You are on page 1of 16

PALM VEIN DETECTION

FOR HUMAN IDENTIFICATION


USING CNN
ABSTRACT
• A person’s finger changes size or form over time and the finger print scanner does
not take this into consideration.
• When these changes occur, an individual can have difficulty identifying
themselves and gaining access.
• In this system palm vein detection is introduced for human identification.
• Palm vein is chosen instead of finger vein because in the case of a finger vein, the
surface area we are dealing with is much smaller than the palm, which makes it
more complex.
• This approach of palm vein detection for human identification is done using
convolutional neural network.
• Every one has a unique vein pattern in their palm, which can be used as an
identification and security technique. Infrared (IR) light is used to illuminate the
vein patterns.
INTRODUCTION
• Biometric based personal identification technology receives more attention.
• Biometric authentication technologies like fingerprint and face recognition
has many problems.
• We need an effective way to reduce the risk caused by fake biometric as
much as possible.
• Here a palm vein-based personal identification system is proposed.
• Palm vein patterns are unique for every individual.
• Palm vein is chosen instead of finger vein .
• This technique is much more preferred over finger print recognition due to
its greater level of security and convenience.
LITERATURE SURVEY
This paper explains about finger print and palm print based biometric technique
which is one of the most developed and easy to use biometric technique[1].
Here a method for personal identification based on finger-vein patterns is
proposed[2].
Vein patterns are different for each finger and for each person[3].
Biometrics is the science of recognizing an individual based on his or her
physical or behavioral traits[4].
Explains and defines all the elements and important issues related to CNN, and
how these elements work[5].
Describes new forms of histogram for image contrast enhancement[6].
Analyzes the contrast modification by three main methods: the histogram
equalization method, local histogram equalization (LHE) method and contrast
limited adaptive histogram equalization (CLAHE) method[7].
Presents two infrared imaging technologies, far-infrared thermography and
near-infrared imaging, to acquire hand vein pattern images for biometric
purposes[8].
Presents an infrared finger vein image acquisition terminal design method[9].
Evaluated some image processing algorithms in Tensor Flow ,which is a novel
framework for deep learning ,but also for image processing[10].
DESIGN METHODOLOGY
TRAINING PROCESS
Coding : Coding is done in Raspberry Pi to continuously capture images of
vein patterns and save them.
Image processing : Captured images are subjected to image processing using
histogram equalization technique( CLAHE) for enhancing the image contrast.
Training: After image processing , these images are used for training the
system.
Program execution: After completion of training, program for execution is
coded in Raspberry Pi in order to detect the vein pattern for human
identification.
IDENTIFICATION PROCESS
WORKING
• Input image is captured by the IR camera. This image will undergo image
processing and it is then compared with the saved images in the trained system.
• The percentage of closeness of the input image to the saved image is set during
training process.
• If the percentage of closeness is equal to or greater than pre-set value, then the
system identifies the person.
WORK PROGRESS


a) Original image b) CLAHE image


:
• Training was done using sample images.

COMPARISON:
• Input image was compared with already trained set of images
• Percentage of closeness is the output.
WORK PLAN

• •
• •
• •
• •
• •
• •
• •
• •
• •
• Human identification using vein patterns provides high security since the data
cannot be duplicated.
• This technique is much more preferred over finger print recognition due to its
greater level of security and convenience.
• As a biometric trait, palm vein pattern-based technology is highly effective for
personal identification.
• In future, vein detection can be used for medical purposes like locating the
patient's hidden veins for vaccinations.
[1] Finger and palm print based multibiometric authentication system with GUI interface” V. Vijayalakshmi ; R. Divya ; K. Jaganath
2013 International Conference on Communication and Signal Processing

[2] Naoto Miura, Akio Nagasaka, Takafumi Miyatake, "Feature Extraction of Finger Vein Patterns Based on Repeated Line Tracking
and Its Application to Personal Identification", Machine Vision and Applications, HITACHI, Ltd, 1-280, 2004.

[3] Junichi Hashimoto, "Finger Vein Authentication Technology and Its Future", Symposium on VLSI Circuits Digest of Technical
Papers, 2006.

[4] A. K. Jain, A. Ross, S. Pankanti, "Biometric: A Tool for Information Security", IEEE Trans. Information Forensics and Security,
vol. 1, no. 2, pp. 125-144, Jun. 2006.

[5] "Understanding of a Convolutional Neural Network" Saad Albawi , Tareq Abed Mohammed, Saad Al-Zawi 2017 International
Conference on Engineering and Technology (ICET).
[6] "Survey of contrast enhancement techniques based on histogram equalization" Manpreet Kaur, Jasdeep Kaur and

Jappreet Kaur, 2011 18th IEEE International Conference on Image Processing

[7] "A Study Of Changes In Image Contrast With Various Algorithms" Catalin Ionescu, Cristian Fosalau and Daniel

Petrisor 2014 International Conference and Exposition on Electrical and Power Engineering (EPE).

[8] "Near- and Far- Infrared Imaging for Vein Pattern Biometrics" Lingyu Wang and Graham Leedham 2006 IEEE

International Conference on Video and Signal Based Surveillance.

[9] "A design of infrared finger vein image acquisition terminal" Chen Liukui, Li Zuojin, Wu Ying and Xiang Yi
, 2011 International Conference on Business Management and Electronic Information.

[10] "Performance of Some Image Processing Algorithms in Tensorflow" Damir Demirović ,Emir Skejić ,Amira
Šerifović–Trbalić -2018 25th International Conference on Systems, Signals and Image Processing (IWSSIP)
THANK YOU

You might also like