<p>An on-line quality assessment system in the industry is essential to prevent artifacts and guide manufacturing processes. Some well-developed systems can diagnose problems and help control the output qualities. However, some of the conventional methods are limited in time consumption and cost of expensive human labor. So, more efficient solutions are needed to guide future decisions and improve productivity. This thesis focuses on developing two image-based machine learning systems to accelerate the manufacturing process: one is to benefit nitrate sensor fabrication, and the other is to help image quality control for inkjet printers.</p> <p><br></p> <p>In the first work, we propose a system for predicting the nitrate sensor's performance based on non-contact images. Nitrate sensors are commonly used to reflect the nitrate levels of soil conditions in agriculture. In a roll-to-roll system, for manufacturing thin-film nitrate sensors, varying characteristics of the ion-selective membrane on screen-printed electrodes are inevitable and affect sensor performance. It is essential to monitor the sensor performance in real-time to guarantee the quality of the sensor. We also develop a system for predicting the sensor performance in on-line scenarios and making the neural networks efficiently adapt to the new data.</p> <p><br></p> <p>Streaks are the number one image quality problem in inkjet printers. In the second work, we focus on developing an efficient method to model and predict missing jets, which is the main contributor to streaks. In inkjet printing, the missing jets typically increase over printing time, and the print head needs to be purged frequently to recover missing jets and maintain print quality. We leverage machine learning techniques for developing spatio-temporal models to predict when and where the missing jets are likely to occur. The prediction system helps the inkjet printers make more intelligent decisions during customer jobs. In addition, we propose another system that will automatically identify missing jet patterns from a large-scale database that can be used in a diagnostic system to identify potential failures.</p>

Sepsis is a life-threatening condition that neonatal patients are especially susceptible to. Fortunately, improved bedside monitoring has enabled the collection and use of continuous vital signs data for the purpose of detecting conditions such as sepsis. While current research has found some success in reducing mortality in neonatal intensive care units with linear directly interpretable models, such as logistic regression, accurate detection of sepsis from inherently noisy time-series data still remains a challenge. Furthermore, previous research has generally relied on pre-defined features extracted from rawvital signs data, which may not be optimal for the detection task. Therefore, assessing the overall feasibility of sequential deep learning models, such as recurrent and convolutional models, could improve the results of current research. This task was tackled in three phases. Firstly, baseline scores were established with a logistic regression model. Secondly, three common recurrent classifiers were tested on pre-defined window based features and compared with each other. Thirdly, a convolutional architecture with a recurrent and non-recurrent classifier was tested on raw low frequency (1Hz) signals in order to examine their capability to automatically extract features from the data. The final results from all phases were compared with each other. Results show that recurrent classifiers trained on pre-defined features do outperform automatic feature extraction with the convolutional models. The best model was based on a long-short term memory unit that achieved an area under the characteristic receiver operating unit curve of 0.806, and outperformed the established baseline results. In comparison with previous research, said model performed on par with the examined simple interpretable baseline models. The low results can likely be attributed to a insufficient sample size of patients with sepsis for the examined models and sub-optimal hyperparameter optimization due to the number of possible configurations. Further avenues of research include examination of high frequency data and more complex models for automatic feature extraction. / Sepsis är ett livshotande tillstånd som neonatala patienter är särskilt mottagliga för. Lyckligtvis har förbättrad patientmonitorering möjliggjort kontinuerlig insamling och andvänding av vitalparametrar i syfte att upptäcka tillstånd som sepsis. Medan aktuell forskning har funnit viss framgång i att minska dödligheten på neonatala intensivvårdsavdelningar med hjälp av linjära tolkbara modeller, såsom logistisk regression, är noggrann detektering av sepsis från brusig tidsseriedata fortfarande en utmaning. Dessutom har tidigare forskning i allmänhet förlitat sig på fördefinierade prediktorer extraherade från rå vitalparameterdata, som kanske inte är optimala för detektionsuppgiften. På grund av detta kan en bedömning av den övergripande användbarheten av sekventiella modeller för djupinlärning, såsom RNN- och CNN-modeller, förbättra resultaten av aktuell forskning. Denna uppgift tacklades i tre faser. Först och främst etablerades baslinjeresultat med en logistisk regressionsmodell. För det andra testades tre RNN-baserad klassificerare på data med fördefinierade fönsterbaserade prediktorer och jämfördes med varandra. För det tredje testades en CNN-arkitektur med både en RNN-klassificerare och MLP-klassificerare på råa lågfrekventa (1Hz) signaler för att undersöka deras förmåga att automatiskt extrahera egna prediktorer från datan. Slutresultaten från alla faser jämfördes med varandra. Resultaten visar att RNN-klassificerare som tränats på fördefinierade prediktorer överträffar automatisk extraktion av prediktorer med CNN-modellerna. Den bäst presterande modellen baserades på en långtidsminnesenhet som uppnådde en AUROC på 0.806, och överträffade de etablerade baslinjeresultaten. I jämförelse med tidigare forskning uppnådde ifrågavarande modell lika hög prestation som de väl undersökta enklare tolkbara baslinjemodellerna. De låga resultaten kan sannolikt tillskrivas en otillräcklig provstorlek av patienter med sepsis för de undersökta modellerna och suboptimal hyperparameteroptimering på grund av antalet möjliga konfigurationer. Ytterligare forskningsvägar inkluderar undersökning av högfrekventa data och mer komplexa modeller för automatisk extraktion av prediktorer.

