H. Badioze Zaman et al. (Eds.): IVIC 2011, Part I, LNCS 7066, pp. 176–182, 2011.
© Springer-Verlag Berlin Heidelberg 2011
Empirical Performance Evaluation of Raster to Vector
Conversion with Different Scanning Resolutions
Bilal Abdulrahman T. Al-Douri, Hasan S.M. Al-Khaffaf,
and Abdullah Zawawi Talib
School of Computer Sciences
Universiti Sains Malaysia
11800 USM Penang, Malaysia
belal_200033@yahoo.com, {hasanm,azht}@cs.usm.my
Abstract. Empirical performance evaluation of raster to vector conversion is a
means of judging the quality of line detection algorithms. Many factors may
affect line detection. This paper aims to study scanning resolution of raster
images and its effects on the performance of line detection. Test images with
three different scanning resolutions (200, 300, and 400 DPI) are vectorised
using available raster to vector conversion software. The Vector Recovery
Index scores calculated with reference to the ground truth images and the
detected vectors are then obtained. These values are analysed statistically in
order to study the effects of different scanning resolutions. From the results,
Vextractor is found to be better (on average) compared to VPstudio and
Scan2CAD. For all the three resolutions, Vextractor and VPstudio perform
better than Scan2CAD. Different scanning resolutions affect the software
differently. The performance of Vextractor and VPstudio increases from low
resolution to moderate resolution, and then decreases with high resolution. The
performance of Scan2CAD decreases with the increase in the resolutions.
Keywords: Empirical Performance Evaluation, Raster to Vector Conversion,
Vector Recovery Index, Statistical Analysis, Visual Informatics.
1 Introduction
Empirical performance evaluation of raster to vector conversion is an important topic
in the area of graphics recognition. It is used as a means to judge the quality of line
detection algorithms. There are many factors that affect quality of line detection and
they may lead to better or poorer line detection rate. Some of these factors have been
studied such as noise level, vectorisation software, and cleaning methods [1] and the
other factors have yet to be studied such as scanning resolution of raster images.
Chhabra and Ihsin [2] used research prototypes and commercial software. The
criterion of the performance was EditCost Index. Liu et al. [3] evaluated the
performance of several research prototypes. They used different types of noise
(Gaussian noise, high frequency, hard pencil, and geometrical noise). VRI was used
to measure the performance of the methods. Only solid circular arcs were included in
their tests.