A5047282647- Non-Destructive Testing Techniques
- Welding Techniques and Residual Stresses
- Ultrasonics and Acoustic Wave Propagation
- Retinal Imaging and Analysis
- Digital Imaging for Blood Diseases
- Power Line Inspection Robots
- Industrial Vision Systems and Defect Detection
- Geophysical Methods and Applications
- Domain Adaptation and Few-Shot Learning
- Multimodal Machine Learning Applications
- AI in cancer detection
- Retinal and Optic Conditions
- Magnetic Field Sensors Techniques
- Advanced Electrical Measurement Techniques
- Electrical Fault Detection and Protection
- Sensor Technology and Measurement Systems
- Power Transformer Diagnostics and Insulation
- Flow Measurement and Analysis
- Magnetic Properties and Applications
- Structural Health Monitoring Techniques
- Hydrogen embrittlement and corrosion behaviors in metals
- Advanced Surface Polishing Techniques
- High voltage insulation and dielectric phenomena
- Astronomical Observations and Instrumentation
University of Manchester
2020-2024
Wuhan University of Technology
2024
University Hospital of Bern
2019-2021
In this work, we propose an AI-based method that intends to improve the conventional retinal disease treatment procedure and help ophthalmologists increase diagnosis efficiency accuracy. The proposed is composed of a deep neural networks-based (DNN-based) module, including identifier clinical description generator, DNN visual explanation module. To train validate effectiveness our DNN-based large-scale image dataset. Also, as ground truth, provide dataset manually labeled by qualitatively...
Lift-off effect, resulting in detection error, which is a significant challenge that needs to be addressed defect using eddy current testing (ECT). A novel array sensor, with each unit composing of 4 excitation coils and receiving coil, proposed implement fast large-area scanning for metal sheet without lift-off interference. In addition, theoretical analysis based on Biot-Savart law applied illustrate the feasibility sensor eliminating effect. However, has also an angular sensitivity means...
Pulsed eddy current (PEC) testing is a non-destructive (NDT) technique that widely used in the industry. The ability to detect corrosion metal materials crucial for ensuring safety and mitigating potential hazards. This paper provides solution from hardware software show of deep learning (DL) method process PEC data under multiple complex distortions, we also deliver real-time DL edge computing system thickness recognition. There are three key contributions. Firstly, constructed new dataset...
Eddy current testing (ECT) is an effective technique for evaluating depth of metal surface defects. However, in practice, evaluation primarily relies on the experience operator and often carried out by manual inspection. In this paper, we address challenges automatic defects virtual state-of-the-art deep learning (DL) techniques. The main contributions are three-fold. Firstly, a highly-integrated portable ECT device developed, taking advantage advanced field programmable gate array...
Fatigue cracking in riveted lap joints on aircraft could cause severe safety consequences flight. The prior inspection for the integrity of second layer structure with rivet removed is necessary due to priority safety. Methods such as radiographic and ultrasonic testing are widely applied but require relatively expensive equipment, mechanical scanning or professional operators. Eddy current method can be adopted non-destructive (NDT) inspection, which safe low-cost. In this paper, a...
The detection of lift-off variation presents a significant challenge in many eddy current testing (ECT) applications, where it is often considered an undesirable signal. To address this issue, the paper proposes novel 4-coil excitation sensor, which can effectively suppress effect when detecting defects varying depths planar structures. proposed sensor design leverages opposite directions between any two arbitrary adjacent coils, cancel out signal due to defect-free plate. However, has...
Eddy current based approaches have been investigated for a wide range of inspection applications. Dodd-Deeds model and the truncated region eigenfunction expansion (TREE) method are widely applied in various occasions, mostly cases that sample is relatively larger than radius sensor coil. The TREE converts integral expressions to summation many terms region. In recent work, impedance co-axial air-cored due plate finite was calculated by modified analytical approach proposed authors. this...
Eddy-current testing technique has been extensively explored for estimating the electromagnetic property of steel plates in various industrial applications. In this article, a physics-guided deep learning (DL) method is proposed to estimate permeability plate high thickness with probe liftoff. A simplified analytical model derived, which realises single multiple frequency inductance transformation and calculates related physical properties measurement configuration. constant found,...
Pulse eddy current (PEC) is one of the key testing (ECT) techniques and it widely used in metal industry. Being able to automatically recognize thickness with PEC can make manufacturing process more efficient convenient. In this paper, main contribution three-fold. Firstly, a novel portable pulse device designed, new dataset constructed wide variety features using our device. Secondly, 1-D convolutional-based deep learning models are utilised achieve automatic recognition high accuracy....
We introduce EddyBot, a novel integration of multi-channel FPGA-based Eddy Current Testing (ECT) scanner within mobile robot designed for autonomous non-destructive testing (NDT). EddyBot distinguishes itself by leveraging advanced signal processing, real-time robotics control, and efficient data synchronization to significantly enhance inspection accuracy efficiency. Our contributions are (i) the deployment Zynq-7020 FPGA fast acquisition (ii) alongside Robot Operating System (ROS) precise...
In non-destructive testing (NDT), inspection of complex metallic parts is a challenge. At the moment, object's surface profile typically obtained from design documents or through separate dedicated measurements such as optical methods. this work, we use contactless sensing and real time feedback to control robot follow unseen surfaces in non-contact autonomous manner. An EC sensor fixed end-effector UR5 robotic arm, its signals are used pose keeping axis perpendicular local maintaining...
Coercivity is the strength of reverse magnetic field required to demagnetize a material after saturation, and it an indication hardness materials. Air gaps cause errors in coercivity measurement referred as lift-off effect. This paper proposes new method address this issue by incorporating additional inductance measurements formulating calibration method. The principle based on fact that both change with variation air gaps. starts finding how changes between sensor sample, then derives...
Eddy current based approaches have been investigated for a wide range of inspection applications. Dodd-Deeds model and the truncated region eigenfunction expansion (TREE) method are widely applied in various occasions, mostly cases that sample is relatively larger than radius sensor coil. The TREE converts integral expressions to summation many terms region. In recent work, impedance co-axial air-cored due plate finite was calculated by modified analytical approach proposed authors. this...
Lift-off variation is a disturbing signal for defect detection using eddy current testing (ECT), hence, novel 4-coil excitation sensor proposed to suppress lift-off effect in detecting defects with different depths this paper. The can hold back effectively the process of due opposite direction between two arbitrary adjacent coils, while structure leads detect signals under angles and sensor. Simulation applied verify effectiveness resisting vibration. In addition, difference also illustrated...
Coercivity is the strength of reverse magnetic field required to demagnetize a material after saturation, and it an indication hardness materials. Airgaps causes errors in coercivity measurement referred as lift-off effect. This paper proposes new method address this issue by incorporating additional inductance measurements formulating calibration method. The principle based on fact that both inductances change with variation airgaps. starts from finding how changes airgaps between sensor...