Worn banknote identification method and ATM using the same

Patent metadata

Field	Value
Publication number	US-10318802-B2
Application number	US-201715657908-A
Country	US
Kind code	B2
Filing date	Jul 24, 2017
Priority date	Jul 27, 2016
Publication date	Jun 11, 2019
Grant date	Jun 11, 2019

Abstract

The present disclosure relates to a worn banknote identification method and an ATM (Automated Teller Machine) using the same, and more particularly, to a worn banknote identification method which acquires an image of a banknote inserted into an ATM, divides pixels of the acquired banknote image into a bright region and dark region depending on a brightness distribution of the banknote image, determines the wear level of the banknote by comparing a difference between the average brightness values of the two regions to a preset reference value, and separately stores the inserted banknote into a reject box depending on the wear level of the banknote, and an ATM using the same.

First claim

What is claimed is:

1. A worn banknote identification method which determines the wear level of a banknote inserted into an ATM using an image of the banknote, comprising:
acquiring an image of the banknote through an image sensor;
calculating brightness values of pixels in the acquired banknote image by converting the banknote image into gray scales;
extracting a threshold value for determining the variance of the brightness values of the pixels in the banknote image;
dividing the pixels of the banknote image into a bright region and a dark region, based on the extracted threshold value;
calculating a difference between the average brightness values of the bright region and the dark region by calculating the average brightness value of the bright region and the average brightness value of the dark region; and
determining whether the banknote is a worn banknote, by comparing the calculated difference to a preset reference value, wherein a minimum wear level required to determine whether the banknote is recyclable is preset through a previous scan operation and stored as the reference value, and the banknote is determined to worn when the calculated difference is equal to or less than the reference value.

2. The worn banknote identification method of claim 1 , wherein the calculating of the brightness values of the pixels in the banknote image comprises extracting only brightness signals by removing color signals from the image of the inserted banknote.

3. The worn banknote identification method of claim 1 , wherein the calculating of the brightness values of the pixels in the banknote image comprises extracting only brightness values of green data from the image of the inserted banknote.

4. The worn banknote identification method of claim 1 , wherein the extracting of the threshold value comprises extracting the threshold value using the Otsu thresholding algorithm.

5. The worn banknote identification method of claim 4 , wherein the extracting of the threshold value using the Otsu thresholding algorithm comprises finding a valley in a distribution graph of the brightness values of the pixels and extracting the valley as the threshold value.

6. The worn banknote identification method of claim 1 , wherein in the calculating of the difference between the average brightness values of the two regions, Equation 2 below is applied:

δ
t

=

∑
t

L
-
1

⁢

i
⁢

⁢

n
i

/

∑
t

L
-
1

⁢

n
i

-

∑
0

t
-
1

⁢

i
⁢

⁢

n
i

/

∑
0

t
-
1

⁢

n
i

[

Equation
⁢

⁢
2

]

where δ t represents the difference between the average brightness values of the bright region and the dark region, L represents the number of brightness value levels, t represents the threshold value, i represents a brightness value, and n i represents the number of pixels having the brightness value i.

Assignees

• Nautilus Hyosung Inc
• Univ Yonsei Iacf
• Hyosung Tns Inc

Inventors

• Choi Eui Sun
• Baek Yoon Kill
• Baek Sang Wook
• Lee Sang Hun
• Lee Chul Hee

Classifications

• G07F19/202 Primary

  Depositing operations within ATMs (depositing in general G07D11/00) · CPC title

• G06V30/40

  Document-oriented image-based pattern recognition · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G07D7/187 Primary

  Detecting defacement or contamination, e.g. dirt · CPC title

• G07D7/2075

  Setting acceptance levels or parameters · CPC title