The Primena prepoznavanja saobraćajnih znakova na skaliranom modelu vozila
Sažetak
Popularnost autonomnih vozila je u poslednjih nekoliko godina značajnije porasla, pošto su sistemi koji omogućavaju određene autonomne funkcije na vozilu sve prisutniji. Studentima mašinskog i softverskog inženjerstva najpristupačniji način da steknu osnovna znanja o autonomnim vozilima predstavlja primena algoritama i sistema neophodnih za autonomnu vožnju na skaliranom modelu vozila. Ovakvi modeli, kao i u ovom slučaju, su opremljeni sistemima neophodnim za autonomnu vožnju, kao što su pogon na sva četiri točka, sistem oslanjanja, elektronski sistem upravljanja, računar kao „mozak” vozila i kamera. Cilj projekata, kao što je i ovaj, je da se osposobi vozilo za autonomnu vožnju na unapred predviđenoj stazi, poštujući sve saobraćajne propise i znakove (na primer, vozilo mora da se potpuno zaustavi ispred znaka stop). Kako bi ovo bilo moguće, neophodno je da vozilo „zna“ koji se saobraćajni znak nalazi u njegovoj neposrednoj blizini, tj. neophodno je prepoznavanje saobraćajnih znakova. U ovom slučaju je korišćena veštačka neuronska mreža za prepoznavanje saobraćajnih znakova. U radu je opisan proces obučavanja odgovarajuće neuronske mreže.
Reference
Alammar J. (2018). Weight (Artificial Neural Network). Retrieved from http://jalammar.github.io/visual-interactive-guide-basics-neural-networks/ in September 2022.
Brownlee J. (2019). Loss and Loss Function for Training Deep Learning Neural Networks. Retrieved from https://machinelearningmastery.com/loss-and-loss-functions-for-training-deep-learning-neural-networks/ in September 2022.
Daoust M., & Wang T. (2022). TensorFlow Lite Python object detection example with Raspberry Pi. Retrieved fromhttps://github.com/tensorflow/examples/tree/master/lite/examples/object_detection/raspberry_pi in February 2022.
Escalera S., Barò X., Pujol O., Vitrià J., & Radeva P. (2011). Traffic-Sign Recognition Systems. London, England: Springer.
Gao X., Podladchikova L., Shaposhnikov D., Hong K., & Shevtsova N. (2005). Recognition of traffic signs based on their colour and shape features extracted using human vision models. Journal of Visual Communication and Image Representation, 17, 675-685. ELSEVIER.
Hymel S. (1995). How to Perform Object Detection with TensorFlow Lite on Raspberry Pi. Retrieved from https://www.digikey.com/en/maker/projects/how-to-perform-object-detection-with-tensorflow-lite-on-raspberry-pi/b929e1519c7c43d5b2c6f89984883588 in December 2021.
Institut Für Nuroinformatik (2010). German Traffic Sign Recognition Benchmar. Retrieved from https://benchmark.ini.rub.de/
Lee H., & Song J. (2019). Introduction to convolutional neural network using Keras; an understanding from statistician. Communications for Statistical Applications and Methods Journal, 26(6), 591-610.
Lopez, L.-D., Fuentes, O. (2007). Color-Based Road Sign Detection and Tracking. In: Kamel, M., Campilho, A. Image Analysis and Recognition. ICIAR 2007, 1138–1147. Lecture Notes in Computer Science, 4633, Springer.
Paul P., Nandi D., Saif S., Zubair K., & Shubho S.-A. (2018). Traffic Sign Detection Based on Colour Segmentation of Obscure Image Candidates: A Comprehensive Study. International Journal Of Modern Education And Computer Science, 6, 35-46.
Rakshit S. (2020). Udacity Car Dataset CrowdAI. Retrieved from https://www.kaggle.com/datasets/soumikrakshit/udacity-car-dataset-crowdai
in April 2022.
Stameković D., (2022). Autonomous motor vehicle control model (Doctoral Dissertation). University of Belgrade, Belgrade, Serbia.
Stamenković D., Popović V., & Blagojević I., (2017). A brief review of strategies used to control an autonomous vehicle. Paper presented at the Second Maintenance Forum 2017, Bečići, Montenegro.
Stameković D., Popović V., & Vorotović G., (2016). Lane detection algorithm using image processing for autonomous vehicle model. Paper presented at Euromaintenance 2016, Athens, Greece.
The TensorFlow Authors (2021). Model Maker Object Detection for Android Figurine. Retrieved from
https://colab.research.google.com/github/khanhlvg/tflite_raspberry_pi/blob/main/object_detection/Train_custom_model_tutorial.ipynb#scrollTo=UNRhB8N7GHXj in February 2022.
Timofte R. (n.d.), Belgium Traffic Sign Dataset. Retrieved from https://btsd.ethz.ch/shareddata/ in January 2022.
Wang L., Shi J., Song G., & Shen I. (2007). Object Detection Combining Recognition and Segmentation. Retrieved from https://www.cis.upenn.edu/~jshi/ped_html/ in March 2022.