Transfer learning and fine‐tuned transfer learning methods' effectiveness analyse in the CNN‐based deep learning models

Celal Ozturk; Murat Tasyurek; Mehmet Uğur Türkdamar

doi:10.1002/cpe.7542

journal article Nov 29, 2022

Transfer learning and fine‐tuned transfer learning methods' effectiveness analyse in the CNN‐based deep learning models

Celal Ozturk

Murat Tasyurek

Mehmet Uğur Türkdamar

Concurrency and Computation: Practice and Experience Vol. 35 No. 4 · Wiley

View at Publisher Save 10.1002/cpe.7542

Abstract

SummaryObject detection is a type of application that includes computer vision and image processing technologies, which deal with detecting, tracking, and classifying desired objects in images. Computer vision is a field of artificial intelligence that enables computers and systems to derive information from digital images and take action or suggestions based on that information. CNN is one of the current methods of object detection due to its ease of use and GPU‐supported parallel working features. Due to the aim of completing deep learning model training quickly or due to insufficient dataset, many studies using the transfer learning method are carried out in fields such as medicine, agriculture, and weapons. However, there are very few studies that use the fine‐tuning method and compare transfer learning in terms of effectiveness. By paying attention to the balanced distribution of the data, approximately 100 images of each chess piece type were included in the analysis and a dataset of at least 1000 images was created. The without transfer learning fine‐tune, fine‐tuned transfer learning, transfer learning, fully supervised learning (FSL) and weakly supervised learning (WSL) applied models performances compared. Experimental results show that the fine‐tuned transfer learning applied YOLO V4 model produces more accurate results than the other models in FSL and the transfer learning applied Faster R‐CNN model produces more accurate results than the other models in WSL.

Topics

No keywords indexed for this article. Browse by subject →

References

60

[1]

BenjdiraB KhursheedT KoubaaA AmmarA OuniK.Car detection using unmanned aerial vehicles: comparison between faster R‐CNN and yolov3. arXiv:1812:10968v1 2019. 10.1109/uvs.2019.8658300

[2]

Deep learning

Yann LeCun, Yoshua Bengio, Geoffrey Hinton

Nature 10.1038/nature14539

[3]

10.48550/arxiv.1512.03385

[4]

LoshchilovI HutterF.SGDR: stochastic gradient descent with warm restarts. arXiv:1608.03983 2016.

[5]

Fırıldak K "Evrişimsel sinir ağlarında kullanılan transfer öğrenme yaklaşımlarının incelenmesi" Anatol J Comput Sci (2019)

[6]

TanC SunF KongT ZhangW YangC LiuC.A survey on deep transfer learning. arXiv:1808.01974v1 2018.

[7]

Koçer B (2012)

[8]

Fchollet.Transfer learning & fine‐tuning;2020.https://keras.io/guides/transfer‐learning/

[9]

Eyiokur FÍ (2018)

[10]

Gürkaynak CDB (2018)

[11]

RuiC YouweiG HuafeiZ HongyuJ.A comprehensive approach for uav small object detection with simulation‐based transfer learning and adaptive fusion. arXiv:01800v1 2021.

[12]

Chen S. Ma K. ZhengY.Med3d: transfer learning for 3d medical image analysis. arXiv:1904.00625v4 2019.

[13]

10.1016/j.jocs.2018.11.008

[14]

10.1016/j.compag.2020.105393

[15]

10.1093/mnras/sty1398

[16]

10.1186/s13635-019-0084-4

[17]

10.3390/rs13010023

[18]

AryaD MaedaH GhoshSK ToshniwalD MrazA KashiyamaT SekimotoY.Transfer learning‐based road damage detection for multiple countries. arXiv preprint arXiv:2008.13101 2020.

[19]

Harsono IW "Lung nodule detection and classification from thorax CT‐scan usingretinanet with transfer learning" J King Saud Univ Comput Inf Sci (2022)

[20]

10.3390/s19163523

[21]

Chaoxia C "Information‐guided flame detection based on faster r‐cnn" IEEE Access (2020)

[22]

Sethy PK "Rice false smut detection based on faster r‐cnn" Ind J Electr Eng Comput Sci (2020)

[23]

A novel deep learning based framework for the detection and classification of breast cancer using transfer learning

SanaUllah Khan, Naveed Islam, Zahoor Jan et al.

