A5102846946- Seismic Imaging and Inversion Techniques
- Hydraulic Fracturing and Reservoir Analysis
- Seismic Waves and Analysis
- Drilling and Well Engineering
- Infrastructure Maintenance and Monitoring
- Advanced Neural Network Applications
- Water Systems and Optimization
- Video Surveillance and Tracking Methods
- Geophysical Methods and Applications
- Hydrocarbon exploration and reservoir analysis
- Automated Road and Building Extraction
Tongji University
2017-2024
Seismic data processing requires careful interpolation or reconstruction to restore the regularly irregularly missing traces. In practice, seismic with consecutively traces are quite common, which will lead a great challenge for conventional methods. To effectively reconstruct successively blank in data, we proposed self-supervised deep learning approach, convolutional neural network is trained supervised manner pseudolabels obtained from unlabeled observed data. The automatically generated...
Summary The need for accurate and efficient seismic data interpolation is increasing. Insufficient spatial sampling may bring negative effects on existing migration inversion methods. In recent years, Deep Learning (DL) techniques have shown great potential various tasks. It can extract high-level features from through self-learning mechanism. this paper, we focus severe under-sampled problem (scale of 4) develop two Convolutional Neural Networks (CNN) based models. We build a deep cascaded...
Summary The accurate separation of single-mode waves from multi-component seismic data is great significance for elastic imaging and inversion. Traditional methods require velocity information do not perform well on the far offset. In this paper, we propose to use deep convolutional neural networks P-S tasks. We design a training testing workflow that can handle arbitrary size. train model one synthetic dataset directly evaluate trained another without re-training or fine-tuning process. Our...
Abstract In order to achieve the governance of urban water environment systems and ensure normal operation drainage systems, it is necessary regularly inspect pipe network detect identify defects. Traditional methods manually identifying defects in Close Circuit Television (CCTV) data annotating them are labor-intensive inefficient. To address this issue, a method for automatic identification based on convolutional neural (CNN) transformer proposed. The model named RFCBAM-CGA-RTDETR has been...