Abstract— field and medical applications. Other biometrics

Abstract— The
appropriate location or detection of vein vessels plays important role in
biometrics like human identification, matching, forensic field and medical
applications. Other biometrics features having particular drawback which will
not bother if we go for vein detection which having its unique features.
Identification of vein vessels is necessary in medical for needle infusion in
the patients. Sometimes many of doctors or nurses find difficult to locate vein
vessels  so it  very complicated for needle infusion and if it
goes wrong then it can cause Trauma, Hematoma, cuts on the bones, Allergic
reactions, Swell in the skin. There are many researches on palm vein, wrist
vein, dorsal vein, Finger vein, forearm vein etc. The framework for all types
of mentioned vein recognition is based on authentication, more or less all
having with same aspiration to detect vein in respective ROI. So system can be
developed which can detect veins off all the mentioned types on the body by
using vein detection system with infrared camera. Camera will take a real time
video with help of infrared rays and every vein on skin will give different contrast,
hence it will be easier to identify.

 

 

Keywords— Vein location, Infrared, Skin anatomy.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

I.    
INTRODUCTION

Vein
pattern recognition has been emerging technique in the field of biometric. Vein
pattern recognition
technique has it’s positives against other technologies. It can successfully
use in various other applications like field of forensic, human recognition
& medical applications such as bruises & burns, intravenous injection,
among children, blood transfusions, geriatrics, enter medications or fluids in
body. The upcoming biometric
techniques include vein detection pattern, DNA and body odor recognition. Vein
recognition also known as vascular biometrics and sub dermal nature of vein,
vascular biometric are consider to be highly secure modality, whereas other
types of biometrics had proven vulnerable to prevention of attacks, but vein
patterns are virtually unspoofable.

 

The technique of vein detection is based on identification of the
subcutaneous vein patterns in individuals. For doctors or nurses, the process
of finding the location of vein vessels is very important to help the patients.
Generally it is easier to find veins in young generation persons, but in
elderly persons it if difficult to find the veins only in visible light. The advantage
of identifying vein is that it does not require the direct contact with the device
and speed of operation is also very high. Vein detection is very important for person
identification, matching, recognition, in medical applications it is mandatory,
and it plays very important role in medical field. The appropriate
location or identification of vein vessels is mandatory for needle infusion in
the patients.

 

Sometimes it is very complicated to insert needle
in the first experiment because many of doctors or nurses find difficult to
locate vein vessels. For patients who suffer from heart disease it is important
to detect veins to provide them intravenous drugs immediately so that it will
help to dissolve clot. Number of trials for needle infusion cause wasting of
time and also it is uncomfortable for patients. For humans it is extremely
important to determine the accurate location of the various vein vessels
because if injecting a needle went wrong then person may suffer badly. While
injecting needle if any errors occurs then it may affects badly on patients
health. Therefor it is need to develop system which will give accuracy in
detecting the veins which contributes for curing patients.

 

Vein pattern recognition gives competition to most of the traditional
methods. It has several advantages corresponds to unique, portable, and
inherent properties compared to other biometrics features, like fingerprint,
retina, voice and face. Fingerprint scanner requires contact with the scanner
and it may possible that pattern can be weak because of other problems like
person age, dryness, disease etc. Retinal scanner is not popular because of its
intrusive nature also it is difficult for installation. Facial recognition has
a disadvantage that it cannot capture without persons knowledge, consent.
Disadvantage of voice recognition is voice changes with different moods,
illness and weakness so, it may cause problem while authentication. To ignore
all above problems we can go with the vein recognition because of its unique
properties those are – high anti counterfeit performance, high recognition
performance, easy acceptability.

 

In medical field to deal with the problems of
needle infusion, a single tool for identifying the veins are required which
will clearly shows the veins. In proposed system for vein detection   infrared lights is used because vein patterns
are not easily observable in visible light. Infrared lights will emit to the
body of patients. The reason why near infrared light is used is because the
light is absorb more strongly by human blood than the surrounding tissues. The
advantageous features of near infrared imaging are it can easily penetrate up
to 3mm into the tissue. When the light emits on the body, camera is set on the
part of the body where we want to detect the veins of that patient. These
results to veins will look darker than the other tissues. The aim is to detect
the veins only not arteries and infrared light absorption in blood vessels
depends on the oxidized level, that is deoxygenated blood absorb more light
than oxygenated blood.

