Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images
Quick response codes (QRCs) are found on many consumer products and often encode security information. However, information retrieval at receiving end may become challenging due to the degraded clarity of QRC images. This degradation may occur because of the transmission of digital images over noise...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-04-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844023023150 |
| _version_ | 1797836931135963136 |
|---|---|
| author | Sadaf Waziry Ahmad Bilal Wardak Jawad Rasheed Raed M. Shubair Khairan Rajab Asadullah Shaikh |
| author_facet | Sadaf Waziry Ahmad Bilal Wardak Jawad Rasheed Raed M. Shubair Khairan Rajab Asadullah Shaikh |
| author_sort | Sadaf Waziry |
| collection | DOAJ |
| description | Quick response codes (QRCs) are found on many consumer products and often encode security information. However, information retrieval at receiving end may become challenging due to the degraded clarity of QRC images. This degradation may occur because of the transmission of digital images over noise channels or limited printing technology. Although the ability to reduce noises is critical, it is just as important to define the type and quantity of noises present in QRC images. Therefore, this study proposed a simple deep learning-based architecture to segregate the image as either an original (normal) QRC or a noisy QRC and identifies the noise type present in the image. For this, the study is divided into two stages. Firstly, it generated a QRC image dataset of 80,000 images by introducing seven different noises (speckle, salt & pepper, Poisson, pepper, localvar, salt, and Gaussian) to the original QRC images. Secondly, the generated dataset is fed to train the proposed convolutional neural network (CNN)-based model, seventeen pre-trained deep learning models, and two classical machine learning algorithms (Naïve Bayes (NB) and Decision Tree (DT)). XceptionNet attained the highest accuracy (87.48%) and kappa (85.7%). However, it is worth noting that the proposed CNN network with few layers competes with the state-of-the-art models and attained near to best accuracy (86.75%). Furthermore, detailed analysis shows that all models failed to classify images having Gaussian and Localvar noises correctly. |
| first_indexed | 2024-04-09T15:17:54Z |
| format | Article |
| id | doaj.art-1a32648dab49494295ca7ee36825da58 |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-04-09T15:17:54Z |
| publishDate | 2023-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-1a32648dab49494295ca7ee36825da582023-04-29T14:55:20ZengElsevierHeliyon2405-84402023-04-0194e15108Performance comparison of machine learning driven approaches for classification of complex noises in quick response code imagesSadaf Waziry0Ahmad Bilal Wardak1Jawad Rasheed2Raed M. Shubair3Khairan Rajab4Asadullah Shaikh5Department of Software Engineering, Istanbul Aydin University, Istanbul 34295, TurkeyDepartment of Software Engineering, Istanbul Aydin University, Istanbul 34295, TurkeyDepartment of Software Engineering, Istanbul Nisantasi University, Istanbul 34398, Turkey; Corresponding author.Department of Electrical and Computer Engineering, New York University (NYU), Abu Dhabi 129188, United Arab EmiratesDepartment of Computer Science, College of Computer Science and Information Systems, Najran University, Najran 61441, Saudi ArabiaDepartment of Information Systems, College of Computer Science and Information Systems, Najran University, Najran 61441, Saudi ArabiaQuick response codes (QRCs) are found on many consumer products and often encode security information. However, information retrieval at receiving end may become challenging due to the degraded clarity of QRC images. This degradation may occur because of the transmission of digital images over noise channels or limited printing technology. Although the ability to reduce noises is critical, it is just as important to define the type and quantity of noises present in QRC images. Therefore, this study proposed a simple deep learning-based architecture to segregate the image as either an original (normal) QRC or a noisy QRC and identifies the noise type present in the image. For this, the study is divided into two stages. Firstly, it generated a QRC image dataset of 80,000 images by introducing seven different noises (speckle, salt & pepper, Poisson, pepper, localvar, salt, and Gaussian) to the original QRC images. Secondly, the generated dataset is fed to train the proposed convolutional neural network (CNN)-based model, seventeen pre-trained deep learning models, and two classical machine learning algorithms (Naïve Bayes (NB) and Decision Tree (DT)). XceptionNet attained the highest accuracy (87.48%) and kappa (85.7%). However, it is worth noting that the proposed CNN network with few layers competes with the state-of-the-art models and attained near to best accuracy (86.75%). Furthermore, detailed analysis shows that all models failed to classify images having Gaussian and Localvar noises correctly.http://www.sciencedirect.com/science/article/pii/S2405844023023150Noisy imagesQuick-response codeNoise identificationNoise classificationMachine learningDeep learning |
| spellingShingle | Sadaf Waziry Ahmad Bilal Wardak Jawad Rasheed Raed M. Shubair Khairan Rajab Asadullah Shaikh Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images Heliyon Noisy images Quick-response code Noise identification Noise classification Machine learning Deep learning |
| title | Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images |
| title_full | Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images |
| title_fullStr | Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images |
| title_full_unstemmed | Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images |
| title_short | Performance comparison of machine learning driven approaches for classification of complex noises in quick response code images |
| title_sort | performance comparison of machine learning driven approaches for classification of complex noises in quick response code images |
| topic | Noisy images Quick-response code Noise identification Noise classification Machine learning Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2405844023023150 |
| work_keys_str_mv | AT sadafwaziry performancecomparisonofmachinelearningdrivenapproachesforclassificationofcomplexnoisesinquickresponsecodeimages AT ahmadbilalwardak performancecomparisonofmachinelearningdrivenapproachesforclassificationofcomplexnoisesinquickresponsecodeimages AT jawadrasheed performancecomparisonofmachinelearningdrivenapproachesforclassificationofcomplexnoisesinquickresponsecodeimages AT raedmshubair performancecomparisonofmachinelearningdrivenapproachesforclassificationofcomplexnoisesinquickresponsecodeimages AT khairanrajab performancecomparisonofmachinelearningdrivenapproachesforclassificationofcomplexnoisesinquickresponsecodeimages AT asadullahshaikh performancecomparisonofmachinelearningdrivenapproachesforclassificationofcomplexnoisesinquickresponsecodeimages |