International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015

ISSN 2229-5518
780

Vein Detection System using Infrared Light

Mayur Wadhwani, Abhinandan Deepak Sharma, Aditi Pillai, Nikita Pisal, Dr. Mita Bhowmick

Abstract – The process of obtaining intravenous (IV) access, Venipuncture, is an everyday invasive procedure in medical settings
and there are more than one billion venipuncture related procedures like blood draws, peripheral catheter insertions, intraven-
ous therapies, etc. performed per year [3]. Excessive venipunctures are both time and resource consuming events causing anxie-
ty, pain and distress in patients, or can lead to severe harmful injuries [8]. The major problem faced by the doctors today is diffi-
culty in accessing veins for intra-venous drug delivery & other medical situations [3]. There is a need to develop vein detection
devices which can clearly show veins. This project deals with the design development of non-invasive subcutaneous vein detec-
tion system and is implemented based on near infrared imaging and interfaced to a laptop to make it portable. A customized
CCD camera is used for capturing the vein images and Computer Software modules (MATLAB & LabVIEW) is used for the
processing [3].

Index Terms— Venipuncture, Near-Infrared imaging, vein detection system, MATLAB, LabVIEW, Webcam, Vein Detection
——————————  ——————————

IJSER
1 INTRODUCTION

A
ccessing veins in cases of elderly or obese as well cause the scarring of the skin. The appearance of
as dark toned & adult patients becomes very diffi- the skin becomes deterred making the skin to ap-
cult sometimes for drug delivery physicians. Al- pear either whiter or darker. The determinations of
though significant work has been done in this area veins become tough in such cases.
and many devices have come up, but the major
problem lies in their cost and portability. A low (B) Intravenous injections: For giving medicines and
cost, portable & efficient infrared imaging detec- drugs to the patients, intra- venous injections are
tion system is the need of the hour. given by doctors and nurses.

Burns and other physical injuries make it difficult (C)Amongst children: Having to puncture them sev-
eral times with a needle is very frightful and ago-
to locate veins and administer lifesaving drugs. In
nizing for the child & so locating veins in young
such cases it becomes very necessary to have a de-
children and infants may be difficult.
vice that detects the exact location of required vein.
Also in case of blood transfusion or withdrawal,
(D) Blood transfusions: It is a process in which blood
etc. it is necessary to know the position of the
is given to the person intravenously. Blood dona-
veins. Even trained nurses and doctors many times
tion, kidney dialysis also need perfect vein detec-
find it difficult to exactly locate the blood veins on
tion.
the first attempt itself. In some medical situations,
the location of vein needs to be identified. Each (E) Geriatrics: Old people often require numerous
second counts when the doctors are treating trau- blood tests or medicinal injections and an efficient
ma patients. The other situations where vein imag- means of puncture would reduce excessive bruise
ing is required are: and enhance the patients overall comfort level.

(A) Bruises and Bums: In case of vein diseases Human eyes can only detect visible light that oc-
bruises appear on the skin like Deep Vein Throm- cupies a very narrow band (400 - 700nm) of the
bosis and Varicose Veins, therefore for the treat- spectrum. However, in other bands of the electro-
ment detection of veins is highly essential. Acci- magnetic spectrum there is much more informa-
dents involving first or second degree of bums tion contained rejected by the objects of interest.
IJSER © 2015
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015
ISSN 2229-5518
781

The visibility under normal visible light conditions Infrared (NIR). Because of its certain attributes as
is very low for human vein patterns on the peri- compared to FIR, NIR gives better results for vein
phery. This can be resolved by Near-Infrared (NIR) detection. Far infrared imaging means red hot im-
imaging techniques. The special properties of age, that recognition rate often greatly influenced
Near-Infrared imaging are: by humidity and temperature, NIR penetrates
most human tissue easily, so it can acquire more
Depth penetration of up to 3mm into biological clear and reliable image quality than FIR.
tissue using NIR. More absorption of radiation by
venous as compared to surrounding tissues. The The designed vein detection takes a snapshot of the
vein image maybe captured by an IR sensitive subject's veins under a source of infrared radiation.
camera, therefore, by shooting the infrared radia- Almost any part of the body could be analyzed in
tion of specific wavelength at the desired body part this method. In many medical practices, X-ray and
results the veins to appear darker than the sur- ultrasonic scanning can also be used to form vein
rounding tissue in the image. images. They are all invasive techniques as they
require injection of agents in blood stream. This is
A spectral window exists from 700 to 900 nm not feasible for general purpose imaging applica-
where light can penetrate deep into the tissues. The tions. The key challenge in a vein pattern biometric
wavelength of the Infrared (IR) light beam coming system obtaining the vein pattern images in a fast

IJSER
out from a light source is selected to be around and non-invasive manner.
850nm. It also avoids undesirable interference from
the radiation (3um – 14um) of human body.

