Pembuatan Alat Inspeksi Visual Jalur PCB Menggunakan Pengolahan Citra Untuk  Kegiatan Praktikum Pengawatan Dan Teknologi PCB

Rangga Ade Julianto; Efrizon Efrizon; Hendrick Hendrick; Laxsmy Devy; Suryadi Suryadi; Yul Antonisfia

doi:10.30630/eji.14.2.295

Rangga Ade Julianto Politeknik Negeri Padang
Efrizon Efrizon Politeknik Negeri Padang
Hendrick Hendrick Politeknik Negeri Padang
Laxsmy Devy Politeknik Negeri Padang
Suryadi Suryadi Politeknik Negeri Padang
Yul Antonisfia Politeknik Negeri Padang

DOI: https://doi.org/10.30630/eji.14.2.295

Keywords: PCB, Logitech c920 Webcam, Raspberry Pi 3b , Image Processing, YOLO CNN

Abstract

PCBs are very influential on the manufacture of electronic devices, for example when there is even a small number of PCB paths that are cut off or damaged, the electronic device cannot be operated properly. Therefore, in this study, the author tried to create and analyze a defect checking tool on PCBs to replace human vision to make it easier and can save costs. This tool is equipped with the help of a Logitech c920 Webcam and a Raspberry Pi 3b+ microprocessor which is used to store and run programs that have been created on Python programming software, so this tool can be used portablely. With these two technologies, Image Processing can be used to detect objects with the OpenCv library and Google Colab. PCB defect detection tool with the help of Image Processing uses yolo convolutional neural network method to help determine path damage on the PCB. You Only Look Once (YOLO) algorithm with five detection classifications, namely short, open circuit, missing hole, mouse bite, and spur. From the results of the study, the results were obtained that the YOLO algorithm was able to detect these five classifications with a value of mAP@0.5 short 90.67%, open circuit 97.86%, Mouse Bite 94.43%, Missing Hole 96.09%, and spur 97.56%.

Downloads

Download data is not yet available.

References

[1] V. A. Adibhatla, H.-C. Chih, C.-C. Hsu, J. Cheng, M. F. Abbod, and J.-S. Shieh, “Defect Detection in Printed Circuit Boards Using You-Only-Look-Once Convolutional Neural Networks,” Electronics (Basel), vol. 9, no. 9, p. 1547, Sep. 2020, doi: 10.3390/electronics9091547.
[2] Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2019). Deep Learning for Generic Object Detection: A Survey. International Journal of Computer Vision, 128(2), 261–318. DOI: https://doi.org/10.1007/s11263-019-01247-4.
[3] J. T. Terpadu, I. Arifin, R. Fakhran Haidi, and M. Dzalhaqi, “PENERAPAN COMPUTER VISION MENGGUNAKAN METODE DEEP LEARNING PADA PERSPEKTIF GENERASI ULUL ALBAB,” Jurnal Teknologi Terpadu, vol. 7, no. 2, pp. 98–107, 2021, [Online]. Available: https://journal.nurulfikri.ac.id/index.php/jtt
[4] JIWOONG, C., DAYOUNG, C., HYUN, K. & LEE, H.-J., 2019. Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving. Seoul, IEEE International Conference on Computer Vision.
[5] FANG, W., WANG, L. & REN, P., 2020. Tinier-YOLO: A Real-Time Object DetectionMethod for Constrained Environments. IEEE Access, Volume 8, pp. 1935 - 1944.
[6] A. Albar, H. Hendrick, and R. Hidayat, “Segmentation Method for Face Modelling in Thermal Images,” Knowl. Eng. Data Sci., vol. 3, no. 2, p. 99, 2020, doi: 10.17977/um018v3i22020p99-105.
[7] R. Park and J. Jo, “Reference class-based improvement of object detection accuracy,” Int. J. Adv. Sci. Eng. Inf. Technol., vol. 10, no. 4, pp. 1526–1535, 2020, doi: 10.18517/ijaseit.10.4.12792.
[8] W. Nugroho, “Deteksi Kerusakan Jalur PCB ( Printed Circuit Board ) Menggunakan Metode Template Matching,” 2014.
[9] M. Metode, P. T. Karya, and M. Nugraha, “Penerapan Metode Template Matching Dalam Menganalisa Cacat Pada Keping PCB” Jurnal Vokasional Teknik Elektronika & Informatika, vol. 1, no. 2, pp. 88–93, 2019.
[10] N. Kurniasari and J. P. Sugiono, “DETEKSI JALUR YANG TERPUTUS PADA RANGKAIAN LISTRIK DALAM PCB MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN),” 2021.