Value bill identifying method

Patent metadata

Field	Value
Publication number	US-10235595-B2
Application number	US-201615564936-A
Country	US
Kind code	B2
Filing date	Apr 6, 2016
Priority date	Apr 13, 2015
Publication date	Mar 19, 2019
Grant date	Mar 19, 2019

Abstract

Provided is a value bill identifying method, which includes: step 1, collecting, by a color collection device including multiple color sensors, color data of a to-be-detected value bill and preprocessing the collected color data; step 2, extracting a feature from the preprocessed color data; step 3, matching the extracted feature with feature template sets corresponding to each type of value bills, to obtain matching scores, and regarding a feature template with the highest score as a matched template of the color data; and step 4, determining a type of the value bill based on a matching result.

First claim

The invention claimed is:

1. A value bill identifying method, comprising:
step 1, collecting, by a color collection device comprising a plurality of color sensors, color data of a to-be-detected value bill and preprocessing the collected color data;
step 2, performing color space conversion on the preprocessed color data, to obtain hue data, detecting a stable sub-segment set in the hue data, wherein a stable sub-segment is a segment with smaller hue variation, calculating a hue mean of the stable sub-segment, wherein hue means of all sub-segments in the stable sub-segment set constitute a feature vector of the color data, wherein the feature vector of the color data which is a one-dimensional vector constitutes a feature extracted from the color data;
step 3, matching the extracted feature with feature template sets corresponding to each type of value bills, to obtain matching scores, and regarding a feature template with the highest score as a matched template of the color data, wherein a feature template set corresponding to one type of value bill is extracted from a true color bill image by following steps 01-07: step 01, dividing, based on complexity of image information, a color image in an orientation of the bill into a plurality of sub-regions; step 02, converting each of the divided sub-regions into color data by simulating an operation mode of a color sensor; step 03, performing color space conversion on the converted color data, to obtain hue data of the region; step 04, detecting a stable sub-segment in the hue data; step 05, calculating a hue mean of the stable sub-segment; step 06, constituting, by hue means of all stable sub-segments, a feature template corresponding to the region; and step 07, constituting, by feature templates corresponding to all the sub-regions in the orientation, a feature template set corresponding to the orientation of the bill, and constituting, by feature template sets corresponding to all orientations of the bill, the feature template set corresponding to the bill; and
step 4, determining a type of the value bill based on a matching result.

2. The value bill identifying method according to claim 1 , wherein before step 1, the value bill identifying method further comprises a step of pre-setting a plurality of the feature template sets corresponding to each type of value bills.

3. The value bill identifying method according to claim 1 , wherein in the step 03, simulated color data is converted into a Hue Saturation Lightness (HSL) color space, hue data SH of color data SS is obtained and expressed as follows:
SH={sh 0 ,sh 1 , . . . sh j , . . . sh L }(1< j<L ),
a converting method is described as follows:

sh
j

=

{

60
×

g
j

-

b
j

Max
-
Min

,

Max
=

r
j

60
×

b
j

-

r
j

Max
-
Min

+
120

,

Max
=

g
j

60
×

r
j

-

g
j

Max
-
Min

+
240

,

Max
=

b
j

.

4. The value bill identifying method according to claim 3 , wherein in the step 04, the positioning the stable sub-segment in the hue data comprises:
calculating an integral image, expressed as follows, of the hue data SH:
S MAP i ={s map 0 ,s map 1 , . . . s map j , . . . s map L }(1< j<L ),

wherein,

smap
j

=

{

sh
j

,

j
=

Assignees

• Grg Banking Equipment Co Ltd

Inventors

• Yue Xuyao
• Xiao Zhuming
• Wang Dandan
• Huang Xiaoqun

Classifications

• G07D7/2033 Primary

  Matching unique patterns, i.e. patterns that are unique to each individual paper · CPC title

• G06V10/56

  relating to colour · CPC title

• G06V10/76

  based on eigen-space representations, e.g. from pose or different illumination conditions; Shape manifolds · CPC title

• G06V10/50

  by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis · CPC title

• G06V10/143 Primary

  Sensing or illuminating at different wavelengths · CPC title