The rise and adaptation of the transformer architecture from natural language processing to visual tasks have proven a useful and powerful tool. Subsequent architectures such as visual transformers (ViT) and shifting window (SWIN) transformers have proven to be comparable and oftentimes exceed convolutional neural networks (CNNs) in terms of accuracy. However, for mobile vision tasks and limited hardware, the computational complexity of the transformer architecture is an impediment. This project aims to answer the question of whether the Swin Transformer can be adapted towards lightweight and low latency classification as a basis for industrial automation, and how it compares to CNNs for a specific task. A case study from the logging industry, binary classification of wooden boards on chain conveyors, will serve as the basis of this evaluation. For these purposes, a novel dataset has been collected and annotated. The results of this project include an overview of the respective architectures and their performance for different implementations on the classification task. Both architectures exhibited sufficient accuracy, while the CNN models performed best for the specific case study.

The need for products has increased during the last few years, this high demand needs to bemet with higher means of production. The use of neural networks can be the key to increasedproduction without having to compromise product quality or human workers well being. This thesislooks into the concept of reliable architectures in convolutional neural networks and how they canbe implemented. The neural networks are trained to recognize the features in images to identifycertain objects, these recognition is then compared to other models to see which of them had the bestprediction. Using multiple models creates a reliable architecture from which results can be produced,these results can then be used in combinations with algorithms to improve prediction certainty. Theaim of implementing the networks with these algorithms are to improve the results without havingto change the networks configurations.

Scalp electroencephalograms (EEGs) are used in a broad range of health care institutions to monitor and record electrical activity in the brain. EEGs are essential in diagnosis of clinical conditions such as epilepsy, seizure, coma, encephalopathy, and brain death. Manual scanning and interpretation of EEGs is time-consuming since these recordings may last hours or days. It is also an expensive process as it requires highly trained experts. Therefore, high performance automated analysis of EEGs can reduce time to diagnosis and enhance real-time applications by identifying sections of the signal that need further review.Automatic analysis of clinical EEGs is a very difficult machine learning problem due to the low fidelity of a scalp EEG signal. Commercially available automated seizure detection systems suffer from unacceptably high false alarm rates. Many signal processing methods have been developed over the years including time-frequency processing, wavelet analysis and autoregressive spectral analysis. Though there has been significant progress in machine learning technology in recent years, use of automated technology in clinical settings is limited, mainly due to unacceptably high false alarm rates. Further, state of the art machine learning algorithms that employ high dimensional models have not previously been utilized in EEG analysis because there has been a lack of large databases that accurately characterize clinical operating conditions. Deep learning approaches can be viewed as a broad family of neural network algorithms that use many layers of nonlinear processing units to learn a mapping between inputs and outputs. Deep learning-based systems have generated significant improvements in performance for sequence recognitions tasks for temporal signals such as speech and for image analysis applications that can exploit spatial correlations, and for which large amounts of training data exists. The primary goal of our proposed research is to develop deep learning-based architectures that capture spatial and temporal correlations in an EEG signal. We apply these architectures to the problem of automated seizure detection for adult EEGs. The main contribution of this work is the development of a high-performance automated EEG analysis system based on principles of machine learning and big data that approaches levels of performance required for clinical acceptance of the technology. In this work, we explore a combination of deep learning-based architectures. First, we present a hybrid architecture that integrates hidden Markov models (HMMs) for sequential decoding of EEG events with a deep learning-based postprocessing that incorporates temporal and spatial context. This system automatically processes EEG records and classifies three patterns of clinical interest in brain activity that might be useful in diagnosing brain disorders: spike and/or sharp waves, generalized periodic epileptiform discharges and periodic lateralized epileptiform discharges. It also classifies three patterns used to model the background EEG activity: eye movement, artifacts, and background. Our approach delivers a sensitivity above 90% while maintaining a specificity above 95%. Next, we replace the HMM component of the system with a deep learning architecture that exploits spatial and temporal context. We study how effectively these architectures can model context. We introduce several architectures including a novel hybrid system that integrates convolutional neural networks with recurrent neural networks to model both spatial relationships (e.g., cross-channel dependencies) and temporal dynamics (e.g., spikes). We also propose a topology-preserving architecture for spatio-temporal sequence recognition that uses raw data directly rather than low-level features. We show this model learns representations directly from raw EEGs data and does not need to use predefined features. In this study, we use the Temple University EEG (TUEG) Corpus, supplemented with data from Duke University and Emory University, to evaluate the performance of these hybrid deep structures. We demonstrate that performance of a system trained only on Temple University Seizure Corpus (TUSZ) data transfers to a blind evaluation set consisting of the Duke University Seizure Corpus (DUSZ) and the Emory University Seizure Corpus (EUSZ). This type of generalization is very important since complex high-dimensional deep learning systems tend to overtrain. We also investigate the robustness of this system to mismatched conditions (e.g., train on TUSZ, evaluate on EUSZ). We train a model on one of three available datasets and evaluate the trained model on the other two datasets. These datasets are recorded from different hospitals, using a variety of devices and electrodes, under different circumstances and annotated by different neurologists and experts. Therefore, these experiments help us to evaluate the impact of the dataset on our training process and validate our manual annotation process. Further, we introduce methods to improve generalization and robustness. We analyze performance to gain additional insight into what aspects of the signal are being modeled adequately and where the models fail. The best results for automatic seizure detection achieved in this study are 45.59% with 12.24 FA per 24 hours on TUSZ, 45.91% with 11.86 FAs on DUSZ, and 62.56% with 11.26 FAs on EUSZ. We demonstrate that the performance of the deep recurrent convolutional structure presented in this study is statistically comparable to the human performance on the same dataset. / Electrical and Computer Engineering

