- Advanced Vision and Imaging
- Image and Object Detection Techniques
- Image Processing Techniques and Applications
- Mineral Processing and Grinding
- Medical Image Segmentation Techniques
- Advanced Sensor and Control Systems
- Data Management and Algorithms
- Video Analysis and Summarization
- Face recognition and analysis
- Robotics and Sensor-Based Localization
- Digital Imaging for Blood Diseases
- Plant Pathogens and Fungal Diseases
- Brain Tumor Detection and Classification
- Face and Expression Recognition
- Adaptive Dynamic Programming Control
- Seismic Waves and Analysis
- Data Visualization and Analytics
- Multimodal Machine Learning Applications
- Medical Imaging and Analysis
- Advanced Image Fusion Techniques
- Remote Sensing and LiDAR Applications
- Time Series Analysis and Forecasting
- Geotechnical Engineering and Underground Structures
- Advanced Neural Network Applications
- Plant Disease Management Techniques
Qingdao University of Technology
2022-2024
Shenzhen University
2024
Shenzhen MSU-BIT University
2024
Qingdao University of Science and Technology
2022-2023
Shandong University of Science and Technology
2018-2020
Abstract The laying trajectory of saline-alkali land subsurface pipes for salt discharge plays a crucial role in assessing the efficiency reclamation. Due to complex soil conditions and physical properties pipes, directly reconstructing using an inertial measurement unit (IMU) often fails achieve sufficient accuracy. To address this issue, paper proposes reconstruction method based on improved extended Kalman filter (EKF) adaptive gradient optimization. This integrates accelerometer...
As a specific form of knowledge distillation (KD), self-knowledge enables student network to progressively distill its own without relying on pretrained, complex teacher network; however, recent studies self-KD have discovered that additional dark captured by auxiliary architecture or data augmentation could create better soft targets for enhancing the but at cost significantly more computations and/or parameters. Moreover, most existing methods extract label as supervisory signal from...
With the rapid use of Android OS in mobile devices and related products, face recognition technology is an essential feature, so that have a strong personal identity authentication. In this paper, we propose based software development framework for real-time detection using OpenCV library, which applicable several applications. Initially, Gaussian smoothing gray-scale transformation algorithm applied to preprocess source image. Then, Haar-like feature matching method used describe...
Geological exploration plays a fundamental and crucial role in geological engineering. The most frequently used method is to obtain borehole videos using an axial view camera system (AVBCS) pre-drilled borehole. This approach surveying the internal structure of based on video playback screenshot analysis. One drawbacks AVBCS that it provides only qualitative description information with forward-looking video, but quantitative analysis data, such as width dip angle fracture, are unavailable....
In this article, we present an efficient coding scheme for LiDAR point cloud maps. As a map consists of numerous single scans spliced together, by recording the time stamp and quaternion matrix each scan during building, cast compression into sequence problem. The architecture includes two techniques: intra-coding inter-coding. For intra-frames, segmentation-based intra-prediction technique is developed. inter-frames, interpolation-based inter-frame network explored to remove temporal...
Analyzing geological drilling hole images acquired by Axial View Panoramic Borehole Televiewer (APBT) is a key step to explore the structure in exploration. Conventionally, borehole are examined technicians, which inefficient and subjective. In this paper, three dominant types of borehole-wall on coal-rock mass structure, namely, border images, fracture intact rock mainly studied. The traditional image classification methods based unified feature extraction algorithm single classifier not...
Due to the influence of shooting environment and inherent image characteristics, there is a large amount interference in process stitching geological borehole video. To accurately match acquired sequences inner part borehole, this paper presents new method an unfolded image, which uses generated from video construct large-scale panorama. Firstly, speeded-up robust feathers (SURF) algorithm used extract feature points complete rough matching. Then, M-estimator sample consensus (MSAC)...
The role played by Axial View Borehole Camera System (AVBCS) in acquiring borehole images and analyzing geographical structures opens new possibilities exploration. Considering the current situations, where are interpreted humans, a image classification method based on NSCT Hu moment invariants is proposed to classify them into three main classes of border, fracture intact rock mass images. Firstly, decomposed one low frequency subband multiple band pass directional coefficients NSCT, obtain...
A new fixed-time adaptive neural network control strategy is designed for pure-feedback non-affine nonlinear systems with state constraints according to the feedback signal of error system. Based on backstepping technology, Lyapunov function each subsystem. The used identify unknown parameters system in a fixed-time, and makes output track expected fixed-time. Through stability analysis, it proved that tracking converges design upper bound setting time only needs modify law controlled...
The use of in-hole imaging to investigate geological structure characteristics is one the crucial methods for study rock mass stability and engineering design. images are usually influenced by lighting characteristics, resulting in presence interference noise regions consequently impacting classification accuracy. To enhance analytical efficacy images, this paper employs proposed optimal non-concentric ring segmentation method establish a new database. This establishes transformation...