In synthetic aperture radar (SAR) and inverse synthetic aperture radar (ISAR), an imaging radar emits electromagnetic waves of varying frequencies towards a target and the backscattered waves are collected. By either moving the radar antenna or rotating the target and combining the collected waves, a much longer synthetic aperture can be created. These radar measurements can be used to determine the radar cross-section (RCS) of the target and to reconstruct an estimate of the target. However, the reconstructed images will suffer from spectral leakage effects and are limited in resolution. Many methods of enhancing the images exist and some are based on deep learning. Most commonly the deep learning methods rely on high-resolution ground truth data of the scene to train a neural network to enhance the radar images. In this thesis, a method that does not rely on any high-resolution ground truth data is applied to train a convolutional neural network to enhance radar images. The network takes a conventional ISAR image subject to spectral leakage effects as input and outputs an enhanced ISAR image which contains much more defined features. New RCS measurements are created from the enhanced ISAR image and the network is trained to minimise the difference between the original RCS measurements and the new RCS measurements. A sparsity constraint is added to ensure that the proposed enhanced ISAR image is sparse. The synthetic training data consists of scenes containing point scatterers that are either individual or grouped together to form shapes. The scenes are used to create synthetic radar measurements which are then used to reconstruct ISAR images of the scenes. The network is tested using both synthetic data and measurement data from a cylinder and two aeroplane models. The network manages to minimise spectral leakage and increase the resolution of the ISAR images created from both synthetic and measured RCSs, especially on measured data from target models which have similar features to the synthetic training data.  The contributions of this thesis work are firstly a convolutional neural network that enhances ISAR images affected by spectral leakage. The neural network handles complex-valued signals as a single channel and does not perform any rescaling of the input. Secondly, it is shown that it is sufficient to calculate the new RCS for much fewer frequency samples and angular positions and compare those measurements to the corresponding frequency samples and angular positions in the original RCS to train the neural network.

<p>  </p> <p>Semantic segmentation is a machine learning task that is seeing increased utilization in multiples fields, from medical imagery, to land demarcation, and autonomous vehicles. Semantic segmentation performs the pixel-wise classification of images, creating a new, segmented representation of the input that can be useful for detected various terrain and objects within and image. Recently, convolutional neural networks have been heavily utilized when creating neural networks tackling the semantic segmentation task. This is particularly true in the field of autonomous driving systems.</p> <p>The requirements of automated driver assistance systems (ADAS) drive semantic segmentation models targeted for deployment on ADAS to be lightweight while maintaining accuracy. A commonly used method to increase accuracy in the autonomous vehicle field is to fuse multiple sensory modalities. This research focuses on leveraging the fusion of long wave infrared (LWIR) imagery with visual spectrum imagery to fill in the inherent performance gaps when using visual imagery alone. This comes with a host of benefits, such as increase performance in various lighting conditions and adverse environmental conditions. Utilizing this fusion technique is an effective method of increasing the accuracy of a semantic segmentation model. Being a lightweight architecture is key for successful deployment on ADAS, as these systems often have resource constraints and need to operate in real-time. Multi-Spectral Fusion Network (MFNet) [ 1 ] accomplishes these parameters by leveraging a sensory fusion approach, and as such was selected as the baseline architecture for this research.</p> <p>Many improvements were made upon the baseline architecture by leveraging a variety of techniques. Such improvements include the proposal of a novel loss function categorical cross-entropy dice loss, introduction of squeeze and excitation (SE) blocks, addition of pyramid pooling, a new fusion technique, and drop input data augmentation. These improvements culminated in the creation of the Fast Thermal Fusion Network (FTFNet). Further improvements were made by introducing depthwise separable convolutional layers leading to lightweight FTFNet variants, FTFNet Lite 1 & 2.</p>

