Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique
Natural water content affects many geotechnical parameters and geological properties of soils, which can reduce cohesion and friction, leading to potential failures in structures such as foundations, retaining walls, and slopes. Identification of the water content helps in designing effective draina...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Infrastructures |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2412-3811/10/6/138 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850168192727515136 |
|---|---|
| author | Mark Miller Yong Fang Yubo Wang Sergey Kharitonov Vladimir Akulich |
| author_facet | Mark Miller Yong Fang Yubo Wang Sergey Kharitonov Vladimir Akulich |
| author_sort | Mark Miller |
| collection | DOAJ |
| description | Natural water content affects many geotechnical parameters and geological properties of soils, which can reduce cohesion and friction, leading to potential failures in structures such as foundations, retaining walls, and slopes. Identification of the water content helps in designing effective drainage and water management systems to prevent flooding and erosion. In tunnel engineering, soil water content plays an important role as the stability of the tunnel face depends on it. This research solves the problem of classifying soil images depending on the natural water content by computer vision technology. First, laboratory soil tests were carried out, and the relationship between the amount of torque on the screw conveyor and the moisture content of the soil was established; photographs of the soil at different conditions were taken at each step of the experiment. Second, the resulting dataset after preprocessing was processed by convolutional neural network algorithms during deep learning; the transfer learning technique was used to obtain better results. As a result, seven algorithms were obtained that allow classifying the soil images, which can later be used to optimize the tunnel construction process. The best classification ability is demonstrated by the algorithm based on the DenseNet architecture (accuracy 0.9302 and loss 0.1980). The proposed model surpasses traditional approaches due to its increased automation and processing speed. Laboratory tests can be carried out only once for one type of soil in order to determine the boundaries of water content for classes labeling, after which only a cheap camera is required from the equipment to transmit new images for processing by the algorithm. |
| format | Article |
| id | doaj-art-2b63b39ac3d94ff1a8ebc7d62a833968 |
| institution | OA Journals |
| issn | 2412-3811 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Infrastructures |
| spelling | doaj-art-2b63b39ac3d94ff1a8ebc7d62a8339682025-08-20T02:21:02ZengMDPI AGInfrastructures2412-38112025-06-0110613810.3390/infrastructures10060138Soil Condition Classification Based on Natural Water Content Using Computer Vision TechniqueMark Miller0Yong Fang1Yubo Wang2Sergey Kharitonov3Vladimir Akulich4Key Laboratory of Transportation Tunnel Engineering of Ministry of Education, School of Civil Engineering, Southwest Jiaotong University, Chengdu 610031, ChinaKey Laboratory of Transportation Tunnel Engineering of Ministry of Education, School of Civil Engineering, Southwest Jiaotong University, Chengdu 610031, ChinaKey Laboratory of Transportation Tunnel Engineering of Ministry of Education, School of Civil Engineering, Southwest Jiaotong University, Chengdu 610031, ChinaDepartment of Bridges and Tunnels, School of Railway Track, Structure and Construction, Russian University of Transport (RUT MIIT), Moscow 127055, RussiaDepartment of Theoretical Mechanics, School of Railway Track, Structure and Construction, Russian University of Transport (RUT MIIT), Moscow 127055, RussiaNatural water content affects many geotechnical parameters and geological properties of soils, which can reduce cohesion and friction, leading to potential failures in structures such as foundations, retaining walls, and slopes. Identification of the water content helps in designing effective drainage and water management systems to prevent flooding and erosion. In tunnel engineering, soil water content plays an important role as the stability of the tunnel face depends on it. This research solves the problem of classifying soil images depending on the natural water content by computer vision technology. First, laboratory soil tests were carried out, and the relationship between the amount of torque on the screw conveyor and the moisture content of the soil was established; photographs of the soil at different conditions were taken at each step of the experiment. Second, the resulting dataset after preprocessing was processed by convolutional neural network algorithms during deep learning; the transfer learning technique was used to obtain better results. As a result, seven algorithms were obtained that allow classifying the soil images, which can later be used to optimize the tunnel construction process. The best classification ability is demonstrated by the algorithm based on the DenseNet architecture (accuracy 0.9302 and loss 0.1980). The proposed model surpasses traditional approaches due to its increased automation and processing speed. Laboratory tests can be carried out only once for one type of soil in order to determine the boundaries of water content for classes labeling, after which only a cheap camera is required from the equipment to transmit new images for processing by the algorithm.https://www.mdpi.com/2412-3811/10/6/138TBMcomputer visiondeep learningCNNwater contentsoil condition classification |
| spellingShingle | Mark Miller Yong Fang Yubo Wang Sergey Kharitonov Vladimir Akulich Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique Infrastructures TBM computer vision deep learning CNN water content soil condition classification |
| title | Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique |
| title_full | Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique |
| title_fullStr | Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique |
| title_full_unstemmed | Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique |
| title_short | Soil Condition Classification Based on Natural Water Content Using Computer Vision Technique |
| title_sort | soil condition classification based on natural water content using computer vision technique |
| topic | TBM computer vision deep learning CNN water content soil condition classification |
| url | https://www.mdpi.com/2412-3811/10/6/138 |
| work_keys_str_mv | AT markmiller soilconditionclassificationbasedonnaturalwatercontentusingcomputervisiontechnique AT yongfang soilconditionclassificationbasedonnaturalwatercontentusingcomputervisiontechnique AT yubowang soilconditionclassificationbasedonnaturalwatercontentusingcomputervisiontechnique AT sergeykharitonov soilconditionclassificationbasedonnaturalwatercontentusingcomputervisiontechnique AT vladimirakulich soilconditionclassificationbasedonnaturalwatercontentusingcomputervisiontechnique |