TY - JOUR
T1 - Intersecting Machining Feature Localization and Recognition via Single Shot Multibox Detector
AU - Shi, Peizhi
AU - Qi, Qunfen
AU - Qin, Yuchu
AU - Scott, Paul
AU - Jiang, Jane
N1 - Funding Information:
Manuscript received May 12, 2020; revised September 9, 2020; accepted October 7, 2020. Date of publication October 13, 2020; date of current version February 22, 2021. This work was supported in part by the EPSRC UKRI Innovation Fellowship (Ref. EP/S001328/1), in part by EPSRC Future Advanced Metrology Hub (Ref. EP/P006930/1), and in part by EPSRC Fellowship in Manufacturing (Ref. EP/R024162/1). Paper no. TII-20-2436. (Corresponding author: Qunfen Qi.) The authors are with the EPSRC Future Advanced Metrology Hub, School of Computing and Engineering, University of Huddersfield, Hud-dersfield HD1 3DH, U.K. (e-mail: p.shi@hud.ac.uk; q.qi@hud.ac.uk; y.qin@hud.ac.uk; p.j.scott@hud.ac.uk; x.jiang@hud.ac.uk).
Publisher Copyright:
© 2005-2012 IEEE.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2021/5/1
Y1 - 2021/5/1
N2 - In Industrie 4.0, machines are expected to become autonomous, self-aware and self-correcting. One important step in the area of manufacturing is feature recognition that aims to detect all the machining features from a 3-D model. In this research area, recognizing and locating a wide variety of highly intersecting features are extremely challenging as the topology information of features is substantially damaged because of the feature intersection. Motivated by the single shot multibox detector (SSD), this article presents a novel deep learning approach named SsdNet to tackle the machining feature localization and recognition problem. The typical SSD is designed for 2-D image objection detection rather than 3-D feature recognition. Therefore, the network architecture and output of SSD are modified to fulfil the purpose of this research. In addition, some advanced techniques are also utilized to further enhance the recognition performance. Experimental results on the benchmark dataset confirm that the proposed method achieves the state-of-the-art feature recognition performance (95.20% F-score), localization performance (90.62% F-score), and recognition efficiency (243.85 ms per model).
AB - In Industrie 4.0, machines are expected to become autonomous, self-aware and self-correcting. One important step in the area of manufacturing is feature recognition that aims to detect all the machining features from a 3-D model. In this research area, recognizing and locating a wide variety of highly intersecting features are extremely challenging as the topology information of features is substantially damaged because of the feature intersection. Motivated by the single shot multibox detector (SSD), this article presents a novel deep learning approach named SsdNet to tackle the machining feature localization and recognition problem. The typical SSD is designed for 2-D image objection detection rather than 3-D feature recognition. Therefore, the network architecture and output of SSD are modified to fulfil the purpose of this research. In addition, some advanced techniques are also utilized to further enhance the recognition performance. Experimental results on the benchmark dataset confirm that the proposed method achieves the state-of-the-art feature recognition performance (95.20% F-score), localization performance (90.62% F-score), and recognition efficiency (243.85 ms per model).
KW - Industrie 4.0
KW - 3d feature localisation
KW - feature recognition
KW - single shot multibox detector
KW - deep learning
KW - 3-D feature localization
KW - Deep learning
KW - single shot multibox detector (SSD)
UR - http://www.scopus.com/inward/record.url?scp=85101754086&partnerID=8YFLogxK
U2 - 10.1109/TII.2020.3030620
DO - 10.1109/TII.2020.3030620
M3 - Article
VL - 17
SP - 3292
EP - 3302
JO - IEEE Transactions on Industrial Informatics
JF - IEEE Transactions on Industrial Informatics
SN - 1551-3203
IS - 5
M1 - 9222288
ER -