Detecting failures in tricone drill bits is crucial in the mining industry due to their potential consequences, including operational losses, safety hazards, and delays in drilling operations. Timely identification of failures allows for proactive maintenance and necessary measures to ensure smooth drilling processes and minimize associated risks. Accurate failure detection helps mining operations avoid financial losses by preventing unplanned breakdowns, costly repairs, and extended downtime. Moreover, it optimizes operational efficiency by enabling timely maintenance interventions, extending the lifespan of drill bits, and minimizing disruptions. Failure detection also plays a critical role in ensuring the safety of personnel and equipment involved in drilling operations. Traditionally, failure detection in tricone drill bits relies on manual inspection, which can be time-consuming and labor-intensive. Incorporating artificial intelligence-based approaches can significantly enhance efficiency and accuracy. This thesis uses machine learning methods for failure inference in tricone drill bits. A classic Convolutional Neural Network (CNN) classification method was initially explored, but its performance was insufficient due to the small dataset size and imbalanced data. The problem was reformulated as an object detection task to overcome these limitations, and a post-processing operation was incorporated. Data augmentation techniques enhanced the training and evaluation datasets, improving failure detection accuracy. Experimental results highlighted the need for revising the initial CNN classification method, given the limitations of the small and imbalanced dataset. However, You Only Look Once (YOLO) algorithms such as YOLOv5 and YOLOv8 models exhibited improved performance. The post-processing operation further refined the results obtained from the YOLO algorithm, specifically YOLOv5 and YOLOv8 models. While YOLO provides bounding box coordinates and class labels, the post-processing step enhanced drill bit failure detection through various techniques such as confidence thresholding, etc. By effectively leveraging the YOLO-based models and incorporating post-processing, this research advances failure detection in tricone drill bits. These intelligent methods enable more precise and efficient detection, preventing operational losses and optimizing maintenance processes. The findings underscore the potential of machine learning techniques in the mining industry, particularly in mechanical drilling, driving progress and enhancing overall operational efficiency

Lane detection is a crucial task in the field of autonomous driving and advanced driver assistance systems. In recent years, convolutional neural networks (CNNs) have been the primary approach for solving this problem. However, interesting findings from recent research works regarding the use of Transformer models and attention-based mechanisms have shown to be beneficial in the task of semantic segmentation of the road lane markings. In this work, we investigate the effectiveness of incorporating a Vision Transformer (ViT) to process feature maps extracted by a CNN network for lane detection. We compare the performance of a baseline CNN-based lane detection model with that of a hybrid CNN-ViT pipeline and test the model over a well known dataset. Furthermore, we explore the impact of incorporating temporal information from a road scene on a lane detection model’s predictive performance. We propose a post-processing technique that utilizes information from previous frames to improve the accuracy of the lane detection model. Our results show that incorporating temporal information noticeably improves the model’s performance, and manages to make effective corrections over the originally predicted lane masks. Our SegNet backbone, exploiting the proposed post-processing mechanism, reached an F1 scoreof 0.52 and Intersection-over-Union (IoU) of 0.36 over the TuSimple test set. However, the findings from the testing of our CNN-ViT pipeline and a relevant ablation study, do indicate that this hybrid approach might not be a good fit for lane detection. More specifically, the ViT module fails to exploit the feature sextracted by our CNN backbone and therefore, our hybrid pipeline results in less accurate lane marking spredictions.

