A Low-Resolution Used Electronic Parts Image Dataset for Sorting Application
Abstract
:1. Summary
2. Data Description
2.1. SNN and SVM Dataset Formats
- The SNN data use an output array format of n × 3. The three columns are used for placing 0 s or 1 s representing the output class. An additional 4th column contains information about the grayscale image filename.
- For the SVM data, the output array has a format of n × 1. A single numerical value (1, 2, or 3) representing the object class is entered into the array. An additional 2nd column contains information about the grayscale image filename.
2.2. CNN Dataset Format
3. Methods
3.1. Object Detection Process
3.2. Validation
- The size of each unclassified object image is checked against an estimated size threshold representing the dimensions of the smallest component to be detected. This eliminates small images that may have been erroneously detected due to noise or tiny holes in components, such as voltage regulators. The pick and place task assumes that objects are physically separated and do not overlap.
- Pixel grayscale shades are checked against a threshold value. Because all the objects in the current sorting task produce dark shades of gray or black in their grayscale images, an 80% or above shade of gray (i.e., <51 grayscale pixel value) is arbitrarily set as the threshold. At least 5% of the unclassified object image pixels need to meet the threshold. Otherwise, the image is flagged for manual inspection or can be automatically ignored.
4. User Notes
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Arruda, E.H.; Melatto, R.A.P.B.; Levy, W.; Conti, D.d.M. Circular economy: A brief literature review (2015–2020). Sustain. Oper. Comput. 2021, 2, 79–86. [Google Scholar] [CrossRef]
- Chand, P.; Sepulveda, J. Automating a Festo Manufacturing Machine with an Allen-Bradley PLC. J. Mechatron. Robot. 2021, 5, 23–32. [Google Scholar] [CrossRef]
- Chand, P.; Foulkes, M.; Kumar, A.; Ariyarathna, T. Using Simulated Work-Integrated Learning in Mechatronics Courses. In Proceedings of the 2021 IEEE International Conference on Engineering, Technology & Education (TALE), Wuhan, China, 5–8 December 2021; pp. 01–06. [Google Scholar]
- Arduino-Tech. DIY LM317 Adjustable Regulated Voltage Module Suite Kit DC/AC Input. Available online: https://www.arduino-tech.com/diy-lm317-adjustable-regulated-voltage-module-suite-kit-dc-ac-input/ (accessed on 20 October 2022).
- Chand, P. Investigating Vision Based Sorting of Used Items. In Proceedings of the 2022 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET), Kota Kinabalu, Malaysia, 13–15 September 2022; pp. 01–05. [Google Scholar]
- Chand, P.; Lal, S. Vision-based Detection and Classification of Used Electronics Parts. Sensors 2022, 22, 9079. [Google Scholar] [CrossRef] [PubMed]
- Reza, M.A.; Chen, Z.; Crandall, D.J. Deep Neural Network–Based Detection and Verification of Microelectronic Images. J. Hardw. Syst. Secur. 2020, 4, 44–54. [Google Scholar] [CrossRef]
- Goobar, L. Machine Learning Based Image Classification of Electronic Components; KTH: Stockholm, Sweden, 2013. [Google Scholar]
- Xu, Y.; Yang, G.; Luo, J.; He, J. An Electronic Component Recognition Algorithm Based on Deep Learning with a Faster SqueezeNet. Math. Probl. Eng. 2020, 2020, 2940286. [Google Scholar] [CrossRef]
- Atik, I. Classification of Electronic Components Based on Convolutional Neural Network Architecture. Energies 2022, 15, 2347. [Google Scholar] [CrossRef]
- Guo, C.; Lv, X.-L.; Zhang, Y.; Zhang, M.-L. Improved YOLOv4-tiny network for real-time electronic component detection. Sci. Rep. 2021, 11, 22744. [Google Scholar] [CrossRef] [PubMed]
- Huang, R.; Gu, J.; Sun, X.; Hou, Y.; Uddin, S. A Rapid Recognition Method for Electronic Components Based on the Improved YOLO-V3 Network. Electronics 2019, 8, 825. [Google Scholar] [CrossRef] [Green Version]
- Acharya, S. Electronics Components and Devices. Available online: https://www.kaggle.com/datasets/aryaminus/electronic-components (accessed on 20 October 2022).
- Goodfellow, I.; Bengio, Y.; Courville, A. Deep Learning; MIT Press: Cambridge, MA, USA, 2016; p. 800. [Google Scholar]
- Jiang, P.; Ergu, D.; Liu, F.; Cai, Y.; Ma, B. A Review of Yolo Algorithm Developments. Procedia Comput. Sci. 2022, 199, 1066–1073. [Google Scholar] [CrossRef]
- Lowe, D.G. Distinctive Image Features from Scale-Invariant Keypoints. Int. J. Comput. Vis. 2004, 60, 91–110. [Google Scholar] [CrossRef]
- Digi-Key. Passives. Available online: https://www.digikey.co.nz/en/products/super-category/passives/21 (accessed on 20 October 2022).
- RS-Components. Passive Components. Available online: https://nz.rs-online.com/web/c/passive-components/ (accessed on 20 October 2022).
- Chand, P. Developing a Matlab Controller for Niryo Ned Robot. In Proceedings of the 2022 International Conference on Technology Innovation and Its Applications (ICTIIA), Virtual, 23–25 September 2022; pp. 01–05. [Google Scholar]
- Niryo. NED User Manual. Available online: https://docs.niryo.com/product/ned/v4.0.0/en/index.html (accessed on 30 June 2022).
- Level. Appropriate Lighting Levels. Available online: https://www.level.org.nz/energy/lighting-design/appropriate-lighting-levels/ (accessed on 21 October 2022).
- Qi, L.; Wang, L.; Huo, J.; Shi, Y.; Gao, Y. GreyReID: A Novel Two-stream Deep Framework with RGB-grey Information for Person Re-identification. ACM Trans. Multimed. Comput. Commun. Appl. 2021, 17, 1551–6857. [Google Scholar] [CrossRef]
- Nixon, M.S.; Aguado, A.S. Feature Extraction and Image Processing for Computer Vision, 4th ed.; Elsevier Academic Press: London, UK, 2019; p. 650. [Google Scholar]
Model | Main Parameters | Dataset Test Accuracy % |
---|---|---|
SNN | 900 inputs Single hidden layer with 80 neurons 3 outputs Scaled conjugate gradient training method | 93.5 |
SVM | 900 inputs 3-class one vs all approach Cubic kernel function | 94.9 |
SVM + PCA | 900 inputs reduced to 20 components with PCA 3-class one vs all approach Cubic kernel function | 94.6 |
CNN | 30 × 30 pixel grayscale image input 3 convolution layers 2 pooling layers 1 fully connected layer | 98.4 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Chand, P. A Low-Resolution Used Electronic Parts Image Dataset for Sorting Application. Data 2023, 8, 20. https://doi.org/10.3390/data8010020
Chand P. A Low-Resolution Used Electronic Parts Image Dataset for Sorting Application. Data. 2023; 8(1):20. https://doi.org/10.3390/data8010020
Chicago/Turabian StyleChand, Praneel. 2023. "A Low-Resolution Used Electronic Parts Image Dataset for Sorting Application" Data 8, no. 1: 20. https://doi.org/10.3390/data8010020