Image Processing Device Specular Reflection Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Character recognition accuracy is compromised by high luminance regions caused by specularly reflected light, which can lead to noise generation and decreased performance in binarization and recognition processes, especially in images with embossed characters like number plates.

Innovation Solution

An image processing device and method that identifies high luminance regions through histogram analysis and clustering, converting pixel values to suppress the influence of specularly reflected light by setting a threshold and replacing high luminance values, thereby improving the accuracy of character recognition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If dynamic binarization is performed to handle partial shadows, then binarization adaptability is improved, but high luminance regions from specular reflection cause noise and reduce character recognition accuracy

Engineering Contradiction:
Improvebinarization adaptabilityVSAvoidcharacter recognition accuracy
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies preliminary luminance correction to the input image before performing binarization and character recognition. By detecting high luminance regions caused by specular reflection and correcting them in advance, the method prevents noise generation in subsequent processing stages, thereby maintaining both binarization adaptability and character recognition accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary luminance correction process between image capture and binarization. This intermediate step acts as a mediator that transforms the input image by suppressing specular reflection effects, enabling both dynamic binarization and accurate character recognition to function properly without interference from high luminance regions

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If specularly reflected light is captured, then image completeness is improved, but luminance saturation creates high luminance regions that degrade character cutout and recognition accuracy

Engineering Contradiction:
Improveimage completenessVSAvoidcharacter recognition accuracy
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The patent converts the harmful effect of specularly reflected light into a beneficial process by detecting high luminance regions and applying targeted luminance correction. Instead of discarding the captured image data, the method utilizes the luminance information to identify and correct problematic regions, thereby preserving image completeness while eliminating the negative impact on character recognition accuracy

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentEP3214579B1Image processing device, character recognition device, image processing method, and program
Publication Date: 2022.11.16 OMRON CORP
  • EP3214579B1 patent drawingFigure 1
  • EP3214579B1 patent drawingFigure 2(a)~2(b)
  • EP3214579B1 patent drawingFigure 3

AI summary

An image processing device for performing a preprocessing of an image recognition process to an input image is provided, which includes: a generation element for generating a histogram of luminance values of the input image; a determination element for determining a reference value for the luminance values on the basis of the histogram, and determining a high luminance pixel that is a pixel having a luminance value greater than the reference value; and a conversion element for converting the luminance value of the high luminance pixel into a luminance value that is lower than or equal to the reference value. Thus, even when a high luminance region is present in the input image due to specularly reflected light or the like, high-accuracy character recognition can be performed.