Vous êtes sur la page 1sur 22

 Software and Hardware

 Introduction
 Offline Signature Verification
 Research Challenges
 Thesis Objective
 Overall Workflow of Signature Verification.
 Related Works
 Binary Image
 Thresholding
 ImageHash
 Convex Hull
 Aspect Ratio
 SVM
 Proposed Method
 Future Works
 Conclusion
 References
 python 3.6
 anaconda 3

 Intel Pentium and Celeron class processor


 Processor Speed – 1.2 GHz or above
 RAM - 512 MB
 HDD - 40 GB
 Monitor-14”SVGA
 Mouse- Normal
 Keyboard- Normal
 A signature is a person's name written in a distinctive way as a form of
identification in authorizing a check or document or concluding a letter.
Signature forgery refers to the act of falsely replicating the signature of
another person.
Signature forgery is done in order to -
 Commit frauds
 Deceive others
 Alter data etc.
One common example of signature forgery is cheque writing.
 Those forgeries can be verified in two
methods-
 Online Signature Verification
• Deals with dynamic features like
speed, pen pressure, directions,
stoke length and when the pen is
lifted from the paper
 Offline Signature Verification
• Uses features(static information)
from the image.
• Deals with shape only.
• Largely used for verifying bank
cheques.
 Differentiating different
parts of signature that
varies with each signing-

 Signature orientation can


be different-

 Threshold value should


be taken wisely so that
False accept and False
Reject occur very less.

 The nature and variety of


the writing pen
 Our main objective of this thesis is to develop a method
that will calculate features of a signature and verify it
comparing with sample prototype in spite of having-
• Different image size.
• Different orientation
• Various writing
 All the methods of signature verification undergoes
the following steps:
Feature is extracted
Features are classified
The system is trained
Matching
 Existing methods so far we studied can be
categorized by the following tree-

Convex Hull Contour area Bounded


Area rectangle Image Hash
area
Aspect
Ratio
 A binary image is a digital image that has only two possible values for
each pixel. Typically, the two colors used for a binary image are black and
white.
 Thresholding is the simplest method of image segmentation. From
a grayscale image, thresholding can be used to create binary images.
 If pixel value is greater than a threshold value, it is assigned
one value (may be white), else it is assigned another value
(may be black).
 Perceptual hash algorithms describe a
class of comparable hash functions.
 Features in the image are used to
generate a distinct (but not unique)
fingerprint.
 These fingerprints are comparable.
 Every perceptual hash algorithm has
the same basic properties:
 Images can be scaled larger or smaller.
Have different aspect ratios, and even
minor coloring differences (contrast,
brightness, etc.) and they will still
match similar images.
 The minimum bounding rectangle (MBR), is an expression of the
maximum extents of a 2-dimensional object (e.g. point, line, polygon) or
set of objects within its (or their) 2-D (x, y) coordinate system, in other
words min(x), max(x), min(y), max(y).
 The MBR is a 2-dimensional case of the minimum bounding box.
 Contour is an outline representing or bounding the
shape or form of something.
 Given a set of points in the plane,
the convex hull of the set is the
smallest convex polygon that
contains all the points of it.
 The area of the hull is calculated.
 The aspect ratio of an image describes the
proportional relationship between its width and its
height.
 Calculated for each image sample.
 In machine learning, support vector
machines are supervised
learning models with associated
learning algorithms that analyze data
used for classification and regression
analysis.
 SVM has a technique called
the kernel trick.
 These are functions which takes low
dimensional input space and
transform it to a higher dimensional
space.
 It is mostly useful in non-linear
separation problem.
Preprocessing
• Grayscale Image
•Thresholding to a Binary Image(Black And White)

Creating Feature descriptor


• Calculating aspect ratio of each sample.
• Calculating bounding rectangle area of each sample.
• Creating a hash number for each sample.
• Calculating hull_area/bounding_area of each sample.
• Calculating contour_area/bounding_area of each
sample.

Classify the descriptors using SVM classifier.

Compare the prototype with the testing signature.


 A high level of skilled forgery may pass the above test but
those can be further detected using the following tests:
• Edge thickness will be calculated to detect overwriting.
• Sudden blobs in the signature need to be detected.
• Straightness of the edges will be checked.
• Noise Detection in Image.

 Further it can be converted to online and real time system.


 https://www.researchgate.net/publication/34
57525_Automatic_singer_recognition_of_pop
ular_music_recordings_via_estimation_and_m
odeling_of_solo_vocal_signals

Vous aimerez peut-être aussi