Pemanfaatan Deep Learning pada Video Dash Cam untuk Deteksi Pengendara Sepeda Motor
Abstract
The number of motorcyclists in Indonesia was 105.15 million in 2016. It made the Indonesian government difficult to monitor motorcyclists on the highways. Dash cam could be used as the alternative tool to detect motorcyclists when given the intelligence. One of the typical drawbacks in detecting objects is complex and varied feature. A convolutional neural networks (CNN) that was capable of detecting motorcyclists was proposed. CNN successfully classified the ship object with f1-score of 0.94. Sliding window and heat map were used in thispaper to search the localization and region of motorcyclists. Two experiments had been done in this paper. The goal of this paper was to set the best combination of CNN architecture and parameter. The first experiment consisted of three trained weights while the second experiment consisted of one trained weight. Weight peformances against test data in experiment 1 and experiment 2 were measured using f1-score of 0.977, 0.988, 0.989, and 0.986, respectively. From the experimental results using the sliding window, experiment 2 had a lower error rate to predict motorcyclists than experiment 1 because the training data on experiment 1 contained more and various images.
References
A. Soin dan M. Chahande, “Moving Vehicle Detection Using Deep Neural Network,” 2017 International Coneference on Emerging Trends in Computing and Communication Technologies (ICETCCT), 2017, hal. 1–5.
H. Tayara, K. G. Soo, dan K. T. Chong, “Vehicle Detection and Counting in High-Resolution Aerial Images Using Convolutional Regression Neural Network,” IEEE Access, Vol. 6, hal. 2220–2230, 2018.
(2017) “2016, Jumlah Sepeda Motor Indonesia Tembus Seratus Juta,” [Online], https://databoks.katadata.co.id/datapublish/2017/12/20/2016-jumlah-sepeda-motor-indonesia-tembus-100-juta, tanggal akses: 12 Januari 2018.
D. Liu, Y. Xiong, K. Pulli, dan L. Shapiro, “Estimating Image Segmentation Difficulty,” International Workshop on Machine Learning and Data Mining in Pattern Recognition, 2011, Vol. 6871, hal. 484–495.
R. T. Ionescu, B. Alexe, M. Leordeanu, M. Popescu, D. P. Papadopoulos, dan V. Ferrari, “How hard can it be ? Estimating the difficulty of visual search in an image,” Computer Vision and Pattern Recognition Conference (CVPR)2, 2016, hal. 2157–2166.
A. Krizhevsky, I. Sutskever, dan G. E. Hinton, “Imagenet Classification With Deep Convolutional Neural Networks,” Proceedings of the 25th International Conference on Neural Information Processing Systems, 2012, Vol. 1, hal. 1097–1105.
J. Schmidhuber, “Deep Learning in Neural Networks: An Overview,” Neural Networks, Vol. 61, hal. 85–117, 2015.
Y. Lecun, Y. Bengio, dan G. Hinton, “Deep learning,” Nature, Vol. 521, No. 7553, hal. 436–444, 2015.
D. Molin, “Pedestrian Detection Using Convolutional Neural Networks,” Thesis, Linkoping University, 2015.
Y. LeCun et al., “Backpropagation Applied to Handwritten Zip Code Recognition,” Neural Computation, Vol. 1, No. 4, hal. 541–551, 1989.
S. Faghih-Roohi, S. Hajizadeh, A. Nunez, R. Babuska, dan B. De Schutter, “Deep Convolutional Neural Networks for Detection of Rail Surface Defects,” Proceedings of the 2016 International Joint Conference on Neural Networks (IJCNN), 2016, hal. 2584–2589.
A. Giusti, D. C. Cire, J. Masci, L. M. Gambardella, and J. Schmidhuber, “Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks,” International Conference on Image Processing (ICIP), 2013, hal. 4034–4038.
N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, dan R. Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting,” Journal of Machine Learning Research, Vol. 15, hal. 1929–1958, 2014.
S. E. Limantoro, Y. Kristian, dan D. D. Purwanto, “Deteksi Pengendara Sepeda Motor Menggunakan Deep Convolutional Neural Networks,” Seminar Nasional Teknologi Informasi dan Komunikasi, 2017, hal. 79–86.
C. Bentes dan D. Velotto, “Ship Classification in TerraSAR-X Images With Convolutional Neural Networks,” IEEE Journal of Oceanic Engineering, Vol. 43, No. 1, hal. 258–266, 2018.
C. H. Lampert, M. B. Blaschko, dan T. Hofmann, “Beyond Sliding Windows : Object Localization by Efficient Subwindow Search,” 2008 IEEE Conference on Computer Vision and Pattern Recognition, 2008, hal. 1–8.
H. Xie, Q. Wu, B. Chen, Y. Chen, dan S. Hong, “Vehicle Detection in Open Parks Using a Convolutional Neural Network,” 2015 Sixth International Conference on Intelligent Systems Design and Engineering Applications (ISDEA), 2015, hal. 927–930.
F. Zhu, Y. Lu, N. Ying, dan G. Giakos, “Fast Vehicle Detection Based on Evolving Convolutional Neural Network,” 2017 IEEE International Conference on Imaging Systems and Techniques (IST), 2017, hal. 1–4.
P. Dollar, C. Wojek, B. Schiele, dan P. Perona, “Pedestrian Detection : A Benchmark,” IEEE Conference on Computer Vision and Pattern Recognition, 2009. CVPR 2009, 2009, hal. 304–311.
N. Hortovanyi, (2017) “Vehicle Detection and Tracking,” [Online]. https://towardsdatascience.com/vehicle-detection-and-tracking-6665d6e1089b, tanggal akses: 10 Februari 2018.
J. Tompson, A. Jain, Y. Lecun, dan C. Bregler, “Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation,” NIPS’14 Proceedings of the 27th International Conference on Neural Information Processing Systems, 2014, Vol. 1, hal. 1799–1807.
F. Chollet, (2015) “Keras,” [Online], https://github.com/keras-team/keras, tanggal akses: 21 Oktober 2017.
X. Y. Chen, S. M. Xiang, C. L. Liu, dan C. H. Pan, “Vehicle Detection in Satellite Images by Hybrid Deep Convolutional Neural Networks,” IEEE Geoscience and Remote Sensing Letters, Vol. 11, No. 10, hal. 1797–1801, 2014.
M. Sokolova dan G. Lapalme, “A Systematic Analysis of Performance Measures for Classification Tasks,” Information Processing & Management, Vol. 45, No. 4, hal. 427–437, 2009.
© Jurnal Nasional Teknik Elektro dan Teknologi Informasi, under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License.
1.png)
