Electronics and Communication
St.Joseph’s college of engineering
Dr. Ezhilarasi. P
Electronics and Communication Engineering
St.Joseph’s college of engineering
introduce a new self adaptive algorithm for segmentation human skin regions in
color images. Skin detection and segmentation is an active research topic, and
many solutions have been proposed so for, especially concerning skin tone
modeling in various color spaces these models are used for image pixel
classification, but its accuracy is limited due to high variance and low human
skin color. With recent advance in multimedia technology, the involvement of
digital images/videos in crimes has been increasing significantly.
Identification of individuals in these images/videos can be challenging. For
example, in cases of child sexual abuse, child pornography, and masked gunman,
the faces of criminals or victims are often hidden or covered and only some
body parts (e.g., back, arm, thigh) can be observed from the digital evidence.
However an efficient skin segmentation algorithm should be capable to detect
skin pixels efficiently by overriding these effects. In this research study, a
RGB based skin segmentation technique is being Presented for extraction of skin
pixels. Therefore, for robust skin pixel detection, a dynamic skin color model
that can cope with the changes must be employed.
Keywords—Vein, Biometric,forensic, vein pattern
Veins are present under the body
surface. They are not visible to the naked eye. Most of the popular methods of
vein imaging are based on infrared (IR) light absorption. The primary
transporter of oxygen in human beings is the hemoglobin molecule. Arteries
carry oxygenated blood and veins carry deoxygenated blood and blood contains
hemoglobin. In recent days scientific research and technology development is take a vital role
in nation growth. Person identification technology is a critical process in
forensic investigation. Considerable effort has been expanded to develop Face,
Fingerprint, Palm print, DNA and Dental identification system. These parameter
are regularly used by law enforcement agents around the world.
However, they are not applicable in where only
partial Non-facial skin of criminal or victims or suspect is observable in
photographic evidence or digital image.ATM robbery, Hijacking, Violent
protester and sex offenses against child are among these cases. Pedophiles are
usually careful not to show their faces in images image for fear of
identification. The problem is child pornography is increasing because of
proliferation of such material electronically and in effective identification
The U.S. Customs Service
estimated that 100,000 websites offer
child pornography1. Traditionally, it was impossible to use vein patterns for
forensic identification, because they were almost invisible to the naked eye in
color images taken by consumer cameras. Recently, we overcame this limitation
by developing a method to uncover vein patterns from color images 34. The application of
skin segmentation is being used widely in all areas of computer science and its
applications. Detecting and tracking face and hands are important for gesture
recognition and human computer interaction. The goal of skin detection is to
extract the skin pixels from an image by applying the good class of skin
segmentation algorithm. Here skin segmentation is a process of segmenting the
pixels of an image into skin and non-skin regions. The algorithm should be
capable enough to make decide about a particular pixel into skin-regions or
non-skin skin regions. The main aim of the proposed system is to identify the
criminal based on the vein pattern recognition by comparing the vein pattern
capture from the crime scene with the image in the database.
Huge numbers of the current skin
division strategies can be classified into the skin shading discovery approach
or the picture division approach. Utilizing the created model, skin areas in
pictures are restricted by thersholding, gaze upward table, credulous bayes
classifier, or more unpredictable example acknowledgment strategies.
In this examination, skin division calculation is utilized as a
preprocessing venture to suspect databases which are normally gathered in
controlled situations, for example, the detainee databases and the sex
wrongdoer registries. Because of their expansive sizes, these database pictures
request a great deal more programmed handling than proof pictures gathered from
wrongdoing scenes, which can be prepared physically or semi naturally by
measurable officers. For calculation improvement and framework assessment,
shading pictures of different body parts were gathered.
To minimize a suspect database, pictures were gathered
in a standard posture and perspective condition, while to mimic proof pictures,
pictures were gathered in shifting stance and perspective conditions.
A. AUTOMATIC MATCHING ALGORITHM
Using an automatic
matching algorithm, we match resultant images from the RGB-NIR mapping approach
and find that its matching result is comparable to the result from matching NIR
section, we present our vein visualizing algorithm. It is composed of two
parts: a mapping model and an automatic neural network weight adjustment
The existing methods are based on the premise that the skin color can
be effectively modeled in various color spaces, which allows segmenting the
skin regions in color images. Using skin color models, every pixel may be
classified to the skin or non-skin class based on its position in the color
space, independently from its neighbors. Alternatively, the probability that
each pixel presents the skin can be determined, which transforms a color image
into a skin probability map (PS).
The map may be binaries using a
certain acceptance threshold in order to extract the skin regions. This problem
has been widely studied, and a large number of skin color models were
introduced over the years. The pixel-wise classification may be improved by
incorporating information extracted from the texture, as well as by spatial
analysis of the pixels that have high skin probability.
