Banknote Recognition Using Lab Color Space

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional banknote recognition systems based on real-time image processing face instability due to distortion and deformation, leading to low recognition rates, especially since they rely on grayscale images that fail to effectively utilize color information.

Innovation Solution

A banknote classification and recognition method utilizing the Lab color space, where RGB images are converted to Lab color space values (Gray a and Gray b), and these values are used in preset formulas to determine banknote denominations, improving recognition accuracy by leveraging color differences between denominations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If geometric features (length and width) are used for banknote recognition, then the recognition process can be performed in real-time, but distortion and deformation occur leading to instability and decreased recognition rate

Engineering Contradiction:
Improvereal-time recognition speedVSAvoidrecognition rate
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent changes the recognition parameters from geometric features (length, width) to color space features (Lab color values). By transforming the feature extraction approach to use color information that is invariant to geometric distortion, the system maintains real-time processing capability while significantly improving recognition stability and accuracy.

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If gray difference in single channel gray image is used for recognition, then processing is simpler, but the gray feature difference between some denominations is not significant leading to misrecognition

Engineering Contradiction:
Improveprocessing complexityVSAvoiddenomination recognition accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent transforms the recognition parameters from single-channel gray values to multi-dimensional Lab color space values. This parameter transformation enables the system to capture significant color differences between denominations while maintaining computational efficiency, thereby improving measurement precision without substantially increasing complexity.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent transitions from one-dimensional gray scale recognition to three-dimensional Lab color space recognition (L, a, b channels). This dimensional expansion provides richer feature information for differentiation, allowing the system to distinguish between denominations with similar gray values by utilizing color channel variations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Productivity

If single channel grayscale image is used, then the processing is simpler and faster, but color information cannot be effectively utilized resulting in low denomination recognition rate

Engineering Contradiction:
Improveprocessing speedVSAvoiddenomination recognition accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent changes the image representation parameters from grayscale to color space (Lab) while maintaining efficient processing. By using color values that can be derived from standard RGB images and transformed to Lab space, the system effectively utilizes color information for denomination recognition without sacrificing processing speed.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3249618B1Banknote classification and identification method and device based on lab color space
Publication Date: 2020.08.05 GRG BANKING EQUIPMENT CO LTD
  • EP3249618B1 patent drawingFigure 1
  • EP3249618B1 patent drawingFigure 2
  • EP3249618B1 patent drawingFigure 3

AI summary

A banknote classification and identification method and device based on a Lab color space, which are used for increasing the recognition rate of face values. The method comprises: acquiring an RGB image of a banknote to be detected; calculating gray values Gray R, Gray G and Gray B of the RGB image in red (R), green (G) and blue (B) wavebands; converting Gray R, Gray G and Gray B into gray values Gray a and Gray b in a Lab color space; and comparing the product of Gray a and Gray b and the difference between the absolute value of Gray a and the absolute value of Gray b with a preset formula to obtain the category of the banknote to be detected.