Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network
Cedar and cypress used for wooden construction have high moisture content after harvesting. To be used as building materials, they must undergo high-temperature drying. However, this process causes internal cracks that are invisible on the outer surface. These defects are serious because they reduce...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/3/1280 |
| _version_ | 1797625241874202624 |
|---|---|
| author | Shigeru Kato Naoki Wada Kazuki Shiogai Takashi Tamaki Tomomichi Kagawa Renon Toyosaki Hajime Nobuhara |
| author_facet | Shigeru Kato Naoki Wada Kazuki Shiogai Takashi Tamaki Tomomichi Kagawa Renon Toyosaki Hajime Nobuhara |
| author_sort | Shigeru Kato |
| collection | DOAJ |
| description | Cedar and cypress used for wooden construction have high moisture content after harvesting. To be used as building materials, they must undergo high-temperature drying. However, this process causes internal cracks that are invisible on the outer surface. These defects are serious because they reduce the strength of the timber, i.e., the buckling strength and joint durability. Therefore, the severity of internal cracks should be evaluated. A square timber was cut at an arbitrary position and assessed based on the length, thickness, and shape of the cracks in the cross-section; however, this process is time-consuming and labor-intensive. Therefore, we used a convolutional neural network (CNN) to automatically evaluate the severity of cracks from cross-sectional timber images. Previously, we used silver-painted images of cross-sections so that the cracks are easier to observe; however, this task was burdensome. Hence, in this study, we attempted to classify crack severity using ResNet (Residual Neural Network) from unpainted images. First, ResNet50 was employed and trained with supervised data to classify the crack severity level. The classification accuracy was then evaluated using test images (not used for training) and reached 86.67%. In conclusion, we confirmed that the proposed CNN could evaluate cross-sectional cracks on behalf of humans. |
| first_indexed | 2024-03-11T09:53:47Z |
| format | Article |
| id | doaj.art-29c8280699a049bea86d4a5ec54a176b |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T09:53:47Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-29c8280699a049bea86d4a5ec54a176b2023-11-16T16:03:09ZengMDPI AGApplied Sciences2076-34172023-01-01133128010.3390/app13031280Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural NetworkShigeru Kato0Naoki Wada1Kazuki Shiogai2Takashi Tamaki3Tomomichi Kagawa4Renon Toyosaki5Hajime Nobuhara6National Institute of Technology, Niihama College, 7-1 Yagumo-cho, Niihama 792-8580, JapanNational Institute of Technology, Niihama College, 7-1 Yagumo-cho, Niihama 792-8580, JapanNational Institute of Technology, Niihama College, 7-1 Yagumo-cho, Niihama 792-8580, JapanEhime Prefectural Forestry Research Center, 2-280-38 Sugou, Kumakougen-cho 792-1205, JapanNational Institute of Technology, Niihama College, 7-1 Yagumo-cho, Niihama 792-8580, JapanNational Institute of Technology, Niihama College, 7-1 Yagumo-cho, Niihama 792-8580, JapanDepartment of Intelligent Interaction Technologies, Graduate School of Systems and Information Engineering, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba 305-8573, JapanCedar and cypress used for wooden construction have high moisture content after harvesting. To be used as building materials, they must undergo high-temperature drying. However, this process causes internal cracks that are invisible on the outer surface. These defects are serious because they reduce the strength of the timber, i.e., the buckling strength and joint durability. Therefore, the severity of internal cracks should be evaluated. A square timber was cut at an arbitrary position and assessed based on the length, thickness, and shape of the cracks in the cross-section; however, this process is time-consuming and labor-intensive. Therefore, we used a convolutional neural network (CNN) to automatically evaluate the severity of cracks from cross-sectional timber images. Previously, we used silver-painted images of cross-sections so that the cracks are easier to observe; however, this task was burdensome. Hence, in this study, we attempted to classify crack severity using ResNet (Residual Neural Network) from unpainted images. First, ResNet50 was employed and trained with supervised data to classify the crack severity level. The classification accuracy was then evaluated using test images (not used for training) and reached 86.67%. In conclusion, we confirmed that the proposed CNN could evaluate cross-sectional cracks on behalf of humans.https://www.mdpi.com/2076-3417/13/3/1280timbercrackconvolutional neural networkResNethuman fuzziness |
| spellingShingle | Shigeru Kato Naoki Wada Kazuki Shiogai Takashi Tamaki Tomomichi Kagawa Renon Toyosaki Hajime Nobuhara Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network Applied Sciences timber crack convolutional neural network ResNet human fuzziness |
| title | Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network |
| title_full | Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network |
| title_fullStr | Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network |
| title_full_unstemmed | Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network |
| title_short | Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network |
| title_sort | crack severity classification from timber cross sectional images using convolutional neural network |
| topic | timber crack convolutional neural network ResNet human fuzziness |
| url | https://www.mdpi.com/2076-3417/13/3/1280 |
| work_keys_str_mv | AT shigerukato crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork AT naokiwada crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork AT kazukishiogai crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork AT takashitamaki crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork AT tomomichikagawa crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork AT renontoyosaki crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork AT hajimenobuhara crackseverityclassificationfromtimbercrosssectionalimagesusingconvolutionalneuralnetwork |