La hausse actuelle du climat pousse les espèces d’arbres tempérés à migrer vers le nord. En vue de comprendre comment certaines espèces réagiront face à cette migration, nous pouvons porter notre regard vers les populations marginales. Les études paléoécologiques de ces populations – situées au-delà de l’aire de répartition continue de l’espèce – peuvent nous informer quant aux conditions écologiques nécessaires à leur migration. Ce mémoire analyse un peuplement d’érables à sucre (Acer saccharum Marsh.) situé à la limite nordique de la répartition de l’espèce, dans la forêt tempérée mixte québécoise. L’objectif de la recherche est d’identifier quand et sous quelles conditions écologiques A. saccharum s’est établi en situation marginale. À ces fins, cette étude propose l’analyse des fossiles extraits des sédiments lacustres d’un lac situé à proximité de l’érablière. Un modèle d’apprentissage-machine est entraîné à l’aide d’images de pollens et permet la classification des pollens extraits des sédiments lacustres – le premier de la sorte. Notre méthode proposée emploi un protocole d’extraction fossile accéléré et des réseaux de neurone convolutifs permettant de classifier les pollens des espèces les plus retrouvées dans les sédiments quaternaires du nord-est de l’Amérique. Bien qu’encore incapable de classifier précisément toutes les espèces présentes dans une telle séquence fossile, notre modèle est une preuve de concept envers l’automatisation de la paléo-palynologie. Les résultats produits par le modèle combinés à l’analyse des charbons fossiles permettent la reconstruction de la végétation et des feux des 10,000 dernières années. L’établissement régional d’A. saccharum est daté à 4,800 cal. BP, durant une période de refroidissement climatique et de feux fréquents mais de faible sévérité. Sa présence locale est prudemment établie à 1,200 cal. BP. Les résultats de ce mémoire soulignent le potentiel de la paléo-palynologie automatique ainsi que la complexité de l’écologie d’A. saccharum. / The current global climate warming is pushing temperate tree species to migrate northwards. To understand how certain species will undergo this migration, we can look at marginal populations. The paleoecological studies of such populations, located beyond the species’ core distribution range, can inform us of the conditions needed for a successful migration. This research thesis analyses a sugar maple (Acer saccharum Marsh.) stand located at the northern boundary of the species’ limit, in Québec’s mixed-temperate forest. The objective of this research is to identify when and under which ecological conditions did A. saccharum establish itself as the stand’s dominant species. To that end, this study analyses the fossil record found in a neighbouring lake’s organic sediments. A machine learning-powered model is trained using pollen images to classify the lacustrine sediment’s pollen record. The first of its kind, our proposed method employs an accelerated fossil pollen extraction protocol and convolutional neural networks and can classify the species most commonly found in Northeastern American Quaternary fossil records. Although not yet capable of accurately classifying a complete fossil pollen sequence, our model serves as a proof of concept towards automation in paleo-palynology. Using results generated by our model combined with the analysis of the fossil charcoal record, the past 10,000 years of vegetation and fire history is reconstructed. The regional establishment of A. saccharum is conservatively dated at 4,800 cal. BP, during a period of climate cooling and frequent, although non-severe, forest fires. Its local presence can only be attested to since 1,200 cal. BP. This thesis’ results highlight both the potential of automated paleo-palynology and the complexity of A. saccharum’s ecological requirements.

Neonatal sepsis is associated with most cases ofmortality in the neonatal intensive care unit. Major challengesin detecting sepsis using suitable biomarkers has lead people tolook for alternative approaches in the form of Machine Learningtechniques. In this project, Random Forest classification wasperformed on a sepsis data set provided by Karolinska Hospital.We particularly focused on tackling class imbalance in the datausing sampling and cost-sensitive techniques. We compare theclassification performances of Random Forests in six differentsetups; four using oversampling and undersampling techniques;one using cost-sensitive learning and one basic Random Forest.The performance with the oversampling techniques were betterand could identify more sepsis patients than the other setups.The overall performances were also good, making the methodspotentially useful in practice. / Neonatal sepsis är orsaken till majoriteten av mortaliteten i neonatal intensivvården. Svårigheten i att detektera sepsis med hjälp av biomarkörer har lett många att leta efter alternativa metoder. Maskininlärningstekniker är en sådan alternativ metod som har i senaste tider ökat i användning inom vård och andra sektorer. I detta project användes Random Forest klassifikations algoritmen på en sepsis datamängd given av Karolinska Sjukhuset. Vi fokuserade på att hantera klassimbalansen i datan genom att använda olika provtagningsoch kostnadskänsliga metoder. Vi jämförde klassificeringsprestanda för Random Forest med sex olika inställningar; fyra av de använde provtagingsmetoderna; en av de använde en kostnadskänslig metod och en var en vanlig Random Forest. Det visade sig att modellens prestanda ökade som mest med översamplings metoderna. Den generella klassificeringsprestandan var också bra, vilket gör Random Forests tillsammans med ingsmetoderna potentiellt användbar i praktiken. / Kandidatexjobb i elektroteknik 2022, KTH, Stockholm

