Method and system for recognition of currency by denomination

• US 6,078,683 A
• Filed: 11/20/1997
• Issued: 06/20/2000
• Est. Priority Date: 11/20/1997
• Status: Expired due to Fees

• Current Assignee: Talaris Holdings Limited (Glory Limited)
• Original Assignee: De la Rue Inc (Glory Limited)
• Inventors: Mikkelsen, John, DeFeo, Michael, Truong, Peter, Denison, Jack, Burgert, Robert J.
• Primary Examiner(s): Au, Amelia
• Assistant Examiner(s): Dastouri, Mehrdad
• Application Number:
  US08/975,475
• Time in Patent Office: 943 Days
• Field of Search: 382/135, 382/156, 382/157, 382/158, 382/205, 382/289, 209/534, 364/148.03, 701/59
• US Class Current: 382/135
• CPC Class Codes: G06V 30/1478 of characters or characters lines
• CPC Class Codes: G07D 7/12 Visible light, infrared or ultraviolet radiation
• CPC Class Codes: G07D 7/17 Apparatus characterised by positioning means or by means responsive to positioning

Abstract

A denomination recognition apparatus and method uses a programmed microelectronic CPU (14) to execute program instructions (19) stored in PROM (18) to read in pixel data from an optical imaging section including LEDs (15) and photodiodes (16) for generating signals which can be converted to a first image of a currency note (13) being transported along the path of travel. The CPU (14) receives position and skew data detected by external sensors for sensing the position and skew of the note (13). The CPU (14) executes the program instructions, first to assemble a first data image of pixel data representing a two-dimensional image of substantially all of the note while averaging the pixel data in groups for further processing; second, to correct the first data image based on the skew data to eliminate skew; and third, to relate the first data image to a plurality of pre-defined data images according to a mathematical function to determine whether the first data image can be classified as matching any one of the pre-defined data images corresponding to a specific currency denomination.

Claims

1. A method of detecting the denomination of a currency note transported along a path of travel, the method comprising:

receiving position and skew data detected by external sensors for sensing the position and skew of the note;

assembling a first data image of pixel data representing a two-dimensional image of substantially all of the note while averaging the pixel data in groups for further processing;

adjusting the first data image to account for anomalies in the pixel data in said first data image;

correcting the first data image based on the skew data to eliminate skew; and

relating the first data image to a plurality of pre-defined data images according to a mathematical function to determine whether the first data image can be classified as matching any one of the pre-defined plurality of data images corresponding to a specific currency denominations;

wherein the adjusting step includes testing each pixel to detect absence of note material, and when an absence of note material is detected, the pixel data is set to a neutral value to identify the absence of note material in order to facilitate comparison of the first data image to each of the plurality of pre-defined data images to determine the specific currency denomination.

2. The method of claim 1, wherein the first data image and the plurality of pre-defined data images comprise big pixels, each of which is at least a 3×

3 array of small pixels.

3. The method of claim 1, wherein the relating step utilizes a neural network function to compare the first data image to each of the plurality of pre-defined data images to determine a denomination of the note.

4. The method of claim 3, wherein a relative entropy is calculated for the denomination resulting from application of the neural network function, and wherein a comparison is made between the relative entropy and a minimum threshold to determine if a tentative match determined by the neural network function is an acceptable match.

5. The method of claim 1, wherein the correction step includes testing for positive skew or negative skew, and wherein upon detecting positive skew, the data for the first image is assembled by scanning the note from left to right, and wherein upon detecting negative skew of the note, the data image for the first data image is assembled by right to left scanning of the note.

6. The method of claim 1, further comprising the step of adjusting light emitters and light detectors prior to assembling a first data image, wherein electrical power to the light emitters is adjusted and power to amplifiers for transmitting signals from the light detectors are adjusted, in response to light transmissivity scaling factors, such that the light emitters and light detectors are calibrated to operate in a predetermined range of operation.

7. An apparatus for detecting the denomination of a currency note transported along a path of travel, the apparatus comprising:

an imaging section for generating signals which can be converted to image data for assembling a first image of the note when transported along the path of travel;

an interface for receiving position data and skew data detected by external sensors for sensing the position and skew of the note;

a processor connected for receiving the position data and the skew data from the interface and for executing a first group of program instructions for reading and assembling a first data image of pixel data representing a two-dimensional image of substantially all of the note while averaging the pixel data in groups for further processing;

wherein the processor executes a second group of program instructions for correcting the first data image based on the skew data to eliminate skew;

wherein the processor executes a third group of program instructions for adjusting the first data image to account for anomalies in said first data image, the third group of program instructions being executed to test each pixel to detect absence of note material, and when an absence of note material is detected, the pixel data is set to a neutral value to identify the absence of note material in order to facilitate comparison of the first data image to each of a plurality of pre-defined data images to determine a specific currency denomination; and

wherein the processor executes a fourth group of program instructions for relating the first data image to the plurality of pre-defined data images according to a mathematical function to determine whether the first data image can be classified as matching any one of the plurality of pre-defined data images corresponding to a specific currency denomination.

8. The apparatus of claim 7, wherein the first data image and the plurality of pre-defined data images comprise big pixels, each of which is a 3×

3 array of small pixels.

9. The apparatus of claim 7, wherein the fourth group of program instructions utilizes a neural network function to compare the first data image to each of the plurality of pre-defined data images to determine a denomination of the note.

10. The apparatus of claim 9, wherein the fourth group of program instructions is executed to calculate a relative entropy for the denomination resulting from application of the neural network function, and wherein a comparison is made by the processor between the relative entropy and a minimum threshold to determine if a tentative match determined by the neural network function is an acceptable match.

11. The apparatus of claim 7, wherein the processor executes a fifth group of program instructions to test for positive skew or negative skew, and wherein upon detecting positive skew, the processor executes further instructions for assembling the data for the first image by scanning the note from left to right, and wherein upon detecting negative skew of the note, the processor executes further instructions for assembling the data for the first image by scanning of the note from right to left.

12. The apparatus of claim 7, wherein the imaging section includes light emitters and light detectors, and wherein the processor executes a sixth group of program instructions to adjust power to the light emitters and to adjust power of the signals from the light detectors prior to assembling a first data image, in response to light transmissivity scaling factors, such that the light emitters and light detectors are calibrated to operate in a predetermined range of operation.