Open Journal of Applied Sciences, 2013, 3, 68-74
Published Online March 2013 (
Copyright © 2013 SciRes. OJAppS
Robust Lane Detection in Shadows and Low Illumination
Conditions using Local Gradient Features
Avishek Parajuli, Mehmet Celenk, H. Bryan Riley
School of Electrical Engineering and Computer Science Stocker Center, Ohio University Athens, Ohio USA
Email: {ap356311, celenk, rileyh1}
Received 2012
This paper presents a method for lane boundaries detection which is not affected by the shadows, illumination and un-
even road conditions. This method is based upon processing grayscale images using local gradient features, characteris-
tic spectrum of lanes, and linear prediction. Firstly, points on the adjacent right and left lane are recognized using the
local gradient descriptors. A simple linear prediction model is deployed to predict the direction of lane markers. The
contribution of this paper is the use of vertical gradient image without converting into binary image(using suitable thre-
shold), and introduction of characteristic lane gradient spectrum within the local window to locate the preciselane
marking points along the horizontal scan line over the image. Experimental results show that this method has greater
tolerance to shadows and low illumination conditions. A comparison is drawn between this method and recent methods
reported in the literature.
Keywords: Local Gradient Features; Lane Detection; Linear Prediction; Characteristic Spectrum
1. Introduction
Lane detection has long been an important area of re-
search for people in autonomous vehicle systems. There
are a large number of lane detection algorithms that are
available in the literature during recent years [1-8]. The
lane detection algorithm can be classified in two broad
categories, namely model based [8,13-18] and feature
based [1,7,11,12].The Model-based lane detection system
begins with selecting a model/template for lane, extract-
ing parameters to represent lanes, and fitting the model to
the extracted parameters and processing. Applications of
these template based methods (linear, spline, parabola)
are limited to some road environments and cannot per-
form well in the presence of shadows and low illumina-
tion condition. The algorithm proposed by Yue et al. [15]
used the cubic B- sna ke to model the lane boundaries.
The use of Canny Hough Estimation of Vanishing Points
(CHEVP) algorithm for locating the initial points is sen-
sitive to choice of threshold. Moreover, this method is
vulnerable to presence of shadows and illumination con-
The most common and simple technique for detecting
road markings is based on Image gradient or edges (fea-
ture based). McDonald and Palmer et al. presented me-
thods based on detecting edges and applying Hough
Transform in [11], [12] respectively. However these edge
methods are sensitive to illumination, heavily affected by
shadows, environmental conditions. Additionally, these
methods have experienced difficulties when choosing a
suitable intensity threshold to remove the unwanted
edges. In [6], a layered approach is presented for robust
lane detection at night which relies on temporal frame
averaging, edge detection and Hough transform. The
algorithm is very fast, robust, and can be used in real
time applications. However, it is susceptible to lighting,
shadows, and environment conditions and focuses only
on straight road segments.
The authors in [3] p re sented a method which combines
input from multiple cameras (data fusion) and line mod-
els are formed from feature points using a RANSAC al-
gorithm. The approach is fast and robust; however, it can
support real time operation only at low speeds, and is
heavily affected by the presence of cars and other ob-
stacles on the road, resulting in generation of false fea-
ture points. There have been efforts by C. Rasmussen in
[4] to use Principal Component Analysis (PCA) for lane
detection where they use gradient within a local window
to obtain the vanishing point, but the directions obtained
from gradients of PCA may not converge all the time.
The algorithm in [19, 22] presented the Likelihood of
Image Shape (LOIS) lane detection system that applies
likelihood function into lane marker detections. Lane
finding in ANother domAin (LANA) [20] uses sets of
frequency domain features to capture information con-
cerning the strength and orientation of spatial edges.
LANA is similar to LOIS in detection stages but differs
only in the use of frequency features of lanes while LOIS
Copyright © 2013 SciRes. OJAppS
uses spatial edge features. The algorithm presented here-
in, also uses the frequency features of the lane markings
that are a characteristic representation of the lane.
Sivaraman et al. presented a method in [15] for lane
detection and tracking in high density traffic scenes using
vehicle localization where lane detection and tracking is
performed by using steerable filters and Kalman filters.
The algorithm in [15]performs well only in high density
traffic scenes but its performance degrades highly in an
empty road. Also good performance of this method in
presence of shadows is doubtable. The research pre-
sented in [10] also fails to perform well in presence of
shado ws.
A survey of the recent lane detection method s are pre-
sented in [9, 10]. There has been progress in detection of
lanes of various shapes but there is a need for robust de-
tection of lanes in presence of shadows and other image
artifacts. The work presented in [7] performs well in
presence of shadows but cannot track the curved sections
of the road in the far field of the image. The work pre-
sented here is compared with the work[7] using the same
datasets [23].The remaining sections of this paper are
organized as follows. Section 2 contains the description
of method and path chosen. Section 3 presents the expe-
rimental results. The conclusion and suggestions for fu-
ture work are presented in section 4.
2. Description of the method
The proposed method begins with extracting the indi-
vidual frames from the video and processing each frame
to detect and track the road lane markers. Then the ver-
tical gradient of the image is calculated and processing is
done in this vertical gradient image. The vertical gradient
is selected to remove the effects of shadows along the
road which are usually horizontal.Additionally, this me-
thod does not require threshold to convert the gradient
image into binary image and is thus unaffected by illu-
mination conditions and shadows.
2.1. Feature extraction
The vertical gradient of an image [21]is obtained by 2D
discrete convolution of an image I (m, n) with h (m, n)
( )( )
,,*( ,)G mnImnhmn=
whe re h (m, n) is a mask given by
( )
, 000
h mn
In the spatial domain, a window of size 15x15 is selected
and placed at randomly selected points s1, s2, s3, s4, and
s5. All the points within these five local windows are
placed in five different matrices namely w1,w2, w3, w4
and w5. Then the magnitude of FT (Fourier Transform)
of these five matrices is computed and their average is
calculated. This average value of the magnitude of FT is
the characteristic spectra for the lane markings and will
be represented as Lspectra. The variable Lspectra used
further in this discussion will refer to the characteristic
spectra of the lane markers. This process of obtaining
characteristic spectra for a point ‘s1’ is shown in Fig. 2.
Figure 1: Characteristic spectra of the left lane markers at point s1, s2,
s3, s4, s5 and the average spectra.
Figure 2: Graphical representation for the process of obtaining charac-
teristic spectra.
Analysis of result for different window dimension indi-
cated the 15x15 size window produced most consistent
and accurate representation of distinguishing road lane
markers from background settings.The spectra for the
lane markings are characteristic to the respective lane
and invariant to the selection of location on a given lane
marking (independent of translation, rotation and scal-
Copyright © 2013 SciRes. OJAppS
ing).The characteristic spectrum is obtained from the first
frame of the video file from the image database[23]. This
spectrum (Lspectra) is consistent for all the images of the
video file from the database. Once the characteristic
spectrum of lane marking is obtained from the first frame
of the, we do not require to obtain the characteristic
spectra from all the successive frames for a period of
2.2. Gradient spectrum match function
The gradient spectrum match function is used to locate
the precise points on the adjacent lanes in which the ve-
hicle is moving. The 15x15 local window is moved along
the horizontal scan line (see Figure 3) and magnitude of
FT over the local window is calculated as represented in
Equations 2-4. Let
( )
be the grad i e nt image
(see Figure 3) with the following dimensions.
( )
are discrete variables in the range
1 ,1nR nC≤≤ ≤≤
The sliding window is placed at each point along the
horizontal scan line (green colored line in Figure 3)and
corresponding gradient values of image I
( )
the window can be represented by variable p as in the
Equation 3. Along the horizontal line the row component
of the image i.e.
( )( )
yp iy i=
is fixed and only the
changes along the scan line from 1 to C.
( )
77, 77kl−≤ ≤−≤≤
{ }(
{ ,}PFTpFTIn knl= =++
Then the variable
{ }
PFT p=
generated at a given
( )
,nn₁₂ is correlated with the characteristic spec-
trum Lspectra and the maximum value is chosen. This
maximum value is R (0, 0) and can be obtained by ele-
ment wise multiplication of Lspectra and P as shown in
Equation 5. The maximum correlation is calculated at
each point along the horizontal scan line (refer to Figure
3) and stored in a variable Sm (
) where
= 1,
The normalized plot of Sm for all the values of
along the horizontal line is called Spectrum matching
plot (Figure 3).
R(0, 0) =
( )( )
iLspectrai jPi j
Sm(n2) =R (0, 0), where
= 1, 2, 3 …C (6)
2.3. Processing of spectrum match (Sm) plot
The spectrum matching plot (Sm) is smoothed with a
Moving Average (MA) filter of order 20, and the max-
ima points on the plot are found using the algorithm
in[24].To remove the unnecessary processing of maxima
Figure 3: Spectrum matching (Sm) plot generated along the horizontal
scan line on the gradient image. Filtered version of spectrum matching
plot with green points representing maxima points in extreme right.
points, the Sm plot is threshold by a suitable value of
0.3(if Sm(i) < 0.3, thenSm(i) = 0). This choice of this
threshold value is not critical and helps only in the re-
moval of unnecessary maxima points.
2.4. Local Maxima Discrimination Via Bayesian
The local maxima points obtained in the Sm plot is de-
cided as either belonging to the left or right. Bayesian
classifier is devised to discriminate the maxima points
into one of the two designated classes, namely, left line
class(C L), and right line class(CR)[22]. The maxima
points which lies on the left of the median line (line
passing through the median of column number) is de-
cided as belonging to left lane class CL, and the maxima
points to the right of median line as designated as right
lane class CR. Let X be the vector containing all the
column number of maxima points and =median(C),
where C is the Column of the image shown in Figure 3.
If Xi then Xi ɛ CL (7)
If Xi then Xi ɛ CR (8)
Copyright © 2013 SciRes. OJAppS
where i =1, 2 …N; N=total number of local maxima.
The Figure 3 shows the first horizontal scan line which
is set to start from at 3/4th of the length of the image
(0.75R). The second scan line will start at 5 pixel points
above the first line and so on. The first scan line has only
one maxima points at x= 60 (N=1). This point is classi-
fied as CL, because Equation. 7 is satisfied. The median
for all the images obtained from the database [23] is con-
stant ( =256/2 + 1 = 129).
Figure 4 shows an instance in which three scan line,
separated by 50 pixels, are produced and the spectrum
matching plot Sm is generated for each of the three hori-
zontal lines. The Sm plot from scan line 1 has only one
maxima point that belongs to left class (CL) as that points
lies to the left of medina line. The scan line 2 intersects
with both the adjacent left lane and right lane markers
and it is reflected in the corresponding Sm plot with two
ma xima points on the either side of median line. The Sm
plot for the scan line 3 has four maxima points with the
left one belonging to the class CL and the other three be-
longs to class CR. In the near field of the image, the Sm
plot along the scan line produces less maxima points. As
the scan line is mover further into the far field the num-
ber of maxima points increase. The maxima points are
classified into one of the two classes.
For all the scan lines separated by a distance of 5
pixels, the left and right lane points are selected as max-
imum values in respective class; i.e,
LT=max (CL) (9)
RT=max (CR) (10)
Due to the effect of filtering, the maxima points are
shifted away from their true location in the Sm plot.
Therefore the precise locations of the lane markers are
found by searching the pixel with maximum intensity in
the neighborhood of the points obtained from Equation 9
and Equation 10 along the respective horizontal line. It
was found that searching for the maximum intensity in
the spatial domain of the gray scale image produced bet-
ter results than searching in the vertical gradient of the
image. Mathematically, it is represented in Equation 11
and 12.
LT= max{img(row,LT-20:LT+20)} (11)
RT= ma x{img(r ow,R T -20:RT+20)} (12)
where row is the row number for a given scan line, and
img (row, column) is the original gray scale image.
After the first five left and right lane points are deter-
mined, next left and right lane points are predicted based
on the last five left and right points, respectively. The
prediction scheme is described in the next section and is
useful in removing the false lane points obtained from
the maxima points of Sm plot.
Figure 4: Image gradient with three scan lines and their respective Sm
plots with maxima points. The maxima points are represented by green
dots on the lower portion of Figure and the median line by yellow and
red line.The distance between the scan lines is set to 50 pixels in this
Figure for the purpose of simplifying the explanation of process. The
distance between the scan lines in the actual algorithm is set to 5. Red
line is the median line.