Pattern Recognition Letters 10.1016/j.patrec.2019.03.022

[24]

Brain tumor classification using deep CNN features via transfer learning

S. Deepak, P.M. Ameer

Computers in Biology and Medicine 10.1016/j.compbiomed.2019.103345

[25]

10.1016/j.irbm.2020.05.003

[26]

10.1016/j.compag.2019.01.041

[27]

10.1002/jsfa.9689

[28]

10.1016/j.measurement.2020.108288

[29]

Vo AH "A novel framework for trash classification using deep transfer learning" IEEE Access (2018)

[30]

10.1016/j.infrared.2019.103048

[31]

10.1016/j.compag.2020.105384

[32]

Geng M. Wang Y. Xiang T. Tian Y.Deep transfer learning for person re‐identification. arXiv preprint arXiv:1611.05244 2016. 10.1109/cvpr.2016.146

[33]

10.1016/j.ecoinf.2019.02.007

[34]

Hosny KM "Skin cancer classification using deep learning and transfer learning" IEEE Access (2018)

[35]

10.1007/s00034-019-01246-3

[36]

Gao Y "Deep transfer learning for image‐based structural damage recognition" Health Monitor Struct (2018)

[37]

10.1016/j.compag.2020.105306

[38]

10.3390/rs10010139

[39]

Zhang D "Weakly supervised object localization and detection: a survey" IEEE Trans Pattern Analysis and Machine Intelligence (2021)

[40]

10.1109/tpami.2020.3046647

[41]

10.1109/tnnls.2020.2969483

[42]

Tan Z (2019)

[43]

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

Ross Girshick, Jeff Donahue, Trevor Darrell et al.

2014 IEEE Conference on Computer Vision and Patter... 10.1109/cvpr.2014.81

[44]

Fast R-CNN

Ross Girshick

2015 IEEE International Conference on Computer Vis... 10.1109/iccv.2015.169

[45]

Ren S. He K. Girshick R.&Sun J.Faster r‐cnn: towards real‐time object detection with region proposal networks. arXiv:1506 01497v3 2016.

[46]

You Only Look Once: Unified, Real-Time Object Detection

Joseph Redmon, Santosh Divvala, Ross Girshick et al.

2016 IEEE Conference on Computer Vision and Patter... 10.1109/cvpr.2016.91

[47]

BochkovskiyA WangCY LiaoHYM.YOLOv4: optimal speed and accuracy of object detection. arXiv 2020;2004.109.

[48]

Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning

Siyu Shao, Stephen McAleer, Ruqiang Yan et al.

IEEE Transactions on Industrial Informatics 10.1109/tii.2018.2864759

[49]

YOLO9000: Better, Faster, Stronger

Joseph Redmon, Ali Farhadi

2017 IEEE Conference on Computer Vision and Patter... 10.1109/cvpr.2017.690

[50]

SolawetzJ.YOLOv4 ‐ ten tactics to build a better model;2020.https://blog.roboflow.com/yolov4‐tactics/

Showing 50 of 60 references

Metrics

56

Citations

60

References

Details

Published: Nov 29, 2022
Vol/Issue: 35(4)
License: View

Authors

C

Celal Ozturk

Department of Computer Engineering Erciyes University Kayseri Turkey

M

Murat Tasyurek

Department of Computer Engineering Kayseri University Kayseri Turkey

M

Mehmet Uğur Türkdamar

Department of Computer Engineering Niğde Ömer Halisdemir University Niğde Turkey

Cite This Article

Celal Ozturk, Murat Tasyurek, Mehmet Uğur Türkdamar (2022). Transfer learning and fine‐tuned transfer learning methods' effectiveness analyse in the CNN‐based deep learning models. Concurrency and Computation: Practice and Experience, 35(4). https://doi.org/10.1002/cpe.7542

Transfer learning and fine‐tuned transfer learning methods' effectiveness analyse in the CNN‐based deep learning models

You May Also Like