148 lines
8.1 KiB
TeX
148 lines
8.1 KiB
TeX
\section{Ball Detection}
|
|
\label{p sec ball detection}
|
|
|
|
The very first task that needed to be accomplished was to detect the ball,
|
|
which is uniformly red-colored and measures about 6 cm in diameter. We decided
|
|
to use a popular algorithm based on color segmentation \cite{ball-detect}. The
|
|
idea behind this algorithm is to find the biggest red area in the image and
|
|
assume that this is the ball. First, the desired color needs to be defined as
|
|
an interval of HSV (Hue-Saturation-Value) \cite{hsv} values. After that, the
|
|
image itself needs to be transformed into HSV colorspace, so that the regions
|
|
of interest can be extracted into a \textit{binary mask}. The contours of the
|
|
regions can then be identified in a mask \cite{contours}, and the areas of the
|
|
regions can be calculated using the routines from the OpenCV library. The
|
|
center and the radius of the region with the largest area are then determined
|
|
and are assumed to be the center and the radius of the ball.
|
|
|
|
\begin{figure}[ht]
|
|
\includegraphics[width=\textwidth]{\fig ball-detection}
|
|
\caption{Ball detection. On the right is the binary mask}
|
|
\label{p figure ball-detection}
|
|
\end{figure}
|
|
|
|
It is sometimes recommended \cite{ball-detect} to eliminate the noise in the
|
|
binary mask by applying a sequence of \textit{erosions} and \textit{dilations},
|
|
but we found, that for the task of finding the \textit{biggest} area the noise
|
|
doesn't present a problem, whereas performing erosions may completely delete
|
|
the image of the ball from the mask, if it is relatively far from the robot and
|
|
the camera resolution is low. For this reason it was decided not to process the
|
|
binary mask with erosions and dilations, which allowed us to detect the ball
|
|
even over long distances.
|
|
|
|
The advantages of the presented algorithm are its speed and simplicity. The
|
|
major downside is that the careful color calibration is required for the
|
|
algorithm to function properly. If the HSV interval of the targeted color is
|
|
too narrow, then the algorithm might miss the ball; if the interval is too
|
|
wide, then other big red-shaded objects in the camera image will be detected as
|
|
the ball. A possible approach to alleviate these issues to a certain degree
|
|
will be presented further in the section \ref{p sec field detect}. To
|
|
conclude, we found this algorithm to be robust enough for our purposes, if the
|
|
sensible color calibration was provided.
|
|
|
|
\section{Goal Detection}
|
|
\label{p sec goal detect}
|
|
|
|
The goal detection presented itself as a more difficult task. The color of the
|
|
goal is white, and there are generally many white areas in the image from the
|
|
robot camera, which have area larger than that of the image of the goal, for
|
|
example the white field lines and the big white wall in the room with the
|
|
field. To deal with the multitude of the possible goal candidates, we
|
|
propose the following heuristic algorithm.
|
|
|
|
\begin{figure}[ht]
|
|
\includegraphics[width=\textwidth]{\fig goal-detection}
|
|
\caption{Goal detection}
|
|
\label{p figure goal-detection}
|
|
\end{figure}
|
|
|
|
First, all contours around white areas are extracted by using a procedure
|
|
similar to that described in the section \ref{p sec ball detection}. Unlike in
|
|
the ball detection, the resulting binary mask undergoes some slight erosions
|
|
and dilations, since in the goal shape detection the noise is undesired. Next,
|
|
the \textit{candidate preselection} takes place. During this stage only $N$
|
|
contours with the largest areas are considered further (in our experiments it
|
|
was empirically determined that $N=5$ provides good results). Furthermore, all
|
|
convex contours are rejected, since the goal is a highly non-convex shape.
|
|
After that, a check is performed, how many points are necessary to approximate
|
|
the remaining contours. The motivation behind this is the following: it is
|
|
clearly visible that the goal shape can be perfectly approximated by a line
|
|
with 8 straight segments. On an image from the camera, the approximation is
|
|
almost perfect when using only 6 line segments, and in some degenerate cases
|
|
when the input image is noisy, it might be necessary to use 9 line segments to
|
|
approximate the shape of the goal. Any contour that requires a different number
|
|
of line segments to be approximated is probably not the goal. The preselection
|
|
stage ends here, and the remaining candidates are passed to the scoring
|
|
function.
