DIML Lane Detection Benchmark
Main

Lane Detection Benchmark


Number of Videos : 470
                   Visitors : 60
                   Downloads : 2

Abstract

  • Description: our database is built to evaluate lane detection performance under various driving scenarios
  • Capture device: OV10630 image sensor
  • Database size: 470 video sequences
  • Image size: 1280 × 800
  • Environment conditions: road type (highway, route, urban roads, tunnel),
                                                  capture time (day, sunset/sunrise, night),
                                                  weather conditions (clear, cloudy, rainy),
                                                  other conditions (lens flare, streetlamps, car lamps, white lamp, yellow lamp)
  • Since our database includes various situations such as traffic jam, pedestrians, and obstacles,
    it can be used to develop and evaluate various vision-based ADAS algorithms and lane detection algorithms.

Data Category

Example Category

DIML-LD-1 (Highway)
  • Sequence featuring a highway scenario.
  • Sub-conditions : capture time (day, sunset/sunrise, night),
                                  weather conditions (clear, cloudy, rainy),
                                  other conditions (lens flare, street lamps, car lamps)


DIML-LD-2 (Route)
  • Sequence featuring a route scenario.
  • Sub-conditions : capture time (day, sunset/sunrise, night),
                                  weather conditions (clear, cloudy, rainy),
                                  other conditions (lens flare, street lamps, car lamps)


DIML-LD-3 (Urban road)
  • Sequence featuring a urban road scenario.
  • Sub-conditions : capture time (day, sunset/sunrise, night),
                                  weather conditions (clear, cloudy, rainy),
                                  other conditions (lens flare, street lamps, car lamps)


DIML-LD-4 (Tunnel)
  • Sequence featuring a tunnel scenario.
  • Sub-conditions : other conditions (white lamp, yellow lamp)


Download

    You may download our dataset for non-commercial research and educational purposes only. In exchange, we require that you provide us your contact information. [Click here]

Citation

    When using this dataset in your research, you must cite one of the following:

    @ARTICLE{Yoo2013,
    author = {Hunjae Yoo and Ukil Yang and Kwanghoon Sohn},
    title = {Gradient-enhancing conversion for illumination-robust lane detection},
    booktitle = {IEEE Transactions on Intelligent Transportation Systems (TITS)},
    volume={14},
    number={3},
    pages={1083--1094},
    year = {2013}
    }

    @ARTICLE{Son2015,
    author = {Jongin Son and Hunjae Yoo and Sanghoon Kim and Kwanghoon Sohn},
    title = {Real-time illumination invariant lane detection for lane departure warning system},
    booktitle = {Expert Systems with Applications (ESWA)},
    volume={42},
    number={4},
    pages={1816--1824},
    year = {2015}
    }

Copyright

    All datasets and benchmarks on this page are copyright by us.
    This means that you must attribute the work in the manner specified by the authors.
    You may not use this work for commercial purposes.
    You may distribute the resulting work only under the same license.

DIML

· © 2018 DIML Lane Detection Benchmark Dataset · by J.Lee