural network representation. Additionally, the interoperability between different machine learning frameworks remains poor. This article seeks to address this issue by proposing a standardized format for convolutional neural networks based on the Predictive Model Markup Language (PMML). A new standardized schema is proposed to represent a range of convolutional neural networks, including classification, regression, and semantic segmentation systems. To demonstrate the practical application of this standard, a semantic segmentation model, which is trained to detect casting defects in X-ray images, is represented in the proposed PMML format. A high-performance scoring engine is developed to evaluate images and videos against the PMML model. The utility of the proposed format and the scoring engine is evaluated by benchmarking the performance of the defect detection models on a range of different computational platforms.

Author Information

Ferguson, Max
Department of Civil and Environmental Engineering, Stanford University, Stanford, CA, USA
Lee, Yung-Tsun Tina
Systems Integration Division, National Institute of Standards and Technology, Gaithersburg, MD, USA
Narayanan, Anantha
Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA
Law, Kincho H.
Department of Civil and Environmental Engineering, Stanford University, Stanford, CA, USA
Pages: 19
Price: Free
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: SSMS20190032
ISSN: 2520-6478
DOI: 10.1520/SSMS20190032