Traffic recognition
所属分类：人工智能/神经网络/深度学习
开发工具：Python
文件大小：43537KB
下载次数：6
上传日期：2018-06-20 20:25:53
上 传 者：
Zun
说明： 通过使用卷积神经网络来实现交通标志识别，其中使用到了tensorflow框架
(Using Convolution Neural Networks to Realize Traffic Sign Recognition)
文件列表:
Traffic recognition\CNN_predict.py (11572, 2018-05-29)
Traffic recognition\datasets\BelgiumTS\Testing\00000\00017_00000.ppm (34683, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00000\00017_00001.ppm (27649, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00000\00017_00002.ppm (31815, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00000\00021_00000.ppm (10444, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00000\00021_00001.ppm (10987, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00000\00021_00002.ppm (12073, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00000\GT-00000.csv (266, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00252_00000.ppm (17773, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00252_00001.ppm (18238, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00252_00002.ppm (20929, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00398_00000.ppm (139299, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00398_00001.ppm (46068, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00398_00002.ppm (74343, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00414_00000.ppm (26509, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00414_00001.ppm (17773, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00414_00002.ppm (13243, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00424_00000.ppm (48747, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00424_00001.ppm (31793, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\00424_00002.ppm (21415, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01983_00000.ppm (34665, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01983_00001.ppm (138039, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01983_00002.ppm (15745, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01987_00000.ppm (10975, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01987_00001.ppm (24748, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01987_00002.ppm (82911, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01988_00000.ppm (123783, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01988_00001.ppm (28442, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01988_00002.ppm (12661, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01991_00000.ppm (10273, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01991_00001.ppm (24265, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\01991_00002.ppm (109446, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\02220_00000.ppm (70683, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\02220_00001.ppm (12847, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\02220_00002.ppm (6085, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00001\GT-00001.csv (1035, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00002\00025_00000.ppm (12154, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00002\00025_00001.ppm (11191, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00002\00025_00002.ppm (10666, 2017-12-28)
Traffic recognition\datasets\BelgiumTS\Testing\00002\02520_00000.ppm (7510, 2017-12-28)
... ...
********************************************************
BelgiumTSC - The Belgium Traffic Sign Classification Benchmark
********************************************************
This archive contains the training set of the
"Belgium Traffic Sign Classification Benchmark".
This training set is a subset of the BelgiumTS - Belgium Traffic Signs database:
http://homes.esat.kuleuven.be/~rtimofte/
http://www.vision.ee.ethz.ch/~timofter/
and follows the structure used in GTSRB Challenge proposed for IJCNN 2011:
http://benchmark.ini.rub.de/
**********************************************
Archive content
**********************************************
This archive contains the following structure:
There is one directory for each of the 62 classes (00000 - 00061).
Each directory contains the corresponding training images and one
text file with annotations, eg. GT-00000.csv.
In total are 4591 images for training. On average for each physically distinct
traffic sign there are 3 images available.
**********************************************
Image format and naming
**********************************************
The images are PPM images (RGB color). Files are numbered in two parts:
XXXXX_YYYYY.ppm
The first part, XXXXX, represents the pole number from where is in the full
annotations of BelgiumTS database. All images of one class
with identical pole numbers originate from the same pole and more represent
one single physical traffic sign.
The second part, YYYYY, is a running number for the views where the traffic
sign is annotated. There is no temporal order
of the images, but they were where extracted in the order of the annotations on
the original pole from BelgiumTS.
**********************************************
Annotation format
**********************************************
The annotations are stored in CSV format (field separator
is ";" (semicolon) ). The annotations contain meta information
about the image and the class id.
In detail, the annotations provide the following fields:
Filename - Image file the following information applies to
Width, Height - Dimensions of the image
Roi.x1,Roi.y1,
Roi.x2,Roi.y2 - Location of the sign within the image
(Images contain a border around the actual sign
of 10 percent of the sign size, at least 5 pixel)
ClassId - The class of the traffic sign
**********************************************
How to cite?
**********************************************
@inproceedings{Timofte-BMVC-2011,
author = {Radu Timofte and Luc Van Gool},
title = {Sparse representation based projections},
booktitle = {British Machine Vision Conference},
year = {2011},
}
@article{Timofte-MVA-2011,
author = {Radu Timofte and Karel Zimmermann and Luc {Van Gool}},
title = {Multi-view Traffic Sign Detection, Recognition, and 3D Localisation},
journal = {Machine Vision and Applications},
year = {2011},
doi = {10.1007/s00138-011-0391-3},
}
**********************************************
Further information
**********************************************
For more information please visit the website at
http://homes.esat.kuleuven.be/~rtimofte/
http://www.vision.ee.ethz.ch/~timofter/
If you have any questions, do not hesitate to contact us
Radu.Timofte@esat.kuleuven.be
Radu.Timofte@vision.ee.ethz.ch
**********************************************
Centre for Processing Speech and Images (PSI/ESAT)
VISion in Industry, Communications, and Services (VISICS)
Katholieke Universiteit Leuven
Belgium
**********************************************
近期下载者：
相关文件：
收藏者：