Real-time rear obstacle detection using reliable disparity for driver assistance

Hunjae Yoo, Jongin Son, Bumsub Ham, Kwanghoon Sohn

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

A vision based real-time rear obstacle detection system is one of the most essential technologies, which can be used in many applications such as a parking assistance systems and intelligent vehicles. Although disparity is a useful feature for detecting obstacles, estimating a correct disparity map is a hard problem due to the matching ambiguity and noise sensitivity, especially in homogeneous regions. To overcome these problems, we leverage reliable disparities only for obstacle detection. A reliability factor is introduced to measure an inhomogeneity of the regions quantitatively. It is computed at each superpixel to consider the noise sensitivity of pixel-wise gradients and to assign similar reliability value within a same object. It includes two major components: firstly, In a feature extraction and combining stage, we extract three features from stereo images such as disparity, superpixel segments and pixel-wise gradient and compute the reliability of disparity from superpixel segments and the pixel-wise gradient. Secondly, In an obstacle detection stage, a disparity feature with reliability votes for localizing obstacles and dominant candidates in voting map are selected as initial obstacle region. The initial obstacle regions are expanded into their neighbor superpixels based on CIELAB color similarity and distance similarity between superpixels. Experimental results show satisfactory performance under various real parking environments. Its detection rate is at least 4% higher than those of other existing methods, and its false detection rate is more than 10% lower and thus, can be used for parking assistance system.

Original languageEnglish
Pages (from-to)186-196
Number of pages11
JournalExpert Systems with Applications
Volume56
DOIs
Publication statusPublished - 2016 Sep 1

Fingerprint

Parking
Pixels
Intelligent vehicle highway systems
Feature extraction
Color

All Science Journal Classification (ASJC) codes

  • Engineering(all)
  • Computer Science Applications
  • Artificial Intelligence

Cite this

@article{65bb7fbe7ba6484aac2342d5899ff2ad,
title = "Real-time rear obstacle detection using reliable disparity for driver assistance",
abstract = "A vision based real-time rear obstacle detection system is one of the most essential technologies, which can be used in many applications such as a parking assistance systems and intelligent vehicles. Although disparity is a useful feature for detecting obstacles, estimating a correct disparity map is a hard problem due to the matching ambiguity and noise sensitivity, especially in homogeneous regions. To overcome these problems, we leverage reliable disparities only for obstacle detection. A reliability factor is introduced to measure an inhomogeneity of the regions quantitatively. It is computed at each superpixel to consider the noise sensitivity of pixel-wise gradients and to assign similar reliability value within a same object. It includes two major components: firstly, In a feature extraction and combining stage, we extract three features from stereo images such as disparity, superpixel segments and pixel-wise gradient and compute the reliability of disparity from superpixel segments and the pixel-wise gradient. Secondly, In an obstacle detection stage, a disparity feature with reliability votes for localizing obstacles and dominant candidates in voting map are selected as initial obstacle region. The initial obstacle regions are expanded into their neighbor superpixels based on CIELAB color similarity and distance similarity between superpixels. Experimental results show satisfactory performance under various real parking environments. Its detection rate is at least 4{\%} higher than those of other existing methods, and its false detection rate is more than 10{\%} lower and thus, can be used for parking assistance system.",
author = "Hunjae Yoo and Jongin Son and Bumsub Ham and Kwanghoon Sohn",
year = "2016",
month = "9",
day = "1",
doi = "10.1016/j.eswa.2016.02.049",
language = "English",
volume = "56",
pages = "186--196",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",

}

Real-time rear obstacle detection using reliable disparity for driver assistance. / Yoo, Hunjae; Son, Jongin; Ham, Bumsub; Sohn, Kwanghoon.

In: Expert Systems with Applications, Vol. 56, 01.09.2016, p. 186-196.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Real-time rear obstacle detection using reliable disparity for driver assistance

AU - Yoo, Hunjae

AU - Son, Jongin

AU - Ham, Bumsub

AU - Sohn, Kwanghoon

PY - 2016/9/1

Y1 - 2016/9/1

N2 - A vision based real-time rear obstacle detection system is one of the most essential technologies, which can be used in many applications such as a parking assistance systems and intelligent vehicles. Although disparity is a useful feature for detecting obstacles, estimating a correct disparity map is a hard problem due to the matching ambiguity and noise sensitivity, especially in homogeneous regions. To overcome these problems, we leverage reliable disparities only for obstacle detection. A reliability factor is introduced to measure an inhomogeneity of the regions quantitatively. It is computed at each superpixel to consider the noise sensitivity of pixel-wise gradients and to assign similar reliability value within a same object. It includes two major components: firstly, In a feature extraction and combining stage, we extract three features from stereo images such as disparity, superpixel segments and pixel-wise gradient and compute the reliability of disparity from superpixel segments and the pixel-wise gradient. Secondly, In an obstacle detection stage, a disparity feature with reliability votes for localizing obstacles and dominant candidates in voting map are selected as initial obstacle region. The initial obstacle regions are expanded into their neighbor superpixels based on CIELAB color similarity and distance similarity between superpixels. Experimental results show satisfactory performance under various real parking environments. Its detection rate is at least 4% higher than those of other existing methods, and its false detection rate is more than 10% lower and thus, can be used for parking assistance system.

AB - A vision based real-time rear obstacle detection system is one of the most essential technologies, which can be used in many applications such as a parking assistance systems and intelligent vehicles. Although disparity is a useful feature for detecting obstacles, estimating a correct disparity map is a hard problem due to the matching ambiguity and noise sensitivity, especially in homogeneous regions. To overcome these problems, we leverage reliable disparities only for obstacle detection. A reliability factor is introduced to measure an inhomogeneity of the regions quantitatively. It is computed at each superpixel to consider the noise sensitivity of pixel-wise gradients and to assign similar reliability value within a same object. It includes two major components: firstly, In a feature extraction and combining stage, we extract three features from stereo images such as disparity, superpixel segments and pixel-wise gradient and compute the reliability of disparity from superpixel segments and the pixel-wise gradient. Secondly, In an obstacle detection stage, a disparity feature with reliability votes for localizing obstacles and dominant candidates in voting map are selected as initial obstacle region. The initial obstacle regions are expanded into their neighbor superpixels based on CIELAB color similarity and distance similarity between superpixels. Experimental results show satisfactory performance under various real parking environments. Its detection rate is at least 4% higher than those of other existing methods, and its false detection rate is more than 10% lower and thus, can be used for parking assistance system.

UR - http://www.scopus.com/inward/record.url?scp=84964049714&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964049714&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2016.02.049

DO - 10.1016/j.eswa.2016.02.049

M3 - Article

AN - SCOPUS:84964049714

VL - 56

SP - 186

EP - 196

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -