Abstract
A novel effective method to detect pill defects during pill manufacturing is proposed in this study. We have developed an analysis program that incorporates deep learning convolutional neural networks to fully automate the image analysis of pills for internal crack detection. The deep learning tool based on YOLO algorithm is effectively implemented into the industrial pharmaceutical workflow. Firstly, we analyze Gauss filtering and smoothing techniques for pill detection. Secondly, Hog feature extraction is introduced to simplify the representation of the image that contains only the most important information about the image. Lastly, improved YOLO is designed for online detection of pill defects. The proposed approach obtains robust quantification of internal pill cracks.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Tobore I et al (2019) Deep learning intervention for health care challenges: some biomedical domain considerations. JMIR Mhealth Uhealth 7(8):e11966 (2019)
Pillbox (2021). http://pillbox.nlm.nih.gov/pillimage/search.php. Accessed 9 Mar 2021
Pill Identifier (2021). http://reference.medscape.com/pill-identifier. Accessed 9 Mar 2021
Pill Identification Tool (2021). http://www.webmd.com/pill-identification. Accessed 9 Mar 2021
Guo Q, Zhang C, Liu H, Zhang X (2016) Defect detection in tire X-ray images using weighted texture dissimilarity. J Sens 2016
Yuan J, Wang Q, Li B (2014) A flexile and high precision calibration method for binocular structured light scanning system. Sci World J 2014(8)
Redmon J, Farhadi A (2018) Yolov3: an incremental improvement. http://arxiv.org/abs/1804.02767
Girshick R (2015) Fast r-cnn. In: 2015 IEEE international conference on computer vision (ICCV), pp 1440–1448, Santiago, Chile, Dec 2015
Liu W et al (2016) Lect Notes Comput Sci 9905:21
Sabri AH et al (2018) J Drug Deliv Sci Technol 46:16
Lin T et al (2017) IEEE conference on computer vision and pattern recognition (CVPR), vol 2117
Acknowledgements
This work was supported by the Autonomous Higher Education Project (SAHEP) grant funded under number T2020-SAHEP-012.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Mac, T.T. (2023). Pill Defect Detection Using an Improved Convolutional Neural Network. In: Conte, G., Sename, O. (eds) Proceedings of 10th International Conference on Mechatronics and Control Engineering . ICMCE 2021. Lecture Notes in Mechanical Engineering. Springer, Singapore. https://doi.org/10.1007/978-981-19-1540-6_7
Download citation
DOI: https://doi.org/10.1007/978-981-19-1540-6_7
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-1539-0
Online ISBN: 978-981-19-1540-6
eBook Packages: EngineeringEngineering (R0)