| 1. |
|
|
| 2. |
|
|
| 3. |
- Cheng, Peng
(författare)
-
Applications of embedded sensors in loader crane positioning and rotor RPM measurement
- 2011
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- In this thesis, two novel applications involving embedded sensors arestudied, one dealing with loader crane positioning and the other involving rotorRevolutions Per Minute (RPM) measurement. The thesis presents a generalintroduction to the embedded sensor, its architecture and its use in mechanicalindustry, and provides the reader with an overview of conventional sensortechnologies within the fields of angle sensors and angular speed sensors, coveringtheir working principles, features, advantages and disadvantages and typicalapplications. The particular problems associated with the use of conventionalsensors in both loader crane positioning and rotor RPM measurement aredescribed and these problems provided the motivation for the designs of theembedded sensor systems developed in this thesis.In the case of the loader crane positioning, the origins of the project and thespecial requirements of the application are described in detail. In addition, apreliminary study is conducted in relation to the idea of a contactless joint angularsensor using MEMS inertial sensors in which four different methods, namely, theCommon-Mode-Rejection with Gyro Integration (CMRGI), Common-Mode-Rejection (CMR), Common-Mode-Rejection with Gyro Differentiation (CMRGD)and Distributed Common-Mode-Rejection (DCMR), are conceived, modeled andtested on a custom-designed prototype experimental setup. The results gatheredfrom these four methods are compared and analyzed in order to identify thedifferences in their performances. The methods, which proved to be suitable, arethen further tested using the prototype sensor setup on a loader crane and theperformance results are analyzed in order to make a decision in relation to the twomost suitable methods for the application of the loader crane positioning. Theresults suggested that the two most suitable were the CMRGD and the DCMR. Thepractical design issues relating to this sensor system are highlighted andsuggestions are made in the study. Additionally, possible future work for thisproject is also covered.In the first case for the rotor RPM measurement, the thesis presents themodeling and simulation of the stator-free RPM sensor idea using the Monte Carlomethod, which demonstrated the special features and performance of this sensor.The design aspects of the prototype sensor are described in detail and theprototype is tested on an experimental setup. The conclusions for the stator-freeRPM sensor are then made from the analysis of the experimental results and futurework in relation to this sensor is also proposed.In the second case of the rotor RPM measurement, the thesis presentsanother idea involving the laser mouse RPM sensor and the main focus of thestudy is on the performance characterization of the laser mouse sensor and theverification of the RPM sensor idea. Experiments are conducted using the test setup and results are gathered and analyzed and conclusions are drawn.Possibilities in relation to future work for this laser mouse RPM sensor are alsoprovided.The summary and the conclusion form the final chapter of the thesis andseveral important aspects of the designs relating to both the loader cranepositioning project and the rotor RPM measurement project are discussed.
|
|
| 4. |
|
|
| 5. |
- Lundgren, Jan, 1977-
(författare)
-
Simulating Behavioral Level On-Chip Noise Coupling
- 2007
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- In this thesis, noise coupling simulation is introduced into the behavioral level. Methods andmodels for simulating on-chip noise coupling at the behavioral level in a design flow are presentedand verified for accuracy and validity. Today, designs of electronic systems are becoming denserand more and more mixed-signal systems such as System-on-Chip (SoC) are being devised. Thisraises problems when the electronics components start to interfere with each other. Often, digitalcomponents disturb analog components, introducing noise into the system causing degradation ofthe performance or even introducing errors into the functionality of the system.Today, these effects can only be simulated at a very late stage in the design process, causinglarge design iterations and increased costs if the designers are required to return and makealterations, which may have occurred at a very early stage in the process.This is why the focus of this work is centered on extracting noise coupling simulation modelsthat can be used at a very early design stage, such as at the behavioral level and then follow thedesign through the various design stages. To achieve this, SystemC is selected as a platform andimplementation example for the behavioral level models. SystemC supports design refinement,which means that when designs are being refined and are crossing the design levels, the noisecoupling models can also be refined to suit the current design.This new method of thinking in primarily mixed-signal designs is called Behavioral levelNoise Coupling (BeNoC) simulation and shows great promise in enabling a reduction in the costsof design iterations due to component cross-talk and simplifies the work for mixed-signal systemdesigners.