Drilling tool wear can significantly affect the performance of the drilling operation and add extra cost to it. Accurate detection of drilling tool condition is very important for enabling proactive maintenance, minimizing downtime, and optimizing drilling processes.  This study investigates the possibility of detecting drilling tool condition of a Raisboring machine using drilling signals with deep learning methods. Given the current situation where the operators of the machine are responsible for detecting drilling abnormalities, which introduces bias and inconsistency to the process, it is crucial to develop an automated machine health monitoring system.  The objectives of this study were to explore the effectiveness of deep learning approaches in detecting drilling tool faults based on sensor data collected during drilling operations; as well as to find out which drilling signal is most effective for this problem.  The working dataset consists of labeled samples representing two drilling tool conditions (new and worn) and includes four channels: RPM, torque, feed force, and ground acceleration signals. To implement this, both time-domain features and frequency-domain features were extracted from the drilling signals and used as input to fully connected neural networks (FCNNs) and convolutional neural networks (CNNs). Performance metrics such as accuracy, precision, recall, and F1-score were used to assess the models’ performance. The results indicate that deep learning has great potential in detecting drilling tool condition. More specifically, the vibration signal, which yielded the highest results with the different algorithms.  The study highlights the potential of deep learning techniques for real-time, automated monitoring of drilling tool condition, enabling timely maintenance interventions and enhanced operationalefficiency.

A needs-based configurator is a system or tool that assists users in customizing products based on their specific needs. This thesis investigates the challenges of obtaining data for a needs-based machine learning house configurator and identifies suitable models for its implementation. The study consists of two parts: first, an analysis of how to obtain data, and second, an evaluation of three models for implementing the needs-based solution. The analysis shows that collecting house review data for a needs-based configurator is challenging due to several factors, including how the housing market operates compared to other markets, privacy concerns, and the complexity of the buying process. To address this, future studies could consider alternative data sources, adding contextual data, and creating surveys or questionnaires. The evaluation of three models: DistilBERT, BERT fine-tuned for Swedish, and a CNN with a Swedish word embedding layer, shows that both the BERT models perform well on the generated dataset, while the CNN model underperformed. The Swedish BERT model performed the best, achieving high recall and precision metrics for k between 2 and 5. This thesis suggests that further research on needs-based configurators should focus on alternative data sources and more extensive datasets to improve performance.

Previous studies have used convolutional neural networks (CNN) to classify synthetic images created by generative adversarial networks (GANs) to confirm images as either being synthetic or natural. Similar to other research, this thesis will cover the classification of synthetic images witha CNN. However, instead of classifying images created by GANs, a latent diffusion based generator is covered instead. This comparative study gathered results from the performance of botha human baseline as well as a CNN’s ability to classify images generated by stable diffusion and real images created by or taken by humans.The results from this study show that the CNN created greatly outperformed the human baseline when classifying the data sets over multipledifferent image domains.

This paper demonstrates the implementation of R-CNN in terms of electromyography-related signals to recognize hand gestures. The signal acquisition is implemented using electrodes situated on the forearm, and the biomedical signals are generated to perform the signals preprocessing using wavelet packet transform to perform the feature extraction. The R-CNN methodology is used to map the specific features that are acquired from the wavelet power spectrum to validate and train how the architecture is framed. Additionally, the real-time test is completed to reach the accuracy of 96.48% compared to the related methods. This kind of result proves that the proposed work has the highest amount of accuracy in recognizing the gestures.

No description available.

A proof-of-concept indirect tire-pressure monitoring system is developed using neural net- works to identify the tire pressure of a vehicle tire. A quarter-car model was developed with Matlab and Simulink to generate simulated accelerometer output data. Simulation data are used to train and evaluate a recurrent neural network with long short-term memory blocks (RNN-LSTM) and a convolutional neural network (CNN) developed in Python with Tensorflow. Bayesian Optimization via SigOpt was used to optimize training and model parameters. The predictive accuracy and training speed of the two models with various parameters are compared. Finally, future work and improvements are discussed.

The CNN-based steganalysis model is able to capture some complex statistical dependencies and also learn feature representations. The proposed model uses a word embedding layer to map the words into dense vectors thus, achieving more accurate representations of the words. The proposed model extracts both, the syntax and semantic features. Files having less than 200 words are referred to as short text. Preprocessing for short text is done through word segmenting and encoding the words into indexes according to the position of words in the dictionary. Once this is performed, the index sequences are fed to the CNN to learn the feature representations. Files containing over 200 words are considered as long texts. Considering the wide range of length variation of these long texts, the proposed model tokenized long texts into their sentence components with a relatively consistent length prior to preprocessing the data. Eventually, the proposed model uses a decision strategy to make the final decision to check if the text file contains stego text or not.