Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method
We proposed an automatic detection method of slope failure regions using a semantic segmentation method called Mask R-CNN based on a deep learning algorithm to improve the efficiency of damage assessment in the event of slope failure disaster. There is limited research on detecting landslides by dee...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/17/6412 |
| _version_ | 1797493197665992704 |
|---|---|
| author | Shiori Kubo Tatsuro Yamane Pang-jo Chun |
| author_facet | Shiori Kubo Tatsuro Yamane Pang-jo Chun |
| author_sort | Shiori Kubo |
| collection | DOAJ |
| description | We proposed an automatic detection method of slope failure regions using a semantic segmentation method called Mask R-CNN based on a deep learning algorithm to improve the efficiency of damage assessment in the event of slope failure disaster. There is limited research on detecting landslides by deep learning, and the lack of training data is an important issue to be resolved, as aerial photographs are not taken with sufficient frequency during a disaster. This study attempts to use CutMix-based augmentation to improve detection accuracy. We also compare the detection results obtained by augmentation of multiple patterns. In the comparison of the not augmented data case, the recall increased by 0.186 in the case using the augmented data with the shape of the slope failure region maintained. When the image data was augmented while maintaining the shape of the slope failure region, the recall score indicated the low oversights in the prediction result is 0.701. This is an increase of 0.186 compared to the case where no augmentation was performed. In addition, the F1 score was 0.740, this also increased by 0.139, and high values were obtained for other indicators. Therefore, the method proposed in this study is greatly useful for grasping slope failure regions because of the detection with high accuracy, as described above. |
| first_indexed | 2024-03-10T01:16:34Z |
| format | Article |
| id | doaj.art-debdc45d716442f99dd2dd29b2d5d153 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T01:16:34Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-debdc45d716442f99dd2dd29b2d5d1532023-11-23T14:08:02ZengMDPI AGSensors1424-82202022-08-012217641210.3390/s22176412Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation MethodShiori Kubo0Tatsuro Yamane1Pang-jo Chun2Institute of Industrial Science, The University of Tokyo, Chiba 277-8574, JapanDepartment of International Studies, The University of Tokyo, Chiba 277-8561, JapanDepartment of Civil Engineering, The University of Tokyo, Tokyo 113-8656, JapanWe proposed an automatic detection method of slope failure regions using a semantic segmentation method called Mask R-CNN based on a deep learning algorithm to improve the efficiency of damage assessment in the event of slope failure disaster. There is limited research on detecting landslides by deep learning, and the lack of training data is an important issue to be resolved, as aerial photographs are not taken with sufficient frequency during a disaster. This study attempts to use CutMix-based augmentation to improve detection accuracy. We also compare the detection results obtained by augmentation of multiple patterns. In the comparison of the not augmented data case, the recall increased by 0.186 in the case using the augmented data with the shape of the slope failure region maintained. When the image data was augmented while maintaining the shape of the slope failure region, the recall score indicated the low oversights in the prediction result is 0.701. This is an increase of 0.186 compared to the case where no augmentation was performed. In addition, the F1 score was 0.740, this also increased by 0.139, and high values were obtained for other indicators. Therefore, the method proposed in this study is greatly useful for grasping slope failure regions because of the detection with high accuracy, as described above.https://www.mdpi.com/1424-8220/22/17/6412image augmentationMask R-CNNslope failureimage segmentationdeep learning |
| spellingShingle | Shiori Kubo Tatsuro Yamane Pang-jo Chun Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method Sensors image augmentation Mask R-CNN slope failure image segmentation deep learning |
| title | Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method |
| title_full | Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method |
| title_fullStr | Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method |
| title_full_unstemmed | Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method |
| title_short | Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method |
| title_sort | study on accuracy improvement of slope failure region detection using mask r cnn with augmentation method |
| topic | image augmentation Mask R-CNN slope failure image segmentation deep learning |
| url | https://www.mdpi.com/1424-8220/22/17/6412 |
| work_keys_str_mv | AT shiorikubo studyonaccuracyimprovementofslopefailureregiondetectionusingmaskrcnnwithaugmentationmethod AT tatsuroyamane studyonaccuracyimprovementofslopefailureregiondetectionusingmaskrcnnwithaugmentationmethod AT pangjochun studyonaccuracyimprovementofslopefailureregiondetectionusingmaskrcnnwithaugmentationmethod |