This helps in detecting required veins accurately
and reducing the chances of faulty needle injecting by the doctor or nurse to
the patients. 

II.     LITERATURE
SURVEY

The appropriate location or detection of vein vessels gives
large contribution in biometrics like human identification, matching, in the
field of forensic and in medical applications. Other biometrics features having
several disadvantages which will not bother if we go for vein detection which
having its unique features. The appropriate location of veins plays very
important role in medical application for needle infusion for the purpose of
drug or medications delivery in patients. The biometrics gives a normalized,
efficient and discriminative representation of feature. There are multiple
types of biometric recognition.

 

Finger
print recognition

 

Pattern of finger print are
unique to each person. Even twins also have different finger print patterns.
This technique has some disadvantage also. Disadvantages of this system are age
of user, dryness and wetness of finger, temporary or permanent cuts on figure
can affect the performance of system because all this above things effect on
pattern hence give a poor pattern for recognition. User may go through the difficulties
in authentication process.

 

Face
recognition

 

There are various types of facial
scan those are by capturing an image of face by using cameras or by using
infrared patterns of facial heat emission. Facial recognition generally used
for identification instead of verification. It includes eyes, nose, lips, and
so on. The advantage of this technique is that it does not require touch of any
device. The disadvantage of this technique is that the quality of capture image
depends on lightening, background, and angle of camera. And another
disadvantage is appearance of person changes time by time.

 

Iris
Recognition

 

The iris is a color portion of
eye which surrounds the pupil and this portion of eye is used for recognition.
The iris pattern can be obtained through the video based images. This
technology can be used for identification and verification both. Advantage of
this recognition system is iris pattern does not change throughout the life and
it has smallest acceptance rate. But the position head and eye is very
important to get accurate results. And people should not move their head during
data acquisition. It requires a very high resolution of captured images because
iris is very small portion of eye.

 

Voice Recognition

This recognition uses different features of acoustics which
are different in individuals. There are various vocal qualities like frequency,
short time spectrum of dialogue and spectrograms and these are used for voice
recognition. In these type of system user selects the expression and repeat it
during identification and verification. This reduces the risk of guessing the
same phrase and entering into system. 
But noise and echoes can reduce the quality and accuracy of system.
Another disadvantage of this system is voice changes illness or in different
moods and it can cause authentication problem.

 

Insertion of
intravenous catheters plays an important role in any surgery. Physicians face
the major problem for accessing vein for IV drug delivery or taking blood
sample for test. It is important for children’s, adults, and critical patients.
Because of poor visibilities it may possible that unnecessary puncturing of
vein can be happen. And this may results in bruise and permanently damage of
veins. This system give an approach for IR vein detection system and it helps
in identifying veins on the back side of hand. IR LED ring is used for infrared
radiation. Smartphone camera is used to capture the image and its preprocessing
can be done by using MATLAB. Noise removal filter and adaptive histogram
equalization technique is used to enhance the image. This system only
implements the acquisition and segmentation of image 1.Vein detection
considered as one of latest biomedical techniques. Different difficulties found
in design and implementation of device regarding to image processing algorithms
and lighting system at low price. Specialists confront the issue while
distinguishing veins for intra-venous medication conveyance. A few issues can
be happens those are wounds, rashes, blood clump. Therefor framework needs to
create which will distinguish non-obtrusive subcutaneous veins. So modified
webcam is utilized for catching vein pictures and PC vision is utilized for
processing. This framework can be utilized for treatment of varicose veins, profound
vein thrombosis and vascular sicknesses. This system can be used for treatment
of varicose veins, deep vein thrombosis. This system has only worked on
acquisition and segmentation and it does not belong to IR imaging which is
stationary it encounters a lot of artifacts 2. Utilizing infrared radiation
for biometric applications is new and energizing idea that convey top of the
line applications in various fields including security and medication. While a
couple of devices have just been actualized, there is a solid requirement of
low cost sensing device. This exploration has been centered around a simple to
utilize, cost effective vein design recognition framework that utilizations
near infrared  radiation yet the
calculations and the equipment lighting and detecting head can effectively be
utilized as a part of different applications in the field of legal sciences,
human acknowledgment or in medicinal applications 3. For intravenous
medication and withdrawal of blood the intravenous access and surgical
perforation are invasive procedure in medical field every day. The numbers of process
which corresponds to vein detection are blood draws, peripheral catheter
insertion, intravenous therapies, etc. For intravenous drug delivery and other
medical situation doctor suffers from difficulty. Therefor device which can
clearly show vein is needed to develop. The system relate to development of
non-invasive subcutaneous vein detection. This system uses a CCD camera for
capturing image and compute software module. This system works on video but
under the particular ideal light conditions and therefor it needs external
adjustment or automatic software calibration. USB cable restrict transfer frame
rate. This system is able to visualise and detect veins from anterior side of
forearm 4. IR radiations for
biometrics provide a promising result in the field of security and biometrics
applications. The use of vein pattern for person identification is new approach
and it uses a vein network present under the skin. The vein pattern is unique
feature in each individual. The features of veins that is uniqueness and
stability can be in biometrics for identification of person. The method is to
detect the veins using an IR illumination then different techniques is used for
enhancement of vein image for identification of person. This work consists of
IR vein detection and matching of systems for identification of persons. The
novel scheme approach is used which capture image by an infrared illumination.
This system is used to detect veins on palm dorsal surface 5.