|
|
| 6. |
- Nie, Yali
(författare)
-
Deep Learning Approaches towards Skin Lesion Classification with Dermoscopic Images
- 2023
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Melanoma is a skin cancer that tends to be deadly. The incidence of melanoma is currently at the highest level ever recorded in Europe, North America and Oceania. The survival rate can be significantly increased if skin lesions are identified in dermoscopic images at an early stage. In the other hand, the classification of skin lesions is incredibly challenging. Skin lesion classification using deep learning approaches has provided better results in classifying skin diseases than those of dermatologist, which is lifesaving in terms of diagnosis.This thesis presents a review of our research articles on classifying skin lesions using deep learning. Regarding the research, I have four goals concerning research frontier work, small datasets, data imbalance, and improving accuracy. In this thesis, I discuss how deep learning can classify skin diseases, summarizing the problems that remain at this stage and the outlook for the future.For the above goals, I first studied and summarized more than 200 highguality articles published over five years. I then used three versions of You only look once (Yolo) to detect skin lesions. Although there were only 200 pictures, the test was very effective for detection. I applied the five-fold algorithm to Vgg_16, trained five models, and fused them so solve the small data problem. To improve the accuracy, I also tried to combine the traditional machine learning method, i.e., the seven-point checklist, with three different backbones. Since the learning rate. Then, I also tried to use the hybrid model, combining convolutional neural networks (CNN) and transformer to train the dataset, and applied focal loss to balance the extremely unbalanced weight of the data.In addition to high-quality data sets and high-performance computers being extremely important in the research and application of deep learning, the optimization of machine learning algorithms for skin lesions can be endless
|
|
| 7. |
- Shaikh, Muhammad Saad
(författare)
-
Hyperspectral imaging for in-situ applications : Methods to improve the classification of materials using hyperspectral imaging
- 2024
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- This thesis addresses several research questions related to in-situ hyperspectral imaging systems, proposes measurement methods for more accurate imaging, and examines the impact of the methods on material classification.First, the thesis investigates the possibility of successfully calibrating a hyperspectral imaging system using a low-cost PTFE reference. A hyperspectral imaging system and practical calibration procedure using an inexpensive calibration reference are introduced. This reference enables accurate measurement of a material’s reflectance spectra independent of lighting and the camera’s spectral distribution of intensity and sensitivity. The study presents experiments conducted on winter roads covered with water, snow, and ice. The results show the robustness of the calibration and the suitability of the system for classifying materials.The thesis further focuses on increasing the dynamic range (DR) of line scanning hyperspectral cameras. A method that relies on the use of multiple exposures is proposed to increase DR, benefiting applications such as plastic detection and polymer sorting. Experiments show that the proposed method can increase the DR for hyperspectral SWIR imaging from 43 dB to 73 dB. Material classification experiments reveal significant accuracy improvements with multiple exposures for large dynamic ranges.The thesis also examines the effect of variations in relative humidity. It shows that even minor changes in humidity can significantly affect measurements. Frequent calibration and pruning of active wavelength bands are proposed as solutions to reduce the classification error rate for polymers from 20% to less than 1%.The thesis also investigates the classification of colored materials by combining visible and infrared imaging. The classification algorithm shows high overall accuracy, close to 99.9% for one test case, which also shows the potential of this approach.Finally, the use of infrared hyperspectral imaging combined with Convolutional Neural Networks (CNN) for the classification of black polymers is evaluated. CNN outperforms all traditional classification algorithms, further demonstrating the potential of the proposed method. Further research on larger and more diversified material samples is recommended.
|
|
