現今無人車駕駛技術中，車輛偵測是一個重要的環節，無論是白天還是夜
晚，都必須能有效偵測。然而目前的無人車，多是依據RGB 影像偵測為主，到
了夜晚RGB 影像容易不清晰。為了解決這個問題，我們提出一個即時車輛檢測
的方法，主要是利用熱影像在夜間進行偵測，因為熱影像即使在昏暗的環境中
仍可保留車輛的特徵和細節。為了能有效地檢測，我們在黃昏和夜間的時段，收
集足夠的熱影像資料集供以訓練，隨後使用Thermal Feature Enhancement (TFE)
來增強影像的對比度和銳化，接著將增強後的影像串接作為模型的輸入，使模
型能更有效地學習特徵。此外，我們改進了一個捲積網路模型用於車輛檢測，
命名為Thermal Image Only Looked Once(TOLO) 模型。然而車輛在行進時容易抖
動，導致拍攝影像模糊，所偵測的車輛候選框的機率降低。為此我們提出了Low
Probability Candidate Filter(LPCF) 的方法來克服這個問題。總結和現有的方法相
比，我們所提出的方法在F1 測量法中，有更良好的表現。

In today’s world, it becomes critical for a self-driving car to detect the vehicles irrespective
of it being a day or night. During the night time, the RGB images captured by the
cameras in the self-driving cars are not clear. Further, to overcome this issue, we propose
a real-time vehicle detection using a sequence of night-time thermal images. Moreover,
the thermal images have the capability of retaining even the minuscule vehicle details
in a dim environment. For an efficient vehicle detection, the thermal image dataset collected
during the dusk and night is used for training purposes. Subsequently, the contrast
enhancement and sharpening of these images are performed using the Thermal Feature
Enhancement (TFE). Then the concatenated images are supplied as the input to allow the
model to learn more effectively. Besides, we also propose an improved convolution network
model entitled as the Thermal Image Only Looked Once (TOLO) model for vehicle
detection. Additionally, juddering of the moving vehicle results in blurred images that are
referred to as low probability candidates. Also, we propose a method called as Low Probability
Candidate Filter (LPCF) to overcome this problem. Our proposed method produces
better results for the F1-measure in comparison with existing methods.

[1] J. Redmon and A. Farhadi, “YOLO9000: better, faster, stronger,” CoRR, vol. abs/
1612.08242, 2016.
[2] J. Redmon and A. Farhadi, “Yolov3: An incremental improvement,” CoRR, vol. abs/
1804.02767, 2018.
[3] A. Yoneyama, C.-H. Yeh, and C.-C. J. Kuo, “Robust vehicle and traffic information
extraction for highway surveillance,” EURASIP Journal on Advances in Signal
Processing, vol. 2005, p. 912501, Aug 2005.
[4] S. Ren, K. He, R. B. Girshick, and J. Sun, “Faster R-CNN: towards real-time object
detection with region proposal networks,” CoRR, vol. abs/1506.01497, 2015.
[5] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. E. Reed, C. Fu, and A. C. Berg,
“SSD: single shot multibox detector,” CoRR, vol. abs/1512.02325, 2015.
[6] Y. Iwasaki, M. Misumi, and T. Nakamiya, “Robust vehicle detection under various
environments to realize road traffic flow surveillance using an infrared thermal camera,”
in TheScientificWorldJournal, 2015.
[7] A. Yoneyama, C. H. Yeh, and C. C. J. Kuo, “Moving cast shadow elimination for
robust vehicle extraction based on 2d joint vehicle/shadow models,” in Proceedings
of the IEEE Conference on Advanced Video and Signal Based Surveillance, 2003.,
pp. 229–236, July 2003.
[8] M. E. Irhebhude, M. A. Ali, and E. A. Edirisinghe, “Pedestrian detection and vehicle
type recognition using centrog features for nighttime thermal images,” in 2015 IEEE
International Conference on Intelligent Computer Communication and Processing
(ICCP), pp. 407–412, Sept 2015.
[9] M. E. Irhebhude, P. O. Odion, and D. T. Chinyio, “Centrog feature technique for
vehicle type recognition at day and night times,” CoRR, vol. abs/1612.00645, 2016.
[10] Y. Iwasaki, M. Misumi, and T. Nakamiya, “Robust vehicle detection under various
environmental conditions using an infrared thermal camera and its application to
road traffic flow monitoring,” in Sensors, 2013.
[11] R. Abbott, J. Del Rincon, B. Connor, and N. Robertson, “Deep object classification
in low resolution lwir imagery via transfer learning,” in Proceedings of the 5th IMA
Conference on Mathematics in Defence, 11 2017.
[12] J. Zhang, M. Huang, X. Jin, and X. Li, “A real-time chinese traffic sign detection
algorithm based on modified yolov2,” Algorithms, vol. 10, no. 4, 2017.
[13] K. Zuiderveld, “Graphics gems iv,” ch. Contrast Limited Adaptive Histogram Equalization,
San Diego, CA, USA: Academic Press Professional, Inc., 1994.
[14] G. RC and R. E. Woods, Digital Image Processing: Introduction. Prentice-Hall,
2007.
[15] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale
image recognition,” CoRR, vol. abs/1409.1556, 2014.
[16] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep
convolutional neural networks,” Curran Associates Inc., 2012.
[17] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke,
and A. Rabinovich, “Going deeper with convolutions,” in 2015 IEEE Conference
on Computer Vision and Pattern Recognition (CVPR), pp. 1–9, June 2015.
[18] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,”
in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
pp. 770–778, June 2016.
[19] M. D. Zeiler and R. Fergus, “Visualizing and understanding convolutional networks,”
in Computer Vision – ECCV 2014 (D. Fleet, T. Pajdla, B. Schiele, and
T. Tuytelaars, eds.), (Cham), pp. 818–833, Springer International Publishing, 2014.
[20] M. Lin, Q. Chen, and S. Yan, “Network In Network,” In Proc. ICLR, 2014.
[21] J. Bian, W. Y. Lin, Y. Matsushita, S. K. Yeung, T. D. Nguyen, and M. M. Cheng,
“Gms: Grid-based motion statistics for fast, ultra-robust feature correspondence,”
in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
pp. 2828–2837, July 2017.