Pipeline Landmark Classification of Miniature Pipeline Robot π-II Based on Residual Network ResNet18
Abstract
:1. Introduction
2. Experimental Platform and Its Detection System of Miniature Pipeline Robot π-II
3. Construction of Typical Pipeline Landmark Dataset
3.1. Shadow Analysis of Typical Pipeline Landmarks
3.1.1. Shadow Characteristics of Straight Pipes
3.1.2. Shadow Characteristics of Curved Pipes
3.1.3. Shadow Characteristics of T-Shaped Pipe Entering from the Side End
3.1.4. Shadow Characteristics of T-Shaped Pipe Entering from the Lower End
3.2. Creation of the Pipeline Landmark Dataset
self.pip_trainsform=torchvision.transforms.Compose([ torchvision.transforms.ToTensor(), torchvision.transforms.Resize((height, width)), torchvision.transforms.Normalize(mean=[B_mean, G_mean, R_mean], std=[B_std, G_std, R_std]), ]) |
train_images,test_images,train_labels,test_labels=sklearn.model_selecti on.train_test_split(images,labels,test_size=0.2,random_state=42) |
4. Pipeline Landmark Classification Based on the Deep Learning Mode ResNet18
4.1. Choice and Modification of the Classification Model
class PIPEResNet18(torch.nn.Module): def __init__(self): super(PIPEResNet18,self).__init__() self.cnn_layers=torchvision.models.resnet18(pretrained=True) num_ftrs=self.cnn_layers.fc.in_features self.cnn_layers.fc=torch.nn.Linear(num_ftrs,4) def forward(self,x): out=self.cnn_layers(x) return out |
4.2. Selection of the Loss Function and Optimization Method during Training
criterion=torch.nn.CrossEntropyLoss(reduction=‘mean’) |
optimizer=torch.optim.Adam(model.parameters(),lr=0.001) |
4.3. Selection of Performance Metrics during Testing
micro_accracy=precision_score(test_labels,test_outputs,average=‘micro’) |
4.4. Training Process and Testing Results
train_dataloader=DataLoader(train_ds,batch_size=64,shuffle=True,drop_la st=True) |
5. Influence of HyperParameter “batch_size”
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Jun, C.; Deng, Z.Q.; Jiang, S.Y. Study of locomotion control characteristics for six wheels driven in-pipe robot. In Proceedings of the 2004 IEEE International Conference on Robotics and Biomimetics, Shenyang, China, 22–26 August 2004; pp. 119–124. [Google Scholar] [CrossRef]
- Kim, H.M.; Choi, Y.S.; Lee, Y.G.; Choi, H.R. Novel mechanism for in-pipe robot based on a multiaxial differential gear mechanism. IEEE/ASME Trans. Mechatron. 2017, 22, 227–235. [Google Scholar] [CrossRef]
- Kwon, Y.S.; Lee, B.; Whang, I.C.; Kim, W.K.; Yi, B.J. A flat pipeline inspection robot with two wheel chains. In Proceedings of the 2011 IEEE International Conference on Robotics and Automation Shanghai International Conference Center, Shanghai, China, 9–13 May 2011; pp. 5141–5146. [Google Scholar] [CrossRef]
- Kwon, Y.S.; Lee, B.; Whang, I.C.; Yi, B.J. A pipeline inspection robot with a linkage type mechanical clutch. In Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, Taiwan, 18–22 October 2010; pp. 2850–2855. [Google Scholar] [CrossRef]
- Wang, J.; Wu, H.; Wang, J.Z. Design of a small-type wheeled pipeline robot driven by monocular vision. In Proceedings of the 2023 IEEE 7th Information Technology and Mechatronics Engineering Conference, Chongqing, China, 15–17 September 2023; pp. 1133–1137. [Google Scholar] [CrossRef]
- Chen, S.; Hu, H.; Guan, W.; Huang, C.; Gong, W.; Xu, X. Research on vision control system of pipeline robot. In Proceedings of the 2021 China Automation Congress, Beijing, China, 22–24 October 2021; pp. 5379–5384. [Google Scholar] [CrossRef]
- Zhang, X.; Chen, H. Visual servo technique of autonomous mobile robot in bended pipe. In Proceedings of the 2002 IEEE International Conference on Industrial Technology, Bangkok, Thailand, 11–14 December 2002; pp. 588–593. [Google Scholar] [CrossRef]
- Ahrary, A.; Tian, L.; Kamata, S.; Ishikawa, M. Anautonomous sewer robots navigation based on stereo camera information. In Proceedings of the 17th IEEE International Conference on Tools with Artificial Intelligence, HongKong, China, 14–16 November 2005. [Google Scholar] [CrossRef]
- Lee, J.S.; Roh, S.G.; Kim, D.W.; Moon, H.; Choi, H.R. In-pipe robot navigation based on the landmark recognition system using shadow images. In Proceedings of the 2009 IEEE International Conference on Robotics and Automation, Kobe, Japan, 12–17 May 2009; pp. 1857–1862. [Google Scholar] [CrossRef]
