Volume 2 No 2 April 2008
|
| 1. |
An Efficient Geometric feature based License Plate Localization and Recognition Download PDF |
|
|
Authors |
C. Nelson Kennedy Babu, Krishnan Nallaperumal |
|
|
Abstract |
License plate location is an important phase in vehicle
license plate recognition for intelligent transport systems. The
objective of this work is to design and implement algorithms for
License Plate Recognition (LPR) of Indian License Plates. This paper
presents a robust time algorithms of license plate location,
segmentation and reorganization of the characters present in the
located plate. The proposed License Plate Localization algorithm
consists of required preprocessing operations leading to successful
plate identification. Geometry-based method is used to locate the
position of the license plate. Thin window scanning method for
Segmentation of Characters and an Artificial Neural Network (ANN)
is designed to recognize characters of letter and characters of
numbers. Experimental results show that the proposed method
improves the effectiveness and robustness for License plate
recognition. |
|
|
Keywords |
License plate localization, recognition, image processing, segmentation, Artificial Neural Network. |
|
| 2. |
A Multiscale Morphological Watershed Segmentation using Color Composite Gradient and Marker Extraction Download PDF |
|
|
Authors |
Krishnan Nallaperumal, Senior Member, IEEE, Krishnaveni. K, Member, IEEE |
|
|
Abstract |
A novel multiscale morphological watershed
segmentation algorithm for the segmentation of color images is
proposed in this paper. The conventional watershed segmentation
algorithm is not suitable for color image directly, since it is hard to
decide a suitable color gradient measurement for color image. The
proposed algorithm uses HSV color space to define color contrast
gradient which is multiplied with multiscale morphological gradient
of the intensity image. The marker extractions of this composite
color gradient image are obtained by a thresholding technique to
avoid over segmentaion vivid in conventional watershed algorithm
and given as input to the watershed algorithm which uses Hill-
Climbing approach to identify and label the neighbourhood pixels
in the marker image. Experimental results show that the proposed
algorithm avoids the over segmentation problem and produce the
segmented result which is good for human perception. |
|
|
Keywords |
Watershed Segmentation |
|
| 3. |
Adaptive Weighted Demosaicing for Single Sensor Camera Images Download PDF |
|
|
Authors |
S.S. Vinsley, Member, IEEE, and Krishnan Nallaperumal, Senior Member, IEEE |
|
|
Abstract |
In order to reduce the hardware size and cost, many digital
still cameras (DSC) use a single sensor array equipped with a color
filter array (CFA) to capture any one of the three primary color
components, R (Red), G (Green), or B (Blue), on each pixel location.
This single band is used to generate the true RGB image. Among the
various suggested CFAs, the Bayer CFA pattern is the most prevalent
one, where G pixels occupy half of all, and R and B pixels share the
others. Demosaicing is a process of interpolating the missed colors,
which aims to reconstruct the missing colors as close as possible by
keeping the computational complexity is very low. In this paper a
novel adaptive edge preserving, edge directed inter-plane weighted
interpolating technique is proposed for color reproduction from
Bayer mosaic images. The proposed interpolation algorithm aims to
effectively estimate the missing green component in the edge and
texture regions with red and blue planes to be estimated based on the
interpolated values of the green plane. The algorithm uses a gradient
test to effectively interpolate though the edge direction. Variance of
color differences is also used as a supplementary criterion to
determine the interpolation direction. Adaptive weight values are
derived for each and every non green pixel locations depending on
the classification as sharpness and smoothness. Experimental results
show that the proposed method performs much better than other
latest techniques in terms of PSNR. |
|
|
Keywords |
Color demosaicing, interpolation, color filter array,single sensor digital camera. |
|
| 4. |
Parallel Algorithm for Labeling a VRML Generalized Scene Graph Download PDF |
|
|
Authors |
M.Mohammed Sathik |
|
|
Abstract |
VR Technology presents simulated 3D scenes to a
user in such a way that he or she perceives that it is a real scene. The
scenes in a VR world are represented using a data structure called
“Scene Graphs”. Inherently Scene Graphs are nothing but Tree data
types. The Scene Graph data type is updated to another object called
“Generalized Scene Graph [3] “ which comprises all the procedures
and methods, which are operating on the nodes of the scene graph as
well as the scene graph itself. While processing a generalized scene
graph, two phases of manipulation namely Inspection Phase and
Evaluation Phase are needed. In the Inspection Phase, all the nodes
are explored and its type and nature are inspected. While doing the
operation one has to search the nodes sequentially, which is time
consuming and elaborate. As time is a crucial factor in producing VR
worlds to the user under multi-user group communication, it is
intended to explore the possibility of adapting parallel algorithms [1]
on searching tree under various conditions.
|
|
|
Keywords |
Scene Graph, Generalize Scene Graph, Labeling Scene Graph, NLP Table, Parallelism |
|
| 5. |
Watershed Segmentation of Cervical Images Using Multiscale Morphological Gradient and HSI Color Space Download PDF |
|
|
Authors |
Krishnan Nallaperumal, Senior Member, IEEE, Krishnaveni. K, Member, IEEE |
|
|
Abstract |
In this paper, a novel watershed segmentation algorithm
is proposed for the segmentation of nucleus from the surrounding
cytoplasm of cervical cancer images. The proposed method converts
the input RGB image into HSI image which contains three
components hue, saturation and intensity. The saturation component
is thresholded to obtain the binary image and each pixel in the binary
image is multiplied with hue component to obtain the product image.
The intensity image is complemented, thresholded and merged with
the product image and smoothened. The local minima are reduced
using extended minima function and the multiscale gradient of this
resultant gray scale image is segmented using watershed algorithm
based on Hill Climbing technique. Experimental results add to the
computational efficiency of the algorithm, its shape maintaining,
edge preserving and scale-calibrating features. The performance is
also superior to most other segmentation techniques. |
|
|
Keywords |
Cervical,Watershed Segmentation |
|
| 6. |
Face Detection using Statistical Models and Explicitly-defined Skin Region Classifiers Download PDF |
|
|
Authors |
Ravi Subban, Krishnan Nallaperumal |
|
|
Abstract |
Human skin detection is the preliminary step in human face detection and face detection is the preliminary step in the human face recognition. This paper proposed a new human skin detection technique and gave a comparative analysis of skin detection techniques based on Bayesian, Gaussian and linear piecewise decision boundary classifiers using YCgCr color space. The new derived color space from YCbCr color space produced very good results for human skin detection. The results obtained using these three skin classification algorithms were compared. The Bayesian color model produced better results than Gaussian model. The skin detection method using linear decision boundary classifier in YCgCr color space produced relatively better results than both Gaussian and Bayesian color models. Lip region is detected to detect facial features using R/G and R/B color models. |
|
|
Keywords |
Bayesian, Gaussian, Skin Cluster Models, Lip Region detection. |
