Abstract
The Single Shot Multibox Detector (SSD) technique is currently among the fastest and most accurate detection algorithms available. However, the majority of research on the accuracy of this approach has focused on noiseless objects. Thus, this study evaluates the algorithm's accuracy with both noisy and noiseless objects. To that goal, the algorithm is trained to recognize ten different flower species. Experiments are then carried out on photographs in four different scenarios: the item is totally lighted, 1/3 of the object is darkened, 1/2 of the object is darkened, and the object is fully darkened. The performance of the algorithm is then evaluated using SPSS 20.0 software and the analysis of variance (ANOVA) and least significant difference (LSD). The experimental results reveal that the algorithm accuracy is strongly dependent on the noise level. The detection accuracy is 100%, 81.3%, 44.7%, and 62%, respectively, when the item is fully lighted, 1/3, 1/2 size of the object is darkened, and the object is fully darkened.
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
Similar content being viewed by others
References
Liu W et al (2016) SSD: single shot multibox detector. Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 9905:21–37
Shuai Q, Wu X (2020) Object detection system based on SSD algorithm. Proc 2020 Int Conf Cult Sci Technol ICCST 2020, 141–144
Abbas SM, Singh SN (2018) Region-based object detection and classification using faster R-CNN. Int Conf Computational Intell Commun Technol CICT 2018, pp 1–6
Liu B, Zhao W, Sun Q (2017) Study of object detection based on Faster R-CNN. Proc 2017 Chinese Autom Congr CAC 2017, 6233–6236
Kanimozhi S, Gayathri G, Mala T (2019) Multiple real-time object identification using single shot multi-box detection. ICCIDS 2019—2nd Int Conf Comput Intell Data Sci Proc, pp 1–5
Kang HJ (2019) Real-time object detection on 640 × 480 image with VGG16+SSD. Proc 2019 Int Conf Field-Programmable Technol ICFPT 2019, 419–422
Hui J (2018) SSD object detection: single shot multibox detector for real-time processing [Online]. Available: https://jonathan-hui.medium.com/ssd-object-detection-single-shot-multibox-detector-for-real-time-processing-9bd8deac0e06
Drive G (2022) Training image for learning model [Online]. Available: https://drive.google.com/file/d/1FRzMiQQsOQ9uHsJSDkxCRCELoHuPFp-d/view?usp=sharing
Drive G (2022) Image included into identification [Online]. Available: https://drive.google.com/file/d/1ugIOpvn9G9ad-aNn0rjLio_FFT7ELNI5/view?usp=sharing
Drive G (2022) Image is identified by the model [Online]. Available: https://drive.google.com/file/d/1cz58OjMJSp8MwcZjObODEzCc20l0jdsG/view?usp=sharing
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 Switzerland AG
About this paper
Cite this paper
Nguyen, VN. (2023). The Recognition Accuracy in the SSD Model. In: Long, B.T., et al. Proceedings of the 3rd Annual International Conference on Material, Machines and Methods for Sustainable Development (MMMS2022). MMMS 2022. Lecture Notes in Mechanical Engineering. Springer, Cham. https://doi.org/10.1007/978-3-031-31824-5_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-31824-5_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-31823-8
Online ISBN: 978-3-031-31824-5
eBook Packages: Chemistry and Materials ScienceChemistry and Material Science (R0)