Abstract
For defense applications, rapid X-ray inspection of propellant samples is essential for the identification and assessment of defects. Automation of this process using artificial intelligence is possible by properly training a neural network model. Convolution Neural Networks (CNNs) have recently demonstrated excellent success in both the tasks of image recognition and localisation using an adequate amount of data. In real-world, it’s not an easy task to produce the correct amount of experimental data required for the deep neural network to operate. In this work, we propose a method for producing synthetic radiographic data that is supported by ray tracing based radiographic simulations for the deep learning algorithms to automatically detect anomaly in X-ray images. The simulation results, which are then supplemented by noise extracted from the experimental data, show a good comparison with the measurements. This Simulation assisted Automatic Defect Recognition (Sim-ADR) system simultaneously perform defect detection and defect instance segmentation. The accuracy of the defect detection system is more than 87% on a testing set included 416 images.
Similar content being viewed by others
References
Brooks, W.: Solid propellant grain design and internal ballistics, vol 8076. National Aeronautics and Space Administration (1972)
Davenas, A.: Solid Rocket Propulsion Technology. Newnes, Oxford (2012)
Halmshaw, R.: Industrial Radiology: Theory and Practice, vol. 21. Chapman Hall, London (2012)
Ghose, B., Kankane, D.: Estimation of location of defects in propellant grain by X-ray radiography. NDT E Int. 41(2), 125–128 (2008)
Deprins, E.: Digital radiography in ndt applications. CINDE J. 25(4), 14–16 (2004)
Ravindran, V.: Digital radiography using flat panel detector for the non-destructive evaluation of space vehicle components. Journal of Non-Destructive Testing & Evaluation 4(2), (2005)
Willems, P., et al.: Image quality comparison for digital radiography systems for ndt. In: 15th WCNDT (2000)
Godoi, W., da Silva, R., Swinka-Filho, V.: Pattern recognition in the automatic inspection of flaws in polymeric insulators. Insight Non-Destruct. Test. Cond. Monitor. 47(10), 608–614 (2005)
Liao, T.W., Li, Y.: An automated radiographic ndt system for weld inspection: part iiflaw detection. Ndt E Int. 31(3), 183–192 (1998)
Wang, G., Liao, T.W.: Automatic identification of different types of welding defects in radiographic images. Ndt E Int. 35(8), 519–528 (2002)
Da Silva, R.R., Siqueira, M.H., de Souza, M.P.V., Rebello, J.M., Calôba, L.P.: Estimated accuracy of classification of defects detected in welded joints by radiographic tests. NDT E Int. 38(5), 335–343 (2005)
Saravanan, T., Bagavathiappan, S., Philip, J., Jayakumar, T., Raj, B.: Segmentation of defects from radiography images by the histogram concavity threshold method. Insight Non-Destruct. Test. Cond. Monitor. 49(10), 578–584 (2007)
Da Silva, R.R., Mery, D.: Radiographic testing: part l image processing. Mater. Eval. 643, 1–9 (2007)
Zahran, O., Kasban, H., El-Kordy, M., Abd El-Samie, F.: Automatic weld defect identification from radiographic images. Ndt E Int. 57, 26–35 (2013)
Lashkia, V.: Automatic weld defect identification from radiographic images. Image Vis. Comput. 19(5), 261–269 (2001)
Da Silva, R., Siqueira, M.H., Caloba, L.P., Rebello, J.: Radiographics pattern recognition of welding defects using linear classifiers. Insight 43(10), 669–74 (2001)
Da Silva, R., Calôba, L., Siqueira, M., Sagrilo, L., Rebello, J.: Evaluation of the relevant characteristic parameters of welding defects and probability of correct classification using linear classifiers. Insight 44(10), 616–22 (2002)
Liao, T.W.: Improving the accuracy of computer-aided radiographic weld inspection by feature selection. Ndt E Int. 42(4), 229–239 (2009)