DISADVANTAGES OF EXISTING SYSTEM
a) A system that
is designed using RGB color space can take an advantage of a large number of
existing software routines, since this color space has been around a number of
b) RGB is not
efficient when dealing with Real-world images and processing an image in RGB
color space is usually not the most efficient method.
c) The most
compelling reason to adopt an RGB workflow is to increase the print provider’s
ability to “match the original”-the RGB color space simply allows for
a wider range of colors that leads to lower accuracy.
d) This algorithm
fully depends on the intensity of the image not the shape and texture. So the accuracy and sensitivity is low.
e) Clearly, the
more data you input to the device, the more you can output. But more
output is not possible in the existing system.
f) It is not
necessary to train a threshold-based method. Hence, the efforts in finding
datasets and implementing training procedures can be omitted.
methods are computationally efficient and easy to implement. The lower
performance of threshold-based methods and the lack of adjustable thresholds
are the disadvantages.
PROPOSED SYSTEM RESULTS
IMAGE DE-NOISING USING 2D HYBRID MEDIAN FILTER
Images capture by the camera and processed
and stored in memory. During this process the images are corrupted due to
impulse noises. The image pixels are getting damaged due to these noises. The
noise occurs due to transmission errors, malfunctioning pixel elements in the
camera sensors, faulty memory locations, and timing errors in analog-to-digital
conversion. Then our goal is to remove that type of noise in maximum amount by
preserving the main image features. Image processing consists of many filters
in order to remove the impulse noises. One of the filters is Hybrid median
filter which is somewhat improved version of median filter, which removes the
noise better than median filter.
The image degradation should
not be there in image processing. For that we have to remove noise in an image
as much as possible. Impulse noises are classified into two major types.
· Salt and pepper noise (equal height impulses) impulse values are
represented as 0 and 255. Typical noise sources include flecks of dust inside
the camera and overheated or faulty CCD elements
· Random-valued impulse noise (unequal height impulses) impulse
values are between 0 and 255.
Fig.1.(a)Input (b) Hybrid median filter
IMAGE ENHANCEMENT USING IMAGE HISTOGRAM ADAPTIVE IMPROVEMENT
This project, we
propose an effective image enhancement algorithm especially suitable for low
quality skin images, which can improve the clarity and continuity of the pixels.
The proposed algorithm estimates the finger elasticity by approximating the
uneven finger skin due to poor skin condition or imperfect acquisitions.
Image enhancement is
considered as one of the most important techniques in image research. The main
aim of image enhancement is to enhance the quality and visual appearance of an
image, or to provide a better transform representation for future automated
image processing. It is necessary to enhance the contrast and remove the noise
to increase image quality. One of the most important stages in medical images
detection and analysis is Image Enhancement Techniques. It improves the clarity
of images for human viewing, removing blurring and noise, increasing contrast,
and revealing details. These are examples of enhancement operations. The
existing techniques of image enhancement can be classified into two categories:
Spatial Domain and Frequency Domain Enhancement. In this project, we present an
overview of Image Enhancement Processing Techniques in Spatial Domain.
Fig.2. (a) Input image (b)
improved contrast and enhanced image
Morphological image processing is a
collection of non-linear operations related to the shape or morphology of
features in an image. Morphological operations rely only on the relative
ordering of pixel values, not on their numerical values, and therefore are
especially suited to the processing of binary images.
Morphological operations can also be applied to grey-scale images such
that their light transfer functions are unknown and therefore their absolute
pixel values are of no or minor interest. Morphological techniques probe an
image with a small shape or template called a structuring element.
Fig.3. (a) Erosion (b) Dilation
operations can also be applied to grey-scale images such that their light
transfer functions are unknown and therefore their absolute pixel values are of
no or minor interest region of interest is segmented based on user requirement.
Background or image which must be projected or highlighted.
ROI –region of Interest
Region of interest is acronym of ROI.
ROI is a portion of an image that you want to filter or perform some other
operation on. There are three types of ROI calculations is available.
ADVANTAGE OF REGION OF INTEREST
advantage of ROI algorithm,
Region of interest method
can correctly separate the regions that have the same properties we define.
This methods can provide
the originals images which have clear edges that segmentation results.
Fig.3. (a) Input image (b) Region of interest
NAME OF THE PARAMETER
hmf(hybrid median filter)
RESULTS & DISCUSSION
This database consists of the right hand
and left hand images from internet. The images of both male and female subjects
are taken for the project. The size of the input image is 720×480 pixels. The
ROI extracted is from the resized image of size 390×189 pixels. Matlab 2015b
was used to obtain the above results. Vein pattern over the original image is
Bio-metric systems are
used to identify and verify human. It analyzes individual’s unique identity.
