0% found this document useful (0 votes)
33 views3 pages

Data Analysis Report of Banknote

This document analyzes data from images of genuine and forged banknotes to determine if machines can detect fake notes. It introduces two continuous variables, V1 and V2, measured from wavelet transformed images representing variance and skewness. Statistical measures and k-means clustering of V1 and V2 are used to differentiate genuine from forged banknotes. The analysis finds that genuine banknotes can be characterized by V1 and V2 values near 0.459 and 1.2 respectively, allowing machines to potentially detect forgeries with higher variance and skewness. It recommends using these values as a standard and collecting more genuine banknote data to improve tolerance thresholds.

Uploaded by

NurujjamanKhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
33 views3 pages

Data Analysis Report of Banknote

This document analyzes data from images of genuine and forged banknotes to determine if machines can detect fake notes. It introduces two continuous variables, V1 and V2, measured from wavelet transformed images representing variance and skewness. Statistical measures and k-means clustering of V1 and V2 are used to differentiate genuine from forged banknotes. The analysis finds that genuine banknotes can be characterized by V1 and V2 values near 0.459 and 1.2 respectively, allowing machines to potentially detect forgeries with higher variance and skewness. It recommends using these values as a standard and collecting more genuine banknote data to improve tolerance thresholds.

Uploaded by

NurujjamanKhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Data Analysis report of banknote-authentication

Two sets of data were collected by measure the images taken from
genuine and forged banknote-like specimens.

These data are compared in order to determine whether can a


machine detect fake notes

Table of contents
1. Introduction of Data
2. Statistical measure of V1 & V2
3. Graph of V1 vs V2
4. K-Mean clustering
5. Summary
6. Recommendation

1. Introduction of Data
V1 = variance of Wavelet Transformed image (continuous)

V2 = skewness of Wavelet Transformed image (continuous)


2. Statistical Measure of V1 & V2

3. Graph of V1 vs V2

4. K-Mean clustering

5. Summary
Mean Median Standard deviation
X 0.434 0.496 2.842
Y 1.922 2.32 5.867

K - mean Value
X Y
Array 1 0.459 1.2
Array 2 1.231 8.352
Array 3 -0.932 -6.944

1. High value of standard deviation (2.842, 5.867) indicates that the data points are
spread out over a large range of values.
2. Array 1 with value (0.459, 1.2) is near to X&Y mean (0.434, 1.922) & X&Y
median (0.496, 2.32)
3. Since K- mean clustering = a collection of data points aggregated together because
of certain similarities.
4. In order to differentiate forged bank note from genuine bank note, we can pick
value of array 1 (0.459, 1.2) as standard. Bank note with variance (V1) and skewness
(V2) more than 0.459 and 1.2 can consider as forged bank note.

6. Recommendation
1. Array 1 (0.459, 1.2) can be set as a standard for genuine bank note.
2. Suggest client to collect more data on variance (V1) and skewness (V2) of genuine
bank note in order to determine the tolerance of the variance and skewness.

You might also like