Subsurface porous media, like aquifers, petroleum reservoirs, and geothermal systems, are vital for natural resources and environmental management. Extensive research has been conducted to understand flow and transport in these media, addressing challenges in hydrocarbon extraction, carbon storage and waste management. Classifying the type of porous media (e.g., sandstone, carbonate) is often the first step in the rock characterization process, and it provides critical information regarding the physical properties of the porous media. Therefore, we utilize multivariate statistical methods with discriminant analysis to categorize porous media samples which proved to be efficient by achieving excellent classification accuracy on testing datasets and served as a surrogate tool to study key porous media characteristics. While recent advances in three-dimensional (3D) imaging of core samples have enabled digital subsurface characterization, the exorbitant computational cost associated with direct numerical simulation in 3D remains a persistent challenge. In contrast, machine learning (ML) models are much more efficient, though their use in subsurface characterization is still in its infancy. Therefore, we introduce a novel 3D convolution neural network (CNN) for end-to-end prediction of permeability. By increasing dataset size, diversity, and optimizing the network architecture, our model surpasses the accuracy of existing 3D CNN models for permeability prediction. It demonstrates excellent generalizability, accurately predicting permeability in previously unseen samples. However, despite the efficiency of the developed 3D CNN model for accurate and fast permeability prediction, its utility remains limited to small subdomains of the digital rock samples. Therefore, we introduce an upscaling technique using a new analytical solution to calculate effective permeability in a 3D digital rock composed of 2 × 2 × 2 anisotropic cells. By incorporating this solution into physics-informed neural network (PINN) models, we achieve highly accurate results. Even when upscaling previously unseen samples at multiple levels, the PINN with the physics-informed module maintains excellent accuracy. This advancement enhances the capability of ML models, like 3D CNN, for efficient and accurate digital rock analysis at the core scale. After successfully applying ML models in permeability prediction, we now extend their application to another important parameter in subsurface engineering projects: effective thermal conductivity, which is a key parameter in engineering projects like radioactive waste repositories, geothermal energy production, and underground energy storage. To address the need for large training data and processing power in ML models, we propose a novel framework based on transfer learning. This approach allows prior knowledge from previous applications to be transferred, resulting in faster and more efficient implementation of new relevant applications. We introduce CNN models trained on various porous media samples that leverage transfer learning to predict porous media sample thermal conductivity accurately. Our approach reduces training time, processing power, and data requirements, enabling effective prediction and analysis of porous media properties such as permeability and thermal conductivity. It also facilitates the application of ML to other properties, improving efficiency and accuracy. / Thesis / Doctor of Philosophy (PhD)

The survival rate for a suddent cardiac arrest (SCA) is incredibly low, with less than one in ten surviving; most SCAs occur outside of a hospital setting. There is a need to develop an effective and efficient system that can sense, communicate and remediate potential SCA situations on a near real-time basis. This research presents a novel Zeolite-PDMS-based optically unobtrusive flexible dry electrodes for biosignal acquisition from various subjects while at rest and in motion. Two zeolite crystals (4A and 13X) are used to fabricate the electrodes. Three different sizes and two different filler concentrations are compared to identify the better performing electrode suited for electrocardiogram (ECG) data acquisition. A low-power, low-noise amplifier with chopper modulation is designed and implemented using the standard 180nm CMOS process. A commercial off-the-shelf (COTS) based wireless system is designed for transmitting ECG signals. Further, this dissertation provides a framework for Machine Learning Classification algorithms on large, open-source Arrhythmia and SCA datasets. Supervised models with features as the input data and deep learning models with raw ECG as input are compared using different methods. The machine learning tool classifies the datasets within a few minutes, saving time and effort for the physicians. The experimental results show promising progress towards advancing the development of a wireless ECG recording system combined with efficient machine learning models that can positively impact SCA outcomes.