A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects
In recent years, due to the shortage of timber resources, it has become necessary to reduce the excessive consumption of forest resources. Non-destructive testing technology can quickly find wood defects and effectively improve wood utilization. Deep learning has achieved significant results as one...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-02-01
|
Series: | Forests |
Subjects: | |
Online Access: | https://www.mdpi.com/1999-4907/12/2/212 |
id |
doaj-d4ee97ec6dcb414eb692974241801764 |
---|---|
record_format |
Article |
spelling |
doaj-d4ee97ec6dcb414eb6929742418017642021-02-12T00:05:26ZengMDPI AGForests1999-49072021-02-011221221210.3390/f12020212A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot DefectsMingyu Gao0Jianfeng Chen1Hongbo Mu2Dawei Qi3College of Science, Northeast Forestry University, Harbin 150040, ChinaCollege of Science, Northeast Forestry University, Harbin 150040, ChinaCollege of Science, Northeast Forestry University, Harbin 150040, ChinaCollege of Science, Northeast Forestry University, Harbin 150040, ChinaIn recent years, due to the shortage of timber resources, it has become necessary to reduce the excessive consumption of forest resources. Non-destructive testing technology can quickly find wood defects and effectively improve wood utilization. Deep learning has achieved significant results as one of the most commonly used methods in the detection of wood knots. However, compared with convolutional neural networks in other fields, the depth of deep learning models for the detection of wood knots is still very shallow. This is because the number of samples marked in the wood detection is too small, which limits the accuracy of the final prediction of the results. In this paper, ResNet-34 is combined with transfer learning, and a new TL-ResNet34 deep learning model with 35 convolution depths is proposed to detect wood knot defects. Among them, ResNet-34 is used as a feature extractor for wood knot defects. At the same time, a new method TL-ResNet34 is proposed, which combines ResNet-34 with transfer learning. After that, the wood knot defect dataset was applied to TL-ResNet34 for testing. The results show that the detection accuracy of the dataset trained by TL-ResNet34 is significantly higher than that of other methods. This shows that the final prediction accuracy of the detection of wood knot defects can be improved by TL-ResNet34.https://www.mdpi.com/1999-4907/12/2/212wood knot defects detectiondeep learningtransfer learningresidual neural networks |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Mingyu Gao Jianfeng Chen Hongbo Mu Dawei Qi |
spellingShingle |
Mingyu Gao Jianfeng Chen Hongbo Mu Dawei Qi A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects Forests wood knot defects detection deep learning transfer learning residual neural networks |
author_facet |
Mingyu Gao Jianfeng Chen Hongbo Mu Dawei Qi |
author_sort |
Mingyu Gao |
title |
A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects |
title_short |
A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects |
title_full |
A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects |
title_fullStr |
A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects |
title_full_unstemmed |
A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects |
title_sort |
transfer residual neural network based on resnet-34 for detection of wood knot defects |
publisher |
MDPI AG |
series |
Forests |
issn |
1999-4907 |
publishDate |
2021-02-01 |
description |
In recent years, due to the shortage of timber resources, it has become necessary to reduce the excessive consumption of forest resources. Non-destructive testing technology can quickly find wood defects and effectively improve wood utilization. Deep learning has achieved significant results as one of the most commonly used methods in the detection of wood knots. However, compared with convolutional neural networks in other fields, the depth of deep learning models for the detection of wood knots is still very shallow. This is because the number of samples marked in the wood detection is too small, which limits the accuracy of the final prediction of the results. In this paper, ResNet-34 is combined with transfer learning, and a new TL-ResNet34 deep learning model with 35 convolution depths is proposed to detect wood knot defects. Among them, ResNet-34 is used as a feature extractor for wood knot defects. At the same time, a new method TL-ResNet34 is proposed, which combines ResNet-34 with transfer learning. After that, the wood knot defect dataset was applied to TL-ResNet34 for testing. The results show that the detection accuracy of the dataset trained by TL-ResNet34 is significantly higher than that of other methods. This shows that the final prediction accuracy of the detection of wood knot defects can be improved by TL-ResNet34. |
topic |
wood knot defects detection deep learning transfer learning residual neural networks |
url |
https://www.mdpi.com/1999-4907/12/2/212 |
work_keys_str_mv |
AT mingyugao atransferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT jianfengchen atransferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT hongbomu atransferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT daweiqi atransferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT mingyugao transferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT jianfengchen transferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT hongbomu transferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects AT daweiqi transferresidualneuralnetworkbasedonresnet34fordetectionofwoodknotdefects |
_version_ |
1724273947271233536 |