Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation
Plastic bottle recycling has a crucial role in environmental degradation and protection. Position and background should be the same to classify plastic bottles on a conveyor belt. The manual detection of plastic bottles is time consuming and leads to human error. Hence, the automatic classification...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/9/1541 |
| _version_ | 1797503799920689152 |
|---|---|
| author | Subhajit Chatterjee Debapriya Hazra Yung-Cheol Byun Yong-Woon Kim |
| author_facet | Subhajit Chatterjee Debapriya Hazra Yung-Cheol Byun Yong-Woon Kim |
| author_sort | Subhajit Chatterjee |
| collection | DOAJ |
| description | Plastic bottle recycling has a crucial role in environmental degradation and protection. Position and background should be the same to classify plastic bottles on a conveyor belt. The manual detection of plastic bottles is time consuming and leads to human error. Hence, the automatic classification of plastic bottles using deep learning techniques can assist with the more accurate results and reduce cost. To achieve a considerably good result using the DL model, we need a large volume of data to train. We propose a GAN-based model to generate synthetic images similar to the original. To improve the image synthesis quality with less training time and decrease the chances of mode collapse, we propose a modified lightweight-GAN model, which consists of a generator and a discriminator with an auto-encoding feature to capture essential parts of the input image and to encourage the generator to produce a wide range of real data. Then a newly designed weighted average ensemble model based on two pre-trained models, inceptionV3 and xception, to classify transparent plastic bottles obtains an improved classification accuracy of 99.06%. |
| first_indexed | 2024-03-10T03:55:30Z |
| format | Article |
| id | doaj.art-4da181058efc498b943509a1b590e378 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T03:55:30Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-4da181058efc498b943509a1b590e3782023-11-23T08:45:53ZengMDPI AGMathematics2227-73902022-05-01109154110.3390/math10091541Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data AugmentationSubhajit Chatterjee0Debapriya Hazra1Yung-Cheol Byun2Yong-Woon Kim3Department of Computer Engineering, Jeju National University, Jeju 63243, KoreaDepartment of Computer Engineering, Jeju National University, Jeju 63243, KoreaDepartment of Computer Engineering, Jeju National University, Jeju 63243, KoreaCentre for Digital Innovation, CHRIST University (Deemed to be University), Bengaluru 560029, Karnataka, IndiaPlastic bottle recycling has a crucial role in environmental degradation and protection. Position and background should be the same to classify plastic bottles on a conveyor belt. The manual detection of plastic bottles is time consuming and leads to human error. Hence, the automatic classification of plastic bottles using deep learning techniques can assist with the more accurate results and reduce cost. To achieve a considerably good result using the DL model, we need a large volume of data to train. We propose a GAN-based model to generate synthetic images similar to the original. To improve the image synthesis quality with less training time and decrease the chances of mode collapse, we propose a modified lightweight-GAN model, which consists of a generator and a discriminator with an auto-encoding feature to capture essential parts of the input image and to encourage the generator to produce a wide range of real data. Then a newly designed weighted average ensemble model based on two pre-trained models, inceptionV3 and xception, to classify transparent plastic bottles obtains an improved classification accuracy of 99.06%.https://www.mdpi.com/2227-7390/10/9/1541deep learninggenerative adversarial networksimage classificationtransfer learningplastic bottle |
| spellingShingle | Subhajit Chatterjee Debapriya Hazra Yung-Cheol Byun Yong-Woon Kim Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation Mathematics deep learning generative adversarial networks image classification transfer learning plastic bottle |
| title | Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation |
| title_full | Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation |
| title_fullStr | Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation |
| title_full_unstemmed | Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation |
| title_short | Enhancement of Image Classification Using Transfer Learning and GAN-Based Synthetic Data Augmentation |
| title_sort | enhancement of image classification using transfer learning and gan based synthetic data augmentation |
| topic | deep learning generative adversarial networks image classification transfer learning plastic bottle |
| url | https://www.mdpi.com/2227-7390/10/9/1541 |
| work_keys_str_mv | AT subhajitchatterjee enhancementofimageclassificationusingtransferlearningandganbasedsyntheticdataaugmentation AT debapriyahazra enhancementofimageclassificationusingtransferlearningandganbasedsyntheticdataaugmentation AT yungcheolbyun enhancementofimageclassificationusingtransferlearningandganbasedsyntheticdataaugmentation AT yongwoonkim enhancementofimageclassificationusingtransferlearningandganbasedsyntheticdataaugmentation |