2.5. Linear prediction of lane points
After the first five horizontal scan lines are drawn, the
left and right lane points on the next scan line are esti-
mated using a simple linear prediction model. Since the
distance between the scan lines are predetermined and
fixed in advance, the row number for each horizontal line
is known. Let y1, y2, y3, y4, and y5 be the column num-
ber of five lane marking points. The 1D displacement
( )
, between the five points is calculated and their
mean value is computed. The next predicted point is then
calculated by adding the mean displacement of last five
to the most recent point y5. The process described here is
presented in Equations 13 -16.
Predicted points,
( )( )
yp iy i=
; i=1,2,… 5s (13)
( )()( )
1;diyp iyp i= +−
where i = 1, 2, 3, 4.
Mean displacement,
( )
Predicted point,
+ D (16)
2.6. Error Correction
Copyright © 2013 SciRes. OJAppS
The estimated points
yp i for the next scan line are
compared with the calculated lane points,
( )
of the
next scan line for both left and right lane points. If the
computed point
( )
deviates very far away from the
predicted point
yp i an error has occurred in the
computation of the actual lane marking points. In such
cases the predicted point is selected as correct point. All
these steps are shown in Equation 17.
If abs {}> threshold, then
( )( )
y iyp i=
( )( )
yp iy i=
where the empirically determined threshold is set to 20.
Figure 5 shows the calculated points and their predicted
values. The linearly predicted points after error correc-
tion yields highly accurate representation of the precise
location of the lane markings.
A 2D graph showing the comparison of computed lane
points with the predicted and corrected points is shown in
Figure 6. The processing steps outlined through sections
2.2 through 2.6 are performed for an image until both the
left and right lane marking point falls to the same class.
That is, algorithm stops when both the lane points lies on
the same side of the median line(same class).
Figure 5. Computed lane markings(blue colored box), and
predicted markings(red color dots) plot on the vertical gra-
dient image and RGB image. Note the presence of shadows.
Figure 6: Comparison of the computed and predicted &
corrected lane marking points as a function of row and
column number.
3. Results
This section, presents the results obtained from the
process described in section 2. The various stages of
feature extraction, generation of spectrum matching plot,
discr i mination via Bayesian learning, and linear
prediction of lane points are implemented using a laptop
computer with a clock rate of 2.24 GHz executing
Matlab software version R2008a. The data sets of images
used for the testing and validation of the proposed
method is same as that has been used in [cmu].
( The
157 image sequence was tested with our algorithm and it
was found that 150 images were successfully detected
with 7 false detections. The succesfully detection rate
was found to be 95.54 % which are better than the
detection rate achieved in [7,15].
Some of the results of lane marker detection in pres-
ence of shadows are presented in Figure 7-8. The pres-
ence of error correction scheme in the algorithm rectifies
the large deviations from the lane markers position as the
lane markers do not change its course abruptly. An im-
portant advantage/contribution of this method is to track
roads lane markers of various shape (curved or straight)
and locate precise lane marking points on each horizontal
scan line which is not affected by presence of shadows
and other low illumination condition. This method has a
disadvantage in that the first five scan line needs to lo-
cate almost precise points because the error in the first
five scan lines cannot be corrected. It was found that any
high dynamic range portion of image leading to errors in
the first five scan lines cannot be rectified.
Copyright © 2013 SciRes. OJAppS
Figure 7: Experimental results for some of the extreme cas-
es of illumination and road conditions. Note that even the-
presence of shadows and vehicle on the lane markers has
little to no affect the detection results.
Figure 8: Some additional experimental results.
Copyright © 2013 SciRes. OJAppS
4. Conclusion
This paper has introduced an approach for the detection
of road lane markers based on local gradient features,
characteristic lane spectrum, and spectrum match plot
(Sm). Further the linear prediction and error correction
schemes leads to robust detection of lane marking points
along the scan line. The experimental result suggest that
this method is robust for lane detection in presence of
shadows and other vehicles.
The future extension of this work can be information
fusion of this method and other methods (e.g. Hough
transform)for higher detection rate in other inclement
weather conditions such as rain, snow, fog. Neverthe-
less, there is always room for improvement.
[1] Y. Wang, N. Dahnoun, and A. Achim, “A novel system
for robust lane detection and tracking,” Signal Processing,
vol. 92, no. 2, pp. 319334, 2012. View at
Publisher · View at Google Scholar.
[2] Z. Kim, “Robust lane detection and tracking in challeng-
ing scenarios,” IEEE Transactions on Intelligent Trans-
portation Systems, vol. 9, no. 1, pp. 1626, 2008.view.
[3] C. Lipski “A fast and robust approach to lane mark-
ing detection and lane tracking” in proceedings of IEEE
SSIAI, 24-26 march 2008.view
[4] C. Rasmussen,”Groupingdominat structure of Ill struc-
tured Road Following” in proceedings of IEEE computer
society, CVPR 2004 27 June-2 July 2004, pages I-470 -
I-477 Vol.1
[5] S. Sivaraman and M. M. Trivedi “Improved vision-based
lane tracker performance using vehicle localization” in
proceedings of IEEE intelligent vehicles symposium,
June 21-24, 2010. view
[6] A. Borkar “A layered approach to robust lane
detection at night” in proceedings of IEEE CIVVS,
March 30-April 2 , 2009
[7] Y. Fan, W. Zhang, X. Li, L. Zhang, and Z. Cheng “A
robust lane boundaries detection algorithm based on gra-
dient distribution features,” Proceedings of the 8th Inter-
national conference on Fuzzy Systems and Knowledge
Discovery (FSKD), pp 1714-1718, July 2011.
[8] J. Wang, F. Gu, C. Zhang, and G. Zhang “Lane boundary
detection based on parabola method,” Proceedings of the
2010 IEEE, International conference on Information and
Automation, pp 1729-1734, china, June 2010.view
[9] J. C. McCall and M. M. Trivedi, “Video-based lane esti-
mation and tracking for driver assistance: survey, system,
and evaluation,” IEEE Transactions on Intelligent Trans-
portation Systems, vol. 7, no. 1, pp. 2037, 2006.
[10] A. Bar Hillel, R. Lerner, D. Levi, and G. Raz, “Recent
progress in road and lane detection: a survey,” Machine
Vision and Applications. In press. View at
r · View at Google Scholar
[11] J McDonald. “Detecting and tracking road markings us-
ing the Hough transform,” Proc. Of the Irish Machine Vi-
sion and Image Processing Conference 2001.
[12] P. L. Palmer, J. Kittler and M. Petrou, “An optimizing
line finder using a Hough transform algorithm,” Comput-
er Vision and Image Understanding, vol. 68, no 1, pp.
1-23, July 1993.
[13] K. Kluge and S. Lakshmanan, “A deformable-template
approach to lane detection,” Proceedings of the Intelligent
Vehicles '95 Symposium, pp. 54-59, September 1995
[14] Y. Wang, et al.,” Lane detection using spline model,”
Pattern Recognition Letters, vol 21, pp.677-689, 2000.
[15] Y. Wang, et al.,” Lane detection and tracking using
B-Snake,” Image and Vision Computing, vol. 22,
pp.269-280, 2004.
[16] J. W. Park, et al., “A lane-curve detection based on LCF,”
Pattern Recognition Letters, vol 24, pp.2301-2313, 2003.
[17] Dong-Joongkang and Mun-Ho Jung,”Road lane segmen-
tation using dynamic programming for active safety ve-
hicles,” Pattern Recognition Letters, vol 24, pp
3177-3185, July 2003.
[18] S.P.Liou, R.C. Jain, “Road following using vanishing
points,“ Computer Vision, Graphics, and Image
Processing 39 (1987) 116-130.
[19] S. Lakshmanan and K. Kluge, “LOIS: A real-time lane
detection algorithm,” in Proceedings 30th Annual Confe-
rence of Information Science Systems, 1996,
[20] C. Kreucher and S. Lakshmanan, “LANA: A lane extrac-
tion algorithm that uses frequency domain features”,
IEEE Transactions on Robotics and Automation, vol. 15,
no. 2, pp.343-350, April 1999.
[21] J.S. Lim, Two Dimensional Signal and Image Processing,
Prentice Hall 1990.
[22] S. Theodoridis and K. Koutroumbas, Pattern Recognition,
3rd ed. New York: Academic, Feb. 2006.
[23] Carnegie-Mellon-University, “CMU/VASC image data-
base19972003,” link
[24] Carlos.Aguilera, “Finding local Extrema in Matlab”
al-extrema), link to matlab files.