Sistem Pembatasan Kecepatan Autonomous Car di Indonesia
DOI:
https://doi.org/10.56706/ik.v17i2.77Keywords:
weather condition detection, CNN, ResNet18, surveillance system, autonomous carAbstract
Perkembangan teknologi mobil yang semakin cepat membutuhkan sistem pengawasan yang lebih canggih dengan artificial intelligence (AI). Kecepatan mobil otomatis perlu dibatasi berdasarkan kondisi cuaca yang terdeteksi oleh kamera dengan penerapan pendekatan AI yang lebih dalam, yaitu deep learning (DL). Metode yang digunakan dalam penelitian ini adalah klasifikasi gambar dengan pre-trained model. Pada penelitian ini, dilakukan eksperimen menggunakan teknologi pengolahan gambar yang terintegrasi dengan AI untuk pendeteksian kondisi cuaca saat mobil bergerak. Keunggulan dari pendekatan DL adalah kemampuannya dalam mempelajari pola dan fitur yang lebih kompleks dalam gambar, sehingga mampu mengenali perubahan cuaca dengan lebih baik. Dalam penelitian ini, dilakukan perbandingan kinerja antara algoritma Convolutional Neural Network (CNN) dan ResNet18 dalam mengklasifikasikan gambar berdasarkan kondisi cuaca yang terdeteksi. Hasilnya menunjukkan bahwa kedua algoritma tersebut memiliki keunggulan dan kelemahan masing-masing. CNN memiliki keunggulan dalam mengenali fitur-fitur umum dalam gambar, sedangkan ResNet18 memiliki kemampuan yang lebih baik dalam mengatasi masalah klasifikasi gambar yang kompleks. Secara keseluruhan, penelitian ini menunjukkan bahwa penggunaan AI dengan pendekatan DL dan teknologi pengolahan gambar dapat menjadi solusi yang efektif dalam meningkatkan sistem pengawasan autonomous car. Dengan kemampuan untuk mendeteksi kondisi cuaca secara akurat, sistem ini dapat memberikan informasi yang relevan kepada pengemudi atau sistem pengendalian otomatis, sehingga meningkatkan keamanan dan kenyamanan dalam berkendara.
The rapid development of automobile technology requires a more sophisticated surveillance system with artificial intelligence (AI). The speed of autonomous cars needs to be limited based on the weather conditions detected by the camera by applying a deeper AI approach, namely deep learning (DL). The method used in this research is image classification with pre-trained models. This research conducted experiments using image processing technology integrated with AI to detect weather conditions while the car is moving. The advantage of the DL approach is its ability to learn more complex patterns and features in images so that it can better recognize weather changes. This study used a performance comparison between Convolutional Neural Network (CNN) and ResNet18 algorithms to classify images based on detected weather conditions. The results show that both algorithms have their advantages and disadvantages. CNN has the advantage of recognizing common features in images, while ResNet18 has a better ability to tackle complex image classification problems. Overall, this research shows that the use of AI with a DL approach and image processing technology can be an effective solution in improving autonomous car surveillance systems. With the ability to accurately detect weather conditions, the system can provide relevant information to the driver or the automatic control system, thereby improving driving safety and comfort.
References
A. Cord and D. Aubert, "Towards rain detection through use of in-vehicle multipurpose cameras," 2011 IEEE Intelligent Vehicles Symposium (IV), Baden-Baden, Germany, 2011, pp. 833-838, doi: 10.1109/IVS.2011.5940484.
K. Yoneda, N. Suganuma, R. Yanase and M. Aldibaja, “Automated driving recognition technologies for adverse weather conditions,” IATSS Research, 2019, pp. 253–262. https://doi.org/10.1016/J.IATSSR.2019.11.005.
L. -W. Kang, K. -L. Chou and R. -H. Fu, "Deep Learning-Based Weather Image Recognition," 2018 International Symposium on Computer, Consumer and Control (IS3C), Taichung, Taiwan, 2018, pp. 384-387, doi: 10.1109/IS3C.2018.00103.
