Objects Detection for Self-Driving Cars using YOLO Algorithm
Abstract
In the world of trending tech and autonomous uprising capacity of machines, the benchmarks set by both of the advancements mentioned above resulted in bleeding-edge technology, namely Self-Driving Cars or Autonomous Vehicles (AV's). An autonomous vehicle is one that navigates between destinations without the assistance of a human operator, using data from sensors, cameras, radar, and artificial intelligence (AI). Self-driving cars' functionality lies in how the developers use vast amounts of data from machine learning and neural networks. They create and store a map of their surroundings based on various sensors placed on different vehicle parts. The sensors used are Radar, Lidar (light detection and ranging), ultrasonic sensors. After processing all of this sensory data, sophisticated software creates a route and sends instructions to the car's actuators, which control acceleration, braking, and steering.
The main difficulty in AVs comes through the detection of objects which were surrounded by the vehicle.As a result, many algorithms were developed, like HOG (Histograms of Oriented Gradients, 2005), R-CNN (Region Convolutional Neural Network), and YOLO (YOU ONLY LOOK ONCE). Out of them, YOLO is a transparent convolutional neural network for detecting the object in real-time. This method has several advantages compared to the algorithms mentioned above because it looks at the image entirely by predicting the bounding boxes to see the kind of object. As of now, there are four versions of YOLO developed from YOLO v1 to YOLO v4. YOLO is fast and accurate, and thus it outperforms the other algorithms and is so used widely across many aspects like self-driving cars.
How to cite this article:
Kumar NL, Maram B. Objects Detection for Self Driving Cars using YOLO Algorithm. J Adv Res Auto Tech Transp Sys 2021; 5(1): 17-23.
References
[2] Yin, Yunhua, Huifang Li, and Wei Fu. "Faster-YOLO: An accurate and faster object detection method." Digital Signal Processing (2020): 102756.
[3]Geethapriya, S., N. Duraimurugan, and S. P. Chokkalingam. "Real-Time Object Detection with Yolo." International Journal of Engineering and Advanced Technology (IJEAT) 8.3S (2019).
[4]Coutinho, Alston Brazil, and Rehan Ali Mirza. "Autonomous Mobility Vehicle." 2020 IEEE International IOT, Electronics and Mechatronics Conference (IEMTRONICS). IEEE, 2020.
[5] Kim, Sung-Dong. "Situation-cognitive traffic light control based on object detection using YOLO algorithm." International Journal of Computational Vision and Robotics 10.2 (2020): 133-142.
[6] K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 770–778.
[7] A.A.M. Al-Saffar, H. Tao, M.A. Talab, Review of deep convolution neural network in image classification, in: 2017 International Conference on Radar, Antenna, Microwave, Electronics, and Telecommunications (ICRAMET), 2017, pp. 26–31.
[8] Y. Li, H. Zhang, X. Xue, et al., Deep learning for remote sensing image classification: a survey, Wiley Interdiscip. Rev. Data Min. Knowl. Discov. 8 (6) (2018) e1264.
[9] A. Ferreyra-Ramirez, C. Aviles-Cruz, E. Rodriguez-Martinez, et al., An improved convolutional neural network architecture for image classification, Pattern Recognit. 11524 (2019) 89–101.
[10] M.A. Kadhim, M.H. Abed, Convolutional neural network for satellite image classification, in: Studies in Computational Intelligence, Intelligent Information and Database Systems: Recent Developments, vol. 830, Springer, Cham, 2019, pp. 165–178.
[11] M.A.O. Ahmed, O. Reyad, B.A. El-Rahiem, An efficient deep convolutional neural network for visual image classification, in: The International Conference on Advanced Machine Learning Technologies and Applications (AMLTA2019), vol. 921, 2019, pp. 23–31.
[12] P. Viola, M. Jones, Rapid object detection using a boosted cascade of simple features, in: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, 2001, pp. I511–I518.
