Vehicle Color Recognition on an Urban Road by Feature Context - 车辆颜色识别数据集
http://cloud.eic.hust.edu.cn:8071/~pchen/project.html
Pan chen, Xiang bai and Wenyu Liu
Department of Electronics and Information Engineering, Huazhong University of Science and Technology
Huazhong University of Science and Technology,HUST:华中科技大学
pdf
http://cloud.eic.hust.edu.cn:8071/~pchen/06777550.pdf
dataset
http://cloud.eic.hust.edu.cn:8071/~pchen/color.rar
Abstract
Vehicle information recognition is a key component of Intelligent Transportation Systems (ITS). Color plays an important role in vehicle identification. As a vehicle has its inner structure, the main challenge of vehicle color recognition is to select the region of interest (ROI) for recognizing its dominant color. In this paper, we propose a method to implicitly select the ROI for color recognition. Preprocessing is performed to overcome the influence of image quality degradation. Then the ROI in vehicle images is selected by assigning the sub-regions with different weights which are learnt by a classifier trained on the vehicle images. We train the classifier by linear SVM for its efficiency and high precision. The experiments are extensively validated on both images and videos, which are collected on urban roads. The proposed method outperforms other competing color recognition methods.
intelligent transportation system,ITS:智能运输系统
key component:主要组成部份,关键组分
vehicle identification:车辆识别
inner structure:内部结构
region of interest,ROI:感兴趣区域
dominant ['dɒmɪnənt]:adj. 显性的,占优势的,支配的,统治的 n. 显性
degradation [,degrə'deɪʃ(ə)n]:n. 退化,降格,降级,堕落
implicitly [ɪm'plɪsɪtlɪ]:adv. 含蓄地,暗中地
Vehicle Color Dataset
Description
The Vehicle Color recognition Dataset contains 15601 vehicle images in eight colors, which are black, blue, cyan, gray, green, red, white and yellow. The images are taken in the frontal view captured by a high-definition camera with the resolution of 1920×1080 on the urban road. The collected data set is very challenging due to the noise caused by illumination variation, haze, and over exposure.
cyan ['saɪən]:n. 蓝绿色 adj. 蓝绿色的
haze [heɪz]:n. 阴霾,薄雾,疑惑 vt. 使变朦胧,使变糊涂 vi. 变朦胧,变糊涂
over exposure:过度曝光,过度照射
illumination variation:光照变化
How to use the dataset
If you use the vehicle color recognition dataset for testing your recognition algorithm you should try and make your results comparable to the results of others. We suggest to choose half of the images in each category to train a model. And use the other half images to test the recognition algorithm.
Papers
[1] Vehicle Color Recognition on an Urban Road by Feature Context. Pan chen, Xiang bai and Wenyu Liu. Intelligent Transportation Systems, IEEE Transactions on (TITS), 2014, Issue: 99, Page(s): 1-7 [pdf] [dataset]
KEY POINTS
Vehicle_Color_Recognition 数据集包含 15601 幅、8 种颜色的车辆图像。Vehicle_Color_Recognition 数据集中所有图像均来自道路监控拍摄得到的正面图像,同时图像采集环境变化较大 (光照、天气等),数据集中同时存在多种车型,例如卡车、轿车、公交车等。
Vehicle_Color_Dataset - 车辆颜色数据集
http://cloud.eic.hust.edu.cn:8071/~pchen/project.html
http://cloud.eic.hust.edu.cn:8071/~xbai/softwares.html