EBVT Database

EBVT: EEE BUET Vehicular Traffic - Video Database

Introduction:

The EBVT video database has been established to facilitate studies on detection, classification and tracking of vehicles from video, especially for developing video-based intelligent transportation systems. The research community is invited to test their algorithms on the EBVT database. On this page, information can be found about the description of the database as well as how to download the database.

Using EBVT Database:

The video clips included in the EBVT database can be used, free of charge only for research and educational purposes. Copying, redistribution, and any unauthorized commercial use are strictly prohibited. The use of this database is restricted to those individuals or organizations that obtained the database directly from this website. Any researcher reporting results which use this database must acknowledge the EBVT database. It is required that the researchers do so by citing the publication:

N. C. Mithun, N. U. Rashid, and S. M. M. Rahman, ''Detection and classification of vehicles from video using multiple time-spatial images,'' IEEE Trans. Intelligent Transportation Systems, vol. 13, no. 3, pp. 1215-1225, 2012.

@Article{EBVT_Database,
author =   "N. C. Mithun and N. U. Rashid and S. M. M. Rahman",
title =    "Detection and classification of vehicles from video using multiple time-spatial images",
journal = "IEEE Trans. Intelligent Transportation Systems",
volume =   "13",
number = "3",
pages =    "1215-1225",
year = 2012}

In addition, we appreciate to hear about any publications that use the EBVT database. Feedbacks on the database and this website are also welcome. We are continuously updating the database.

Data Description:

The video clips for the EBVT database were captured at different locations of Dhaka, Bangladesh and Suwon, Republic of Korea, under various day, night and traffic conditions using a fixed camera of model Sony W-110. The total length of these video clips is more than hours in duration. Both the incoming and outgoing traffic flows are considered, with variations in the elevation of camera, so that different appearances of similar vehicles are captured in the video. Examples of four scenes from the captured video sequences having different elevations of camera are shown in Fig. 1. From this figure, it may be observed that scenes 1 and 2 represent outgoing traffic flows and scenes 3 and 4 represent incoming traffic.

Figure 1: Sample scenes from different video sequences.

The lengths of the roads and camera positions are such that video resolution may vary between 1/6 and 1/12 m/pixel. The traffic consists of vehicles of seven types, i.e., Type I (motorbike), Type II (rickshaw), Type III (auto-rickshaw), Type IV (car or station wagon), Type V (jeep or minivan), Type VI (covered-van), and Type VII (bus). Type I represents the vehicular class 2-Wheeler, Types II and III represent the class 3-Wheeler, Types IV and V represent the class 4-Wheeler, and Type VI and VII represent the class 6-Wheeler. The minimum and maximum speeds of the vehicles in traffic were approximately 15 and 60 km/h, respectively. Tables 1 and 2 present a brief overview of video clips in the EBVT database. The tables contain clip ID, frame size, frame rate, length, file-size of the traffic video in the two cities, namely, Suwon and Dhaka. A few remarks on the video clips are also provided. We define the Clip ID as Place_Period_SerialNo.FileFormat. Here, the Place is Dhaka/Suwon, Period is Day/Night, and the video file format is wmv/mpg.

Table 1: Details of video clips captured in Suwon

Clip ID	Frame Size	Frame Rate (fps)	Length(min:sec)	Remarks			File-Size
Clip ID	Frame Size	Frame Rate (fps)	Length(min:sec)	Traffic	Viewing Angle	Illumination	File-Size
Suwon_Day_001.wmv	320X240	30	02:41	Incoming	Top-View,Angle:	Sunny	9.98MB
Suwon_Day_002.wmv	320x240	30	13:21	Incoming	Top-View,Angle:	Sunny + Normal	49.3MB
Suwon_Day_003.wmv	320x240	30	20:49	Outgoing	Top-View, Angle:	Sunny	77.0MB
Suwon_Day_004.wmv	320x240	30	13:14	Outgoing	Top-View, Angle:	Sunny + Normal	48.9MB
Suwon_Day_005.wmv	176x144	20	01:54	Incoming	Top-View, Angle:	Sunny	1.47MB
Suwon_Day_006.wmv	320x240	30	00:18	Outgoing	Top-View, Angle:	Sunny	1.16MB
Suwon_Day_007.wmv	320x240	30	01:48	Outgoing	Top-View, Angle:	Sunny	6.67MB
Suwon_Night_001.wmv	320X240	30	05:52	Incoming	Top-View, Angle:	Low Street-Light	21.7MB
Suwon_Night_002.wmv	320x240	30	11:21	Incoming	Top-View, Angle:	Low Street-Light	42MB
Suwon_Night_003.wmv	320x240	30	05:59	Incoming	Top-View, Angle:	Low Street-Light	22.1MB
Suwon_Night_004.wmv	320x240	30	21:06	Outgoing	Top-View, Angle:	No Street-Light	46.1MB
Suwon_Night_005.wmv	320x240	30	00:23	Incoming	Top-View, Angle:	Low Street-Light	1.44MB
Suwon_Night_006.wmv	320x240	30	00:22	Incoming	Top-View, Angle:	Low Street-Light	1.42MB

Table 2: Details of video clips captured in Dhaka

Clip ID	Frame Size	Frame Rate (fps)	Length (min:sec)	Remarks			File-Size
Clip ID	Frame Size	Frame Rate (fps)	Length (min:sec)	Traffic	Viewing Angle	Illumination	File-Size
Dhaka_Day_001.wmv	320x240	30	09:59	Incoming	Top-View, Angle:	Normal	36.9MB
Dhaka_Day_002.MPG	320x240	25	09:59	Outgoing	Top-View, Angle:	Sunny	52.0MB
Dhaka_Day_003.wmv	176x144	25	00:45	Incoming	Top-View, Angle:	Cloudy	1.90MB
Dhaka_Day_004.MPG	320x240	25	02:24	Outgoing	Top-View, Angle:	Sunny	12.4MB
Dhaka_Day_005.MPG	320x240	25	00:40	Outgoing	Top-View, Angle:	Sunny	3.49MB
Dhaka_Day_006.MPG	320x240	25	01:33	Outgoing	Top-View, Angle:	Sunny	8.15MB
Dhaka_Day_007.MPG	320x240	25	01:39	Outgoing	Top-View, Angle:	Sunny	8.59MB
Dhaka_Day_008.MPG	208x160	20	00:30	Outgoing	Top-View, Angle:	Sunny	0.84MB
Dhaka_Day_009.MPG	320x240	25	00:08	Outgoing	Top-View, Angle:	Sunny	0.765MB
Dhaka_Day_010.MPG	320x240	25	03:27	Outgoing	Top-View, Angle:	Sunny	17.9MB
Dhaka_Day_011.MPG	320x240	25	07:23	Outgoing	Top-View, Angle:	Sunny	38.4MB
Dhaka_Day_012.MPG	320x240	25	05:38	Outgoing	Top-View, Angle:	Sunny	29.4MB
Dhaka_Day_013.wmv	320x240	30	09:59	Outgoing	Top-View, Angle:	Normal	36.9MB
Dhaka_Day_014.wmv	320x240	30	09:59	Outgoing	Top-View, Angle:	Cloudy+Sunny	36.9MB
Dhaka_Day_015.wmv	320x240	30	09:59	Outgoing	Top-View, Angle:	Sunny+Cloudy	36.9MB
Dhaka_Day_016.wmv	320x240	30	09:59	Outgoing	Top-View, Angle:	Sunny+Cloudy	36.9MB
Dhaka_Day_017.wmv	320x240	30	09:59	Outgoing	Top-View, Angle:	Cloudy+Sunny	36.9MB
Dhaka_Day_018.wmv	320x240	30	06:25	Outgoing	Top-View, Angle:	Cloudy	23.7MB
Dhaka_Day_019.wmv	320x240	30	06:25	Outgoing	Top-View, Angle:	Cloudy+Sunny	23.7MB
Dhaka_Day_020.wmv	320x240	30	03:34	Outgoing	Top-View, Angle:	Cloudy+Sunny	13.2MB
Dhaka_Day_021.wmv	320x240	30	00:40	Outgoing	Top-View, Angle:	Normal+Sunny	2.53MB
Dhaka_Day_022.wmv	320x240	25	00:51	Outgoing	Side-View	Normal	2.22MB
Dhaka_Day_023.wmv	320x240	30	00:05	Outgoing	Side-View	Normal	0.399MB
Dhaka_Day_024.wmv	320x240	30	00:25	Outgoing	Side-View	Normal	1.70MB

Download:

The clips captured in Suwon during day- and night-times are available in EBVT_Suwon_Day (193 MB) and EBVT_Suwon_Night (133 MB), respectively. The clips obtained from Dhaka are available in EBVT_Dhaka_BUET (344 MB) and EBVT_Dhaka_Shahbag (103 MB). These clips are available in password protected ZIP files. The password can be requested to the author of this website by writing in mahbubur@eee.buet.ac.bd or mahbuburbd@gmail.com.