Valavanis, I., Kosmopoulos, D.: Multiclass defect detection and classification in weld radiographic images using geometric and texture features. Expert Syst. Appl. 37(12), 7606–7614 (2010)
Zapata, J., Vilar, R., Ruiz, R.: Performance evaluation of an automatic inspection system of weld defects in radiographic images based on neuro-classifiers. Expert Syst. Appl. 38(7), 8812–8824 (2011)
Boaretto, N., Centeno, T.M.: Automated detection of welding defects in pipelines from radiographic images dwdi. Ndt E Int. 86, 7–13 (2017)
Baniukiewicz, P.: Automated defect recognition and identification in digital radiography. J. Nondestruct. Eval. 33(3), 327–334 (2014)
Zapata, J., Vilar, R., Ruiz, R.: Automatic inspection system of welding radiographic images based on ann under a regularisation process. J. Nondestruct. Eval. 31(1), 34–45 (2012)
Ferguson, M.K., Ronay, A., Lee, Y.T.T., Law, K.H.: Detection and segmentation of manufacturing defects with convolutional neural networks and transfer learning. Smart Sustain. Manuf. Syst. 2, 10 (2018)
Mery, D., Riffo, V., Zscherpel, U., Mondragón, G., Lillo, I., Zuccar, I., Lobel, H., Carrasco, M.: Gdxray: The database of X-ray images for nondestructive testing. J. Nondestruct. Eval. 34(4), 42 (2015)
Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp. 91–99 (2015)
Duvauchelle, P., Freud, N., Kaftandjian, V., Babot, D.: A computer code to simulate x-ray imaging techniques. Nucl. Instrum. Methods Phys. Res. Sect. B 170(1–2), 245–258 (2000)
Lazos, D., Bliznakova, K., Kolitsi, Z., Pallikarakis, N.: An integrated research tool for X-ray imaging simulation. Comput. Methods Progr. Biomed. 70(3), 241–251 (2003)
Fanti, V., Marzeddu, R., Massazza, G., Randaccio, P.: A simulation tool to support teaching and learning the operation of X-ray imaging systems. Med. Eng. Phys. 27(7), 555–559 (2005)
Bellon, C., Jaenisch, G.R.: Artist-analytical rt inspection simulation tool. In: Proc DIR, pp. 25–27 (2007)
Tucker, D., Barnes, G., Chakraborty, D.: Semiempirical model for generating tungsten target X-ray spectra. Med. Phys. 18, 211–8 (1991a). https://doi.org/10.1118/1.596709
Tucker, D.M., Barnes, G.T., Wu, X.: Molybdenum target X-ray spectra: a semiempirical model. Med. Phys. 18(3), 402–407 (1991b). https://doi.org/10.1118/1.596686
He, K., Gkioxari, G., Dollr, P., Girshick, R.: Mask r-cnn. In: 2017 IEEE International Conference on Computer Vision (ICCV), pp. 2980–2988 (2017)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition, pp. 770–778 (2016). https://doi.org/10.1109/CVPR.2016.90
Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L.: Microsoft coco: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) Computer Vision - ECCV 2014, pp. 740–755. Springer, Cham (2014)
Abdulla, W.: Mask r-cnn for object detection and instance segmentation on keras and tensorflow (2017). https://github.com/matterport/Mask_RCNN
Sun, C., Shrivastava, A., Singh, S., Gupta, A.: Revisiting unreasonable effectiveness of data in deep learning era (2017)
Joulin, A., van der Maaten, L., Jabri, A., Vasilache, N.: Learning visual features from large weakly supervised data (2015)
Acknowledgements
This study was funded by Armaments Research Board (ARMREB) and supported by High Energy Material Research Laboratory (HEMRL). We would like to thank Anirudha Sane and Deepak Patil from HEMRL, Pune for helping us in procuring the experimental data.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This document is the results of the research project funded by Armaments Research Board (ARMREB) in collaboration with the High Energy Material Research Laboratory (HEMRL).
Rights and permissions
About this article
Cite this article
Gamdha, D., Unnikrishnakurup, S., Rose, K.J.J. et al. Automated Defect Recognition on X-ray Radiographs of Solid Propellant Using Deep Learning Based on Convolutional Neural Networks. J Nondestruct Eval 40, 18 (2021). https://doi.org/10.1007/s10921-021-00750-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10921-021-00750-4