- Thielemann, J.T.; Breivik, G.M.; Berge, A. Pipeline landmark detection for autonomous robot navigation using time-of-flight imagery. In Proceedings of the 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, Anchorage, AK, USA, 23–28 June 2008; pp. 1–7. [Google Scholar] [CrossRef]
- Chand, A.N.; Zuhdi, N.; Mansor, A.; Iqbal, A.; Rustam, F.; Baur, W. An industrial robot for fire water piping inspection and mapping. In Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems, Prague, Czech Republic, 27 September 2021; pp. 2337–2344. [Google Scholar] [CrossRef]
- Choi, Y.S.; Kim, H.M.; Suh, J.S.; Mun, H.M.; Yang, S.U.; Park, C.M.; Choi, H.R. Recognition of inside pipeline geometry by using PSD sensors for autonomous navigation. In Proceedings of the 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, Chicago, IL, USA, 14–18 September 2014; pp. 5024–5029. [Google Scholar] [CrossRef]
- Lee, D.H.; Moon, H.; Choi, H.R. Landmark detection of in-pipe working robot usingline-laser beam projection. In Proceedings of the ICCAS 2010, Gyeonggi-do, Republic of Korea, 27–30 October 2010; pp. 611–615. [Google Scholar] [CrossRef]
- Lee, D.H.; Moon, H.; Koo, J.C.; Choi, H.R. Map building method for urban gas pipelines based on landmark detection. Int. J. Control Autom. Syst. 2013, 11, 127–135. [Google Scholar] [CrossRef]
- Lee, D.H.; Moon, H.; Choi, H.R. Autonomous navigation of in-pipe working robot in unknown pipeline environment. In Proceedings of the 2011 IEEE International Conference on Robotics and Automation, Shanghai, China, 9–13 May 2011; pp. 1559–1564. [Google Scholar] [CrossRef]
- Choi, Y.S.; Kim, H.M.; Mun, H.M.; Lee, Y.G.; Choi, H.R. Recognition of pipeline geometry by using monocular camera and PSD sensors. Intel. Serv. Robot. 2017, 10, 213–227. [Google Scholar] [CrossRef]
- Haurum, J.B.; Bahnsen, C.H.; Pedersen, M.; Moeslund, T.B. Water level estimation in sewer pipes using deep convolutional neuralnetworks. Water 2020, 12, 3412. [Google Scholar] [CrossRef]
- Patil, R.R.; Mustafa, M.Y.; Calay, R.K.; Ansari, S.M. S-BIRD: A novel critical multi-class imagery dataset for sewer monitoring and maintenance systems. Sensors 2023, 23, 2966. [Google Scholar] [CrossRef] [PubMed]
- Zhao, Y.; Su, Z.; Zhao, H. Micro-leakage image recognition method for internal detection in small, buried gas pipelines. Sensors 2023, 23, 3956. [Google Scholar] [CrossRef] [PubMed]
- Wang, J.; Mo, Z.; Cai, Y.; Wang, S. Kinematic analysis of a wheeled-leg small pipeline robot turning in curved pipes. Electronics 2024, 13, 2170. [Google Scholar] [CrossRef]
Parameter | Value |
---|---|
Resolution | 640 × 480 |
Frame Rate | 30 fps |
View Angle | 70° |
Waterproof Level | IP67 |
Photo Format | JPEG |
Device | Specification |
---|---|
CPU | Intel i5-7300HQ 2.50GHz |
Memory | 20 GB |
GPU | NVIDIA GeForce GTX 1050 |
GPU Memory | 4 GB |
Parameter | Value | |
---|---|---|
Training | Test | |
Datasets | 704 | 182 |
Resolution | 200 × 200 | 200 × 200 |
Epochs | 100 | 100 |
Batch size | 64 | 64 |
Steps per epoch | 11 | 3 |
Learning rate | 0.001 | --- |
Optimizer | Adam | --- |
Fps | 142 | 100 |
Straight Pipe | Curved Pipe | T-Shaped Pipe I | T-Shaped Pipe II | |
---|---|---|---|---|
Straight pipe | 43 | 0 | 0 | 0 |
Curved pipe | 0 | 53 | 0 | 0 |
T-shaped pipe I | 0 | 0 | 42 | 0 |
T-shaped pipe II | 0 | 0 | 0 | 44 |
Straight Pipe | Curved Pipe | T-Shaped Pipe I | T-Shaped Pipe II | |
---|---|---|---|---|
TP | 43 | 53 | 42 | 44 |
FN | 0 | 0 | 0 | 0 |
FP | 0 | 0 | 0 | 0 |
TN | 139 | 129 | 140 | 138 |
Presicion | 1 | 1 | 1 | 1 |
Recall | 1 | 1 | 1 | 1 |
F1 score | 1 | 1 | 1 | 1 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Wang, J.; Chen, C.; Liu, B.; Wang, J.; Wang, S. Pipeline Landmark Classification of Miniature Pipeline Robot π-II Based on Residual Network ResNet18. Machines 2024, 12, 563. https://doi.org/10.3390/machines12080563
Wang J, Chen C, Liu B, Wang J, Wang S. Pipeline Landmark Classification of Miniature Pipeline Robot π-II Based on Residual Network ResNet18. Machines. 2024; 12(8):563. https://doi.org/10.3390/machines12080563
Chicago/Turabian StyleWang, Jian, Chuangeng Chen, Bingsheng Liu, Juezhe Wang, and Songtao Wang. 2024. "Pipeline Landmark Classification of Miniature Pipeline Robot π-II Based on Residual Network ResNet18" Machines 12, no. 8: 563. https://doi.org/10.3390/machines12080563