D. A. Padilla, J. F. Villaverde, J. J. T. Magdaraog, A. J. L. Oconer and J. P. Ranjo, "Vehicle and Weather Detection Using Real Time Image Processing Using Optical Flow and Color Histogram," 2019 5th International Conference on Control, Automation and Robotics (ICCAR), Beijing, China, 2019, pp. 880-883, doi: 10.1109/ICCAR.2019.8813346.
M. Hnewa and H. Radha, "Object Detection Under Rainy Conditions for Autonomous Vehicles: A Review of State-of-the-Art and Emerging Techniques," in IEEE Signal Processing Magazine, vol. 38, no. 1, pp. 53-67, Jan. 2021, doi: 10.1109/MSP.2020.2984801.
KENK, Mourad (2020), “DAWN”, Mendeley Data, V3, doi: 10.17632/766ygrbt8y.3
V. Abhyankar, A. G. Singh, P. Paul, A. Mehta and S. Vidhya, "Portable Autonomous Rain Prediction Model Using Machine Learning Algorithm," 2019 International Conference on Vision Towards Emerging Trends in Communication and Networking (ViTECoN), Vellore, India, 2019, pp. 1-4, doi: 10.1109/ViTECoN.2019.8899704.
Jayanthi, K. B., K. Vanitha, M. Malathi, J. Chandrasekar, and R. Khavya, “An Efficient Weather Recognition Algorithm on Highway Roads for Vehicle Guidance,” International Journal of Health Sciences, vol. 6, no. S8, 2022, pp. 1212-23, doi:10.53730/ijhs.v6nS8.9882.
X-Z. Chen, C-M. Chang, C-W. Yu and Y-L. Chen, "A Real-Time Vehicle Detection System under Various Bad Weather Conditions Based on a Deep Learning Model without Retraining," Sensors 20, 2020, no. 20: 5731. https://doi.org/10.3390/s20205731.
Z. Paschalidi, W. Acevedo, M. Hellweg, T. Kratzsch, R. Potthast and J. Nachtigall, “Real time weather for autonomous driving and precise road weather forecasts based on floating car data and seamless integration of Ultra-Rapid Data Assimilation and Nowcasting,” EMS Annual Meeting 2021, online, 2021, EMS2021-385, https://doi.org/10.5194/ems2021-385.
M. Elhoseiny, S. Huang and A. Elgammal, "Weather classification with deep convolutional neural networks," 2015 IEEE International Conference on Image Processing (ICIP), Quebec City, QC, Canada, 2015, pp. 3349-3353, doi: 10.1109/ICIP.2015.7351424.
A. Sharma and Z. S. Ismail, “Weather Classification Model Performance: Using CNN, Keras-Tensor Flow,” ITM Web Conf., 2022, doi: 10.1051/itmconf/20224201006.
C. Zhang, E. Nateghinia, L. F. Miranda-Moreno and L. Sun, “Winter road surface condition classification using convolutional neural network (CNN): visible light and thermal image fusion,” Canadian Journal of Civil Engineering., 2021, pp. 569-578, doi: 10.1139/cjce-2020-0613.
Q. A. Al-Haija, M. A. Smadi and S. Zein-Sabatto, "Multi-Class Weather Classification Using ResNet-18 CNN for Autonomous IoT and CPS Applications," 2020 International Conference on Computational Science and Computational Intelligence (CSCI), Las Vegas, NV, USA, 2020, pp. 1586-1591, doi: 10.1109/CSCI51800.2020.00293.
M. N. Khan and M. M. Ahmed, “Weather and surface condition detection based on road-side webcams: Application of pre-trained Convolutional Neural Network,” International Journal of Transportation Science and Technology, 2022, pp. 468-483, doi: 10.1016/j.ijtst.2021.06.003.
M. Sandler, A. Howard, M. Zhu, A. Zhmoginov and L-C. Chen, “MobileNetV2: Inverted Residuals and Linear Bottlenecks,” arXiv:1801.04381v4, 2019.
A. V. S. Abhishek, V. R. Guralla and L. Sahoo, “ResNet18 Model With Sequential Layer For Computing Accuracy On Image Classification Dataset,” International Journal of Creative Research Thoughts (IJCRT), 2022.
Downloads
Submitted
Accepted
Published
Issue
Section
License
Copyright (c) 2023 Info Kripto
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