Basically, bio-metric systems are used for security purpose, the main objective
of biometric system is to achieve of transparent identity. Bio-metric measure
individual’s unique physical or behavioral characteristic to recognize and
authenticate their identity. Image skin segmentation is used as a preprocessing
step to suspect database which are usually collected as a database.
are used to improve the quality and increase the accuracy of the pixel which
are present in the evidence images. If the image is for away from the camera
picture quality will be low and data might not clear. Morphological image
processing is a collection of non-linear operations related to the shape or
morphology of features in an image. Morphological operations can
also be applied to grey-scale images such that their light transfer functions
are unknown and therefore their absolute pixel values are of no or minor
interest region of interest is segmented based on user requirement. Background
of image which must be projected or highlighted. ROI is define and classified
in different types. Region of interest and Non-Region of interest.
describes the use of vein pattern and its trait.The first step in vein pattern
recognition is image pre-processing which includes image enhancement,
filtering, de-noising with the help of Hybrid median filter and adaptive mean
shift algorithm. The processed image is under gone segmentation by histogram
adaptive improvement algorithm followed by morphological operation. Thus Region
of interest is obtained and non-region of interest is eliminated. The proposed
system is an efficient method to identify criminals or victims. In future, vein
pattern matching with large data base for identifying criminals or victims will
The author would like to thank Dr.Noah craft
for providing constructive comments on the preliminary version of this work and
would like to thank the department of electronics and communication
engineering, St.Joseph’s college of engineering.
Ashwini J. Kolar, et al., “Experimental Prototype of Non- Invasive
Low Cost Subcutaneous Vein Detection Device”, In IEEE International conference
on Signal and Image Processing ICSIP, pp. 170-179. 2013.
Arfika nurhudatiana, student member IEEE, and adams wai-kin kong
proposed”On criminal identification in color skin images using skin marks
(RRPVSM) and fusion with inferred vein patterns” -2015.
Chakravorty, Tanushri, et al. “Low-cost subcutaneous vein
detection system using ARM9 based single board computer.” Electronics
Computer Technology (ICECT), 2011 3rd International Conference on. Vol. 2.
Deepak Prasanna.R, et al., “Enhancement of vein patterns in hand
image for biometric and biomedical application using various image enhancement
techniques” International conference on Modeling optimization and Computing,
pp. Procedia Engineering 38,elsevier1174 – 1185, 2012.
Dipti Verma and Dr.Sipi Dubey, “A survey on biometric
authentication techniques using Palm vein feature”, Journal of global research
in computer science JGRCS ,vol-5 no-8,ISSN-2229 371X,2014.
Dipti Verma and Dr.Sipi Dubey, ” Two Level Centre of Gravity
Computation –An Important Parameter for Offline Signature Recognition” in IJCA
2012,SEPT.VOL54, ISSUE 6. 31
Graham Leedham, “Near- and- Far- Infrared Imaging for Vein Pattern
Biometrics”, Proceedings of the IEEE International Conference on Video and
Signal Based Surveillance, 2006.
Hyeon Chang LEE, Byung Jun KANG, Eui Chul LEE, Kang Ryoung PARK,
«Finger vein recognition using weighted local binary pattern,» Journal of
Zhejiang University-SCIENCE C (Computers & Electronics), pp. 514- 524,
Jae Hee Song et al., “Vein
Visualization using a Smart Phone with Multispectral Wiener Estimation for
Point-care-Application.” IEEE Journal of Biomedical and Health Informatics,
Manam Mansoor., et al., “Real Time Low Cost Infrared Vein Imaging
System.”, In IEEE International conference on Signal Processing, Image
Processing and Pattern Recognition ICSIPR, Coimbatore, pp. 117 – 121. Feb.
Shi Zhao, Yiding Wang and Yunhong Wang, “Extracting Hand Vein
Patterns from Low-Quality Images: A New Biometric technique Using Low-Cost Devices”,
IEEE, 4th International Conference on Image and Graphics, 2007.2 Wang Lingyu,
Rafael, C. Gonzalez and
Richard E. Woods “Digital Image Processing” Second Edition.
Tarik Arici., et al., “A Histogram Modification Framework and Its
Application for Image Contrast Enhancement.”, IEEE Transaction on image
processing., vol. 18,9,pp. 1921 – 1935, September 2009. 32
Verma, Dipti, and Sipi Dubey. “Processing and Enhancement of
Palm Vein Image in Vein Pattern Recognition System.” (2015).
Crisan, Septimiu, Joan Gavril Tarnovan, and Titus Eduard Crisan.
“A low cost vein detection system using near infrared radiation.”
Sensors Applications Symposium, 2007. SAS’07. IEEE. IEEE, 2007.
Wang Kejun, Ding Yuhang, Wang Dazhen, “A Study of Hand Vein-based
Identity Authentication Method”, Science & Technology Review, 23(1): 35-37,
Yuksel.A, et al., “Hand Vein Biometry Based on Geometry and
Appearance Method.”, IET computer vision ., vol. 5, no. 6 pp. 398- 406,2011.