Bezier-Exemplar Hybrid based Image Inpainting Method in Computerized Colorimetric Protein Estimation System

During the quality assurance process of latex gloves, the modified Lowry method is considered as one of the standard methods to determine the protein concentration for quality assurance. Nevertheless, this traditional method requires several tedious chemical processes, which results in time-consumin...

Full description

Saved in:
Bibliographic Details
Published inIAENG international journal of computer science Vol. 49; no. 2; p. 463
Main Authors Sim, K S, Toa, C K, Lim, Z Y, Kathy, L C Y
Format Journal Article
LanguageEnglish
Published Hong Kong International Association of Engineers 20.05.2022
Subjects
Accuracy
Algorithms
Chemical reactions
Chemical tests
Colorimetry
Computer vision
Digital imaging
Gloves
Image processing
Image restoration
Latex
Proteins
Quality assurance
Quality control
Online AccessGet full text

Cover

More Information
Summary:During the quality assurance process of latex gloves, the modified Lowry method is considered as one of the standard methods to determine the protein concentration for quality assurance. Nevertheless, this traditional method requires several tedious chemical processes, which results in time-consuming. To overcome this weakness, our previous work has implemented a new method based on image processing namely Computerized Colorimetric Protein Estimation Method (CCPE). This method provides fast and efficient in determining the protein concentration with an uncomplicated chemical test and the aid of computer vision technology. In the CCPE method, there will be a section where a scanner will be used to scan glove samples and convert samples into a digital image. Afterward, the digitalized glove sample image will go through computerized algorithms to estimate the protein concentration. However, during the scanning process, wrinkles on the image will affect the accuracy of the estimation results. Thus, this paper presents our implemented Bezier-exemplar Hybrid based Image Inpainting (BeHbII) method that is used to eliminate out the wrinkle of the image efficiently. The results shows that the inpainting method can restore the wrinkled image up to 99% similarity based on the unwrinkled original image of the same glove model and improve the accuracy of protein concentration detection up to 96%. This method does not limit to the latex glove imaging process but contributes to other imaging fields that face the wrinkles issue too.
ISSN:1819-656X
1819-9224