The basic phenomenon is radiation of the wave- 2 DESIGN

length region 740 nm-760nm is to detect veins but
not arteries because of selective absorption of NIR ring of eight LEDs were used for illuminating
infrared radiation in blood vessels. The reason is the desired body part with infrared light. The de-
the deoxidized haemoglobin [deoxy-Hb or Hb] in sign of the light source should provide perfect il-
the veins almost completely absorb the radiation lumination so that the vein images can be captured
while the oxidized haemoglobin [HbO] become and there should be a contrast between the veins
almost transparent. and the surrounding tissue. So a ring of LED is
chosen as the camera lens is circular, making the
centre as webcam lens. The experiments conducted
had shown that near infrared LED array provided
illumination but it had a ringing effect light had
alternate light and dark bands. So, to correct this,
each LED was diffused by using sandpaper, and
rinsed by water. The result was an almost uniform
intensity of illumination. Also, NIR vein imaging
technique does not depend upon the skin color and
pigmentation of the person and it does not inter-
fere with the imaging process. In case of people
with several tattoos due to which the radiation
Figure 1. Absorption of light by blood were not able to pass through the skin.

There are two types: Far-Infrared (FIR) & Near The camera should have sufficient spatial resolu-

IJSER © 2015
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015
ISSN 2229-5518
782

tion so as to identify the vein details. The CCD is nected visible LEDs. The LEDs were soldered into
perfectly capable of detecting NIR up to a wave the board, they were diffused & tested. In addition,
length of approximately 1mm but all modern cam- the filter was replaced with a freshly developed
eras have an infrared cut-off filter in front of the filter especially for this purpose. The result of this
sensor since the main purpose of the camera is to was an illumination less intense than the LED grid
see the maximum amount of visible radiation. This & enabled us to take close up images of the fo-
filter was eliminated in order to gain access to the rearm.
infrared part of the radiation spectrum for the de-
tection system.

Figure 6: Final Modification of camera

IJSER
The designed subcutaneous vein detection device used
USB-based customized webcam .The output of the web-
cam was fed to computer. The images obtained were of
640X480 8-bit dimensions. MATLAB was used for cod-
ing. The following figure gives the outline of the algo-
rithm.

Figure 7 Software Algorithm

Figure 2-5 from L-R, Assembly, LED output. Ring Effecy. Diffused
LED output images

The main goal of the project was to make a porta-

ble and cost effective device. Visible LEDs inbuilt
in the webcam were replaced with Infrared LEDs.
The added benefit was that batteries were redun-
dant, as the LEDs were driving power from the IC
board of the webcam itself. The LEDs were sol-
dered in the same position as the previously con-

IJSER © 2015
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015
ISSN 2229-5518
783

2.1 Image Acquisition and Preprocessing (c )LabVIEW Implementation

The image of the arm was captured using an ordi- For obtaining real time images LabVIEW (Labora-
nary webcam that has been modified to only allow tory Virtual Instrument Engineering Workbench)
infrared light to reach the image detectors. After which is a system-design platform and develop-
taking the image, preprocessing was applied to the ment environment for a visual programming lan-
image. The purpose of this step was to improve the guage from National Instruments was used.
imaging quality so that vein patterns can be more
easily detected during the segmentation. This was NI Vision Acquisition Software is driver software
done by first cropping the image to isolate the ROI, for displaying, acquiring, monitoring images, &
applying filters to reduce noise and enhance the logging from a multitude of camera types. This

contrast software is includes all NI vision hardware and all

NI vision software licenses. NI Vision Acquisition
Software is sold separately for security and moni-
2.2 Segmentation and Post-processing toring applications not requiring image processing
or specialized hardware. The special block in Lab-
Since the noise had been reduced and the contrast VIEW is crucial in the working of our project. The
enhanced, segmentation permitted to separate the block selects automatically the connected camera

IJSER
vein pattern from the background. The vein pat- thereby removing the need for an interfacing code
tern was located and isolated from the rest of the between camera and the software. It also gives a
image. This was the most crucial step in the entire variety of configuration options associated with
recognition. If the veins are not properly detected, the webcam like resolution, frame rate of the web-
the errors increases. cam. Use NI-IMAQ to acquire from analog, parallel
digital, NI Smart Cameras & Camera Link. Vision
acquisition made it easy to perform filtering,
grayscale conversion and allowing frame by frame
viewing.

The preprocessing step served two main purposes.

The first was noise removal & smoothing. Since the
images were captured using a modified webcam,
considerable noise were present in the images.
Gaussian and median filters were used to remedy
the effect of this noise. The second was contrast
enhancement, which was necessary as the vein
pattern was faint.