|
|
|
|
The scoring function calculates, how different are the properties of the
|
|
candidates are from the properties, that an idealized goal contour is expected
|
|
to have. The evaluation is happening based on two properties. The first
|
|
property is based on the observation, that the area of the goal contour is much
|
|
smaller than the area of its \textit{enclosing convex hull} \cite{convex-hull}.
|
|
The second observation is that all points of the goal contour must lie close to
|
|
the enclosing convex hull. The mathematical formulation of a corresponding
|
|
scoring function can then look like the following:
|
|
|
|
\begin{equation*}
|
|
S(c)=\frac{A(c)}{A(Hull(c))}+\displaystyle\sum_{x_i \in c}\min_{h \in Hull(c)
|
|
}(||x_i-h||)
|
|
\end{equation*}
|
|
|
|
The contour, that minimizes the scoring function, while keeping its value under
|
|
a certain threshold is considered the goal. If no contour scores below the
|
|
threshold, then the algorithm assumes that no goal was found. An important note
|
|
is that the algorithm is designed in such a way, that the preselection and
|
|
scoring are modular, which means that the current simple scoring function can
|
|
later be replaced by a function with a better heuristic, or even by some
|
|
function that employs machine learning models.
|
|
|
|
Our tests have shown, that when the white color is calibrated correctly, the
|
|
algorithm can detect the goal almost without mistakes, when the goal is present
|
|
in the image. Most irrelevant candidates are normally discarded in the
|
|
preselection stage, and the scoring function improves the robustness further.
|
|
Figure \ref{p figure goal-detection} demonstrates the algorithm in action. On
|
|
the right is the binary mask with all found contours. On the left are the goal,
|
|
and one contour that passed preselection but was rejected during scoring.
|
|
|
|
One downside of this algorithm, is that in some cases the field lines
|
|
might appear to have the same properties, that the goal contour is expected to
|
|
have, therefore the field lines can be mistaken for the goal. We will describe,
|
|
how we dealt with this problem, in the section \ref{p sec field detect}.
|
|
|
|
\section{Field Detection}
|
|
\label{p sec field detect}
|
|
|
|
The algorithm for the field detection is very similar to the ball detection
|
|
algorithm, but some concepts introduced in the section \ref{p sec goal detect}
|
|
are also used here. This algorithm extracts the biggest green area in the
|
|
image, finds its enclosing convex hull, and assumes everything inside the hull
|
|
to be the field. In here, when we extract the field, we apply strong Gaussian
|
|
blurring and erosions-dilations combination to the binary mask, so that the
|
|
objects on the field are properly consumed.
|
|
|
|
\begin{figure}[ht]
|
|
\includegraphics[width=\textwidth]{\fig field-detection}
|
|
\caption{Field detection}
|
|
\label{p figure field-detection}
|
|
\end{figure}
|
|
|
|
Such rather simple field detection has two important applications. The first
|
|
one is that the robot should be aware, where the field is, so that it doesn't
|
|
try to walk away from the field. Due to time constraints, we didn't implement
|
|
this part of the behavior. The second application of field detection is the
|
|
improvement of the quality of goal and ball recognition. As was mentioned in
|
|
the section on ball detection, the current algorithm might get confused, if
|
|
there are any red objects in the robot's field of view. However, there
|
|
shouldn't be any red objects on the field, except the ball itself. So, if
|
|
everything that's not on the field is ignored, when trying to detect the ball,
|
|
the probability of identifying a wrong object decreases. On the other hand, the
|
|
problem with the goal detection algorithm was that it could be distracted by
|
|
the field lines. So, if everything on the field is ignored for goal
|
|
recognition, then the accuracy can be improved.
|
|
|
|
\begin{figure}[ht]
|
|
\includegraphics[width=\textwidth]{\fig combined-detection}
|
|
\caption{Using field detection to improve ball and goal detection}
|
|
\label{p figure combined-detection}
|
|
\end{figure}
|