Skip to content
Snippets Groups Projects
Commit 2d1ca18d authored by even's avatar even
Browse files

Article: intro revisited

parent e5b79712
No related branches found
No related tags found
No related merge requests found
......@@ -102,3 +102,53 @@
series = {LNCS},
publisher = {Springer}
}
@article{GioiAl10,
title = {{LSD}: A Fast Line Segment Detector with a False Detection Control},
author = {Gioi, R. G. von and Jakubowicz, J.
and Morel, J.-M. and Randall, G.},
journal = {IEEE Trans on Pattern Analysis and Machine Intelligence},
volume = {32},
number = {4},
month = apr,
year = {2010},
pages = {722--732},
doi = {10.1109/TPAMI.2008.300}
}
@article{MatasAl00,
title = {Robust detection of lines using the progressive probabilistic
{H}ough transform},
author = {Matas, Jiri and Galambos, Charles and Kittler, Josef},
journal = {Computer Vision and Image Understanding},
volume = {78},
number = {1},
year = {2000},
pages = {119--137}
}
@inproceedings{LuAl15,
title = {CannyLines: A parameter-free line segment detector},
author = {Lu, Xiaohu and Yao, Jian and Li, Kai and Li, Li},
booktitle = {International Conference on Image Processing (ICIP)},
publisher = {IEEE},
year = {2015},
pages = {507--511}
}
@article{AkinlarTopal12,
title = {EDPF: a real-time parameter-free edge segment detector
with a false detection control},
author = {Akinlar, Cuneyt and Topal, Cihan},
journal = {International Journal of Pattern Recognition
and Artificial Intelligence},
volume = {26},
number = {01},
year = {2012},
pages = {1255002},
doi = {10.1142/S0218001412550026}
}
\section{Conclusion and perspectives}
\label{sec:conclusion}
In this paper we introduced a new edge detector based on a local analysis of
the image gradient and on the use of blurred segments to vehiculate an
estimation of the edge thickness.
......@@ -35,8 +37,15 @@ But this default remains quite sensible in unsupervised context.
In future works, we intend to provide some protection against this drawback
by scoring the detection result on the base of a characterization of the
initial context.
Then experimental validation of the consistency of the estimated
width and orientation values on real situations are planned in
different application fields.
Then experimental validation of the consistency of the estimated width and
orientation values on real situations are planned in different application
fields.
In particular, straight edges are rich visual features for 3D scene
reconstruction from 2D images.
The preimage of the detected blurred segments,
i.e. the space of geometric entities which numerization matches this
blurred segment, may be used to compute some confidence level in the 3D
interpretations delivered, as a promising extension of former works
on discrete epipolar geometry \cite{NatsumiAl08}.
%\section*{Acknowledgements}
\section{Experimental validation}
\label{sec:expe}
The evaluation stage aims at quantifying the advantages of the new detector
compared to the former one.
For a fair comparison, the process flow of the former method (the initial
......
\section{Introduction}
\label{sec:intro}
\subsection{Motivations}
Straight edge detection is a preliminary step of many image analysis
processes. Therefore it is always an active reasearch topic centered
on the quest of still faster, more accurate or more robust-to-noise
methods.
{\it TOWRITE : petit \'etat de l'art en r\'esumant IWCIA'09 et en ajoutant
quelques id\'ees perso (Hough, local, ...).
Parameter-space-based methods : robust to noise, well suited to
supervided context \cite{EvenMalavaud00}.
Most of works aim at reducing their time complexity. }
processes. Therefore it is always an active research topic centered on
the quest of still faster, more accurate or more robust-to-noise methods
\cite{MatasAl00,GioiAl10,AkinlarTopal12,LuAl15}.
However they seldom provide an exploitable measure of the output edge
quality, based on intrinsic properties such as sharpness, connectivity
or scattering.
%Some information may sometimes be drawn from their specific context,
%for example through an analysis of the peak in a Hough transform accumulator.
These methods rarely provide a direct measure of the quality of the output
edge, such as sharpness, connectivity or scattering.
Some information may often be drawn from their specific context, for example
through an analysis of the peak in a Hough transform accumulator, or
TO COMPLETE.