A Gaussian filter is a smoothing filter based on the

Figure 8-11:From L-R: Thresholded Images, Equalized Images
Gaussian distribution. It is suitable because it acts
as a low pass filter, attenuating high frequency
The resulted segmented grayscale images were
noise while leaving the lower frequency features
with some unwanted information such as noise,
unchanged. Another source of noise in the images
shadows and faint veins. Therefore it was not al-
ways a true representation of the actual vein pat- was hairs that show up as very thin dark lines. A
tern. way to remove these was is to use a median filter.
The use of IR image capture makes the veins stand

IJSER © 2015
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015
ISSN 2229-5518
784

out, it is often necessary to further improve the

contrast before segmenting the image.

Global threshold separates the foreground and the

background pixels. Global threshold is better than
other techniques because it uses bimodal histo-
gram and calculates the inter-variance between
them. Adaptive histogram splits the image into
different tiles of 8x8 pixels and finds the histogram
in each tile to give a better contrast.

3. RESULT

The following real-time video snapshots were ob-

tained from the designed system. The videos ob-
tained provided a very vein pattern, however they
are not perfect. The snapshots obtained are ob-
tained in a contained environment with ideal light-

IJSER
ing conditions. Thus there’s a need for external
adjustment or automatic software calibration. Plus,
the USB cable restricts the frame rate of the transfer
but it is suitable for diurnal purposes.

Figure 12-15: Real Time Video Snapshots

IJSER © 2015
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015
ISSN 2229-5518
785

4. CONCLUSION Image Processing, 39,355-368(1987)

This paper investigates near infrared techniques

8. Simon Juric and Borut Zalik,” An innovative approach to
for vein imaging. The stand-alone, portable NIR near-infrared spectroscopy using a standard mobile device
vein detection system was able to visualize and and its clinical application in the real-time visualization of
detect vessels from anterior forearm. Various im- peripheral veins” BMC

age algorithms were tried on still images the histo- Medical Informatics and Decision Making 14(100), 2014

gram equalization worked best on MATLAB. 9. R. Miyake “Vein Imagine: A new method of NIR,
LABVIEW real time processing was implemented Where a Processed Image Is Projected onto the Skin for the
and samples obtained. Taking sample images from Enhancement of Vein Treatment”,
people of different skin complexities and muscle ISSN,32,1031-1038(2006)

built, the desired image output were successfully

10. Ravi Kumar A.V, Dr. Nataraj K.R, Dr.Rekha
obtained. Since we have made portable IR imaging
K.R, “Morphological Real Time Video Edge Detection
due to that we have encountered motion artifacts in Labview”, IJCSIT, 3(2), 3808-3811(2012)
issues. Therefore, our main goal of obtaining a
11. Thiusius Rajeeth Savarimuthu ,Anders Kjær-
portable efficient vein imaging system at a very
Nielsen, Anders Stengaard Sørensen, “Real-time
low cost is accomplished. medical video processing, enabled by hardware acce-
lerated correlations”, J Real-Time Image Proc, DOI
10.1007/s11554-010-0185-2(2010)

IJSER
5. CONCLUSION
12. Neethu M. Sasi, V. K. Jayasree, “
Contrast Limited Adaptive Histogram Equalization for
Qualitative Enhancement of Myocardial Perfusion Im-
1. F. Wang, A. Behrooz , M. Morris, A. Adibi, “High-
ages” Engineering, 2013, 5, 326-331
contrast subcutaneous vein detection and localization using
multispectral imaging” Biooptic. SPIE. 18(5), 050504-1 (2013).
13. Ajay P. Dhawale, Prof. S. R. Hirekhan, “Real-
Time Image Processing for Biological Applications
2. V. Gaikwad, S. Pardeshi,” Vein detection using infrared Through Morphological Operations using Lab-
imaging system” ISSN (Online):, 2(3), 2347-2820(2014) VIEW”, IJERT, 3(5), ISSN: 2278-0181(2014)

3. N. Limbad, Prof. G. D. Parmar, “Vein Pattern

Detection System Using Cost-effective Modified IR
Sensitive Webcam”ISSN (Online),1(9), 2347 – 4718(2014)

4. TanushriChakravorty,D.NSonawane,
Suchakrapani Datt Sharma, Tushar Patil, “Low-Cost Subcu-
taneous Vein Detection System using ARM9 based Single
Board Computer” IEEE, 3(11), 978-1-4244-8679 (2011)

5. Wang Lingyu and Graham Leedham, “Near- and Far-

Infrared Imaging for Vein Pattern Biometrics,”
Proceedings of the IEEE International Conference on
Video and Signal Based Surveillance (AVSS’06)

6. Septimiu Crisan, Joan Gavril Tarnovan, and Titus Eduard

Crisan, “A Low Cost Vein Detection System Using Near
Infrared Radiation ,” SAS 2007 - IEEE Sensors Applications
Symposium.

7. S. Pizer, “Adaptive Histogram Equilizations and its Varia-

tions”, Computer Visions, Graphics, and

IJSER © 2015
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 6, Issue 12, December-2015
ISSN 2229-5518
786

IJSER

IJSER © 2015
http://www.ijser.org