 

III.     Proposed
Method

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1: Block Diagram of Proposed Method

 

The block diagram of proposed method for vein
detection is as shown in figure.1. In proposed framework image is taken and region
of interest is selected from input image. The various image enhancement
techniques are available to improve the quality of image. Noise removal filter
are used to remove noise from input image. Adaptive thresholding and
morphological operations are used. Morphological operations reduce the unwanted noise from an image. Segmentation process is used for
image extraction.

 

 

 

Image capturing and Region
of interest

Image
is captured and interested region of interest is selected from input image.  

 

Image enhancement

To
improve the quality of image this image enhancement technique is used. There is
possibility of noise in input image so to remove noise from image different
filters can be used. In this system median filter is used improve quality of
image.

 

Generally
input image taken from human body have low contrast and to improve this
contrast adaptive histogram equalization is used.

 

Adaptive thresholding

To
apply single threshold on complete image intensity distribution and background pixels
should be sufficiently distinct. Adaptive threshold convert gray scale image into a binary image.

 

Morphological
operations

Morphological
open operations are used to remove noise from an image and it is used to estimate
image background. This process is helpful for selecting particular area.

 

After
morphological operations interested region will be selected. Output of this
will be a detected vein image which can be used for application.

IV.     Results

In vein detection system image
enhancement is the important part to detect veins in interested region. Image
enhancement is a process to remove the unwanted distortion and determine the
hidden information that is present in images. The performance of the image
enhancement techniques to detect vein are given.

 

Adaptive
Histogram Equalization

 

 

(a-1)

(b-1)

 

(a-2)

(b-2)

 

(a-3)

(b-3)

 

(a-4)

(b-4)

Figure.2 (a) Original and contrast
adjusted image and its histogram equalization contrast adjusted (b) contrast
enhance by Adaptive histogram equalization for all four images

Median
Filter

     

       
(a)                     (b)                               (c)

 

                        

 

        
(a)                     (b)                               (c)

 

  

        
(a)                     (b)                               (c)

 

  

        
(a)                     (b)                               (c)

 

 

Figure 3: Noise removal by
filtering (a) original image (b) noisy image (c) filter image left side result
of averaging filter and right side of medianfilter2

 

V.     Conclusion

Vein
lies underneath the skin of human body and to detect it successfully image
enhancement is important part. Capture images mostly have low contrast and
contain noise. For removal of noise and to improve quality, image enhancement
techniques that is adaptive histogram equalization and median filtering is   used.

 

References 

1 Anagha
B. Bawase, Pro. Dr. Mrs. S. D. Apte, “Infrared
Hand Vein Detection System,” IOSR Journal of Electronics and Communication Engineering
(IOSR-JECE), PP 48-52.

 

 2 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. 2013.

 

3 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.

 

4 Mayur
Wadhwani, Abhinandan Deepak Sharma, Aditi Pillai,” Vein Detection System using Infrared Light,” International Journal of Scientific & Engineering Research,
Vol. 6, Issue 12, Dec 2015.

 

5  Manjiree S. Waikar, and Dr.S. R. Gengaje, ” Infrared
Vein Detection System for Person Identification – An Image Processing Approach” International Research Journal of Engineering and
Technology (IRJET), vol.
4, no. 6, June 2017.