In particular, the accuracy of the edge orientation may be quite critical
in some application contexts, such as computer vision.
Digital geometry is a recent research domain where new mathematical definitions
of quite classical geometric objects, such as lines or circles, are introduced
to better fit to the discrete nature of most of todays data to process.
In particular, the notion of blurred segment \cite{DebledAl05,Buzer07} was
introduced to cope with the image noise or other sources of imperfections
from the real world by the mean of a width parameter.
Efficient algorithms were designed to recognize these digital objects in
binary images.
In digital geometry, the notion of blurred segment \cite{DebledAl05,Buzer07}
was introduced to cope with the image noise or other sources of
imperfections from the real world. The preimage of that geometrical object,
ie the space of geometric entities which numerization matches this
blurred segment, may convey useful information to evaluate possible moves in
the 3D interpretations drawn, as a promising extension of former works
on discrete epipolar geometry \cite{NatsumiAl08}.
Our work aims at designing a flexible tool to detect such blurred segment
in gray-level images for as well supervised as unsupervised contexts.
Our work aims at designing a flexible tool to detect blurred segments
with optimal width and orientation in gray-level images for as well
supervised as unsupervised contexts.
We seek for user-friendly solutions with ideally no parameter to set,
or at least quite few values with intuitive meaning to an end user.
\subsection{Method overview and previous work}
\subsection{Previous work}
The method we propose is based on a first rough detection in a local area
In a former paper \cite{KerautretEven09}, we already introduced an efficient
tool to detect blurred segments of fixed width in gray-level images.
It is based on a first rough detection in a local area
of the image either defined by the user in supervised context or blindly
explored in automatic mode. The goal is to disclose the presence of an edge.
Therefore, a simple test as the gradient maximal value is performed.
In case of success, refinement steps are run through an exploration of
the image in the direction of the detected edge.
In order to prevent local disturbances, for instance the intersection with
a sharper edge, all the local gradient maxima are successively tested,
and the gradient orientation consistency is checked.
In order to prevent local disturbances such as the presence of a sharper
edge nearby, all the local gradient maxima are successively tested
untill a correct candidate with an acceptable gradient orientation is found.
Only the gradient information is processed as it provides a good information
on the image dynamics, and hence the presence of edges.
Trials to also use the intensity signal were made though costly correlation
techniques, but they were mostly successful for detecting objects with
stable appearance such as metallic pipes \cite{AubryAl17}.
We already designed and experimented an exploratory detector
\cite{KerautretEven09}.
Despite of good performances obtained compared to other methods from the
literature, several drawbacks remain.
First, the blurred segment width is not measured, but initially set by the
user to meet the application requirements, so that no quality information
can be derived from the computed segment.
Moreover, the blurred segment hull is left free to shift sidewards, or worst,
to rotate around a thin edge in the image, so that the produced orientation
value can be largely biased.
Despite of good performances obtained compared
to other methods from the literature, several drawbacks remained.
At first, a fixed width value was set by the user according to the
application requirements, and detected segments were embedded in that
fixed tolerence whatever their dispersion be. When this dispersion is low,
the blurred segment is free to shiff sidewards, or worst, to rotate, thus
degrading the provided position and rotation measures.
Then, two refinement steps are systematically run to cope with most of the
tested data, although this is useless when the first detection is successfull.
Beyond, there is no guarantee that this could treat all kinds of data.
The search direction is fixed by the detected direction at the former step,
and there is necessarily a limit on this direction accuracy - at least
linked to the restricted directions encoded in a limited grid - so that
other steps would have been necessary to deal with high resolution images.
Then two refinement steps were arbitrarily run to cope with most of
the tested data, uselessly when the first one was successfull.
Beyond, there was no guarantee that this could treat all kinds
of data. The search direction is fixed by the detected direction at the
former step, and there is necessarily a limit on this direction
accuracy - at least linked to the restricted directions encoded
in a limited grid - so that other steps would have been necessary
to deal with high resolution images.
\subsection{Main contritions}
Our study relies only on the use of the image gradient, as it provides a
good information on the signal dynamics, and hence the presence of edges.
Trials were made to use also the intensity signal though expensive
correlation techniques, but it was mostly successful for tracking objects
with stable appearance such as metallic pipes \cite{AubryAl17}.
The work presented in this paper aims at solving both former mentioned
drawbacks through two main contributions:
the concept of adaptive directional scanner designed to get some compliance
to the unpredictable orientation problem;
the control of the assigned width to the blurred segment recognition algorithm,
intended to derive more reliable information on the edge orientation and
quality.
As a side effect, this two major evolutions led to a noticeable improvement
of the execution time.
Organisation of the paper : TO WRITE.
In the next section, the main theoretical notions this work relies on are
introduced, with a specific focus on the new concept of adaptive directional
scanner.
Then the new detector workflow and its integration into both supervised and
unsupervised contexts are presented and discussed in \RefSec{sec:method}.
Experiments led to assess the expected increase of performance are decribed
in \RefSec{sec:expe}.
Finally achieved results are summarized in \RefSec{sec:conclusion},
followed by some open perspectives for future works.
\newcommand{\RefFig}[1]{Fig.\,\ref{#1}}
\newcommand{\RefSec}[1]{Section\,\ref{#1}}
\newcommand{\RefSec}[1]{section\,\ref{#1}}
\newcommand{\RefTab}[1]{Tab.\,\ref{#1}}
\documentclass[runningheads]{llncs}
%\usepackage[utf8]{inputenc}
%\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{graphicx}
%\graphicspath{{./Fignotions/}{./Figmethod}}
......@@ -25,10 +25,11 @@
\authorrunning{P. Even et al.}
\institute{Universit\'e de Lorraine, LORIA, UMR 7503, Nancy, France
\email{philippe.even,hoai-diem-phuc.ngo\{at\}loria.fr}
\and Universit\'e de Lyon 2, LIRIS, Lyon, France
\email{bertrand.kerautret\{at\}univ-lyon2.fr}}
\institute{Universit\'e de Lorraine, LORIA (UMR 7503), Nancy, France
\email{philippe.even@loria.fr},
\email{hoai-diem-phuc.ngo@loria.fr}
\and Universit\'e Lyon 2, LIRIS (UMR 5205), Lyon, France
\email{bertrand.kerautret@univ-lyon2.fr}}
\maketitle
......
\section{The detection method}
\label{sec:method}
\subsection{Workflow of the detection process}
The work-flow of the blurred segment detection process is summerized
The workflow of the blurred segment detection process is summerized
in the following figure.
\begin{figure}[h]
......@@ -33,7 +35,7 @@ in the following figure.
\put(330,18){\scriptsize $\mathcal{B}_3$}
\put(322,15){\vector(1,0){22}}
\end{picture}
\caption{The detection process main work-flow.}
\caption{The detection process main workflow.}
\label{fig:workflow}
\end{figure}
......@@ -123,7 +125,7 @@ the awaited one.
\label{fig:voisins}
\end{figure}
This detection procedure can be used to dectect as well straight edges
This detection procedure can be used to detect as well straight edges
as thin straight objects. In the first case, the gradient vectors of all
edge points are assumed to be oriented in the same direction. But if the
sign of the gradient direction is not considered, points with gradient in
......
\section{Theoretical background}
\label{sec:notions}
\subsection{Blurred segment}
Our work relies on the notion of digital straight line as classically
......@@ -23,9 +25,9 @@ of points in $\mathbb{Z}^2$ that all belong to a digital line of
arithmetical width $\varepsilon$.
\end{definition}
Linear time algorithms have been developed to recognize a blurred segment
of assigned width $\varepsilon$ \cite{DebledAl05,Buzer07}.
They are based on an incremental growth of the convex hull of the blurred
In this work, we use a linear-time algorithm that was developed to recognize
a blurred segment of assigned width $\varepsilon$ \cite{DebledAl05}.
It is based on an incremental growth of the convex hull of the blurred
segment when adding each point successively.
The minimal width $\mu$ of the blurred segment $\mathcal{B}$ is the
arithmetical width of the narrowest digital straight line that contains
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment