Mastitis is a common disease among cows in dairy farms. Diagnosis of the infection is today done manually, by analyzing bacteria growth on agar plates. However, classifiers are being developed for automated diagnostics using images of agar plates. Input images need to be of reasonable quality and consistent in terms of scale, positioning, perspective, and rotation for accurate classification. Therefore, this thesis investigates if a combination of image processing techniques can be used to match each input image to a pre-defined reference model. A method was proposed to identify important key points needed to register the input image to the reference model. The key points were defined by identifying the agar plate, its compartments, and its rotation within the image. The results showed that image registration with the correct key points was sufficient enough to match images of agar plates to a reference model despite any varieties in scale, position, perspective, or rotation. However, the accuracy depended on the identification of the salient features of the agar plate. Ultimately, the work proposes an approach using image registration to transform images of agar plates based on a pre-defined reference model, rather than a reference image.

The human factor accounts for nine out of ten out of all traffic accidents, and because more vehicles are being deployed on the roads, the number of accidents will increase. Because of this, various automated functions have been implemented in vehicles in order to minimize the human factor in driving. In recent year, this development has accelerated and vehicles able to perform the complete driving task without any human assistance have begun to emerge from different projects around the world. However, the autonomous vehicle still has many barriers to overcome before safe driving in traffic becomes a reality. One of these barriers is the difficulty to visually perceive the surrounding. This is partly because of the fact that something can cover the camera sensors, but it is also problematic to translate the perceived data, that the sensors are collecting, into something valuable for the passenger. The situation could be improved if wireless communications were available to the autonomous vehicle. Instead of trying to understand the surrounding by the use of camera sensors, the autonomous vehicle could obtain the necessary data via wireless communication, which was the subject of this study. The study showed that wireless communication will be significant for the autonomous vehicle in the future. The conclusion is based on the fact that wireless communication was a solution in other transport systems that have had the similar barrier as for the autonomous vehicle. There are also plans on managing the barrier via wireless communication in pilot projects related to autonomous vehicles. / Den mänskliga faktorn står för nio av tio utav alla trafikolyckor, och eftersom att allt fler fordon kommer ut på vägarna så leder det till att olycksantalet ökar. På grund av detta så har olika automatiserade funktioner applicerats i fordonet för att undvika den mänskliga faktorn i körningen. Denna utveckling har accelererat och fordon som ska kunna utföra hela det dynamiska framförandet utan mänsklig assistans har börjat utvecklas i olika projekt runt om i världen. Dock så har det autonoma fordonet många barriärer kvar att övervinna, för säkert framförande, varav en av dessa barriärer är fordonets förmåga att visuellt uppfatta omgivningen. Dels genom att något kan täcka kamerasensorerna men även att kunna omsätta det sensorerna uppfattar till något värdefullt för passageraren. Situationen skulle dock kunna förbättras om trådlös kommunikation gjordes tillgänglig för det autonoma fordonet. Istället för att försöka uppfatta omgivningen via kamerasensorer, skulle det autonoma fordonet kunna få den information som behövs via trådlös kommunikation, vilket är vad denna studie behandlade. Studien visade att trådlös kommunikation kommer att ha en betydelse för det autonoma fordonet i framtiden. Slutsatsen grundar sig på att trådlös kommunikation varit en lösning inom andra transportsystem som haft en liknande barriär som för det autonoma fordonet. Man planerar dessutom på att hantera det autonoma fordonets barriär via trådlös kommunikation i pilotprojekt i dagsläget

Ceramic materials contain several defects, one of which is porosity. At the time of writing, porosity measurement is a manual and time-consuming process performed by a human operator. With advances in deep learning for computer vision, this thesis explores to what degree convolutional neural networks and semantic segmentation can reliably measure porosity from microscope images. Combining classical image processing techniques with deep learning, images were automatically labeled and then used for training semantic segmentation neural networks leveraging transfer learning. Deep learning-based methods were more robust and could more reliably identify porosity in a larger variety of images than solely relying on classical image processing techniques.

Wheel loaders are heavy duty machines that are ubiquitous on construction sites and in mines all over the world. Fully autonomous wheel loaders remains an open problem but the industry is hoping that increasing their level of autonomy will help to reduce costs and energy consumption while also increasing workplace safety. Operating a wheel loader efficiently requires dig plans that extend over multiple dig cycles and not just one at a time. This calls for a model that can predict both the performance of a dig action and the resulting shape of the pile. In this thesis project, we use simulations to develop a data-driven artificial neural network model that can predict the outcome of a dig action. The model is able to predict the wheel loader’s productivity with an average error of 7.3% and the altered shape of the pile with an average relative error of 4.5%. We also show that automatic differentiation techniques can be used to accurately differentiate the model with respect to input. This makes it possible to use gradient-based optimization methods to find the dig action that maximises the performance of the wheel loader.

Man overboard is one of the most common and dangerous accidents that can occur whentraveling on a boat. Available research on man overboard systems with cameras have focusedon man overboard taking place from larger ships, which involves a fall from a height.Recreational boat manufacturers often use cord-based kill switches that turns of the engineif the wearer falls overboard. The aim of this thesis is to create a man overboard warningsystem based on state-of-the-art object detection models that can detect man overboard situationthrough inputs from a camera. Awell performing warning system would allow boatmanufactures to comply with safety regulations and expand the kill-switch coverage to allpassengers on the boat. Furthermore, the aim is also to create two new datasets: one dedicatedto human detection and one with man overboard fall sequences. YOLOv5 achievedthe highest performance on a new human detection dataset, with an average precision of97%. A Mobilenet-SSD-v1 network based on weights from training on the PASCAL VOCdataset and additional training on the new man overboard dataset is used as the detectionmodel in final warning system. The man overboard warning system achieves an accuracyof 50% at best, with a precision of 58% and recall of 78%.

Autonomous vehicles rely on perceiving the environment using on-board sensor. These sensors have inherent limitations in terms of their effective range and risk occlusion due to their placement in the environment. These constraints limit the operational design domain of autonomous vehicles due to reliability and safety concerns. This report aims to show how an off-board sensor can be used as a complement to a vehicles on-board sensors. The goal of this sensor complement is to achieve an extension of the vehicle’s operational design domain and to relax constraints on the on-board sensors. Off-board sensors are less constrained in terms of sensor placement and allow for a more optimized location to perceive the environment. An autonomous vehicle is implemented and limitations in terms of sensing range and reliability is analyzed. An off-board camera based positioning system is also implemented and tested together with the autonomous vehicle in order to explore how an extension of the sensing range can be achieved. The extension of the operational design domain is tested by implementing a lane merge scenario which utilize both on and off-board sensor information. The lane merge scenario is also tested using different types of radio communication, namely 4G hotspot, 5G and wifi. The results of the lane merge scenario indicate that it is possible to achieve a range extension using an off-board sensor and thereby allow for a possible extension of the operational design domain of the autonomous vehicle. Although a range extension is possible, sending off-board sensor data over wireless radio raises questions on how applicable the solution is considering safety requirements in the automotive industry. / Autonoma fordon förlitar sig på att kunna uppfatta omgivningen med hjälp av sensorer ombord på fordonet. Dessa sensorer har begränsningar vid vilka avstånd de är tillförlitliga samt riskerar att bli ockluderade på grund av hur sensorn är placerad på fordonet. Dessa begränsningar försvårar fordonets användningsområde till följd av tillförlitlighet och säkerhetsaspekter. Denna rapport försöker visa hur en extern sensor kan användas för att komplettera sensorer ombord ett fordon. Målet med detta komplement är att åstadkomma en utökning av fordonets användningsområde samt minimera begränsningarna av fordonets förmåga att uppfatta omgivningen. Externa sensorer kan placeras med större frihet vilket möjliggör en mer optimal placering för att maximera förmågan att iaktta trafiken. Ett autonomt fordon implementeras och dess begränsningar i form av sensorkänslighet och pålitlighet analyseras. Ett externt kamera-baserat positioneringssystem är också utvecklat och testat tillsammans med det autonoma fordonet för att undersöka hur en utökning av användningsområdet kan genomföras. Utökningen av fordonets användningsområde testas genom att genomföra ett scenario där det autonoma fordonet ska dela körfält med en annan trafikant. I detta scenario används både sensorer ombord på det autonoma fordonet samt externa sensorer. Sensorinformationen delas genom olika typer av radiokommunikation, såsom, 4G hotspot, 5G och wifi för att se om nätverksfördröjningen har påverkan på resultaten. Resultatet tyder på att det är möjligt att uppfylla en utökning av det autonoma fordonets användningsområde genom att använda en extern sensor som utökar perceptionen av omgivningen. En utökning av användningsområdet är möjlig men väcker frågor om huruvida trådlös kommunikation kan uppfylla de krav och säkerhetsregulationer som finns inom bilindustrin.

In one of the facilities at the Stena Recycling plant in Halmstad, Sweden, about 300 tonnes of metallic waste is processed each day with the aim of sorting out all non-ferrous material. At the end of this process, non-ferrous materials are manually sorted out from the ferrous materials. This thesis investigates a computer vision based approach to identify and localize the non-ferrous materials and eventually automate the sorting.Images were captured of ferrous and non-ferrous materials. The images areprocessed and segmented to be used as annotation data for a deep convolutionalneural segmentation network. Network models have been trained on different kinds and amounts of data. The resulting models are evaluated and tested in ac-cordance with different evaluation metrics. Methods of creating advanced train-ing data by merging imaging information were tested. Experiments with using classifier prediction confidence to identify objects of unknown classes were per-formed. This thesis shows that it is possible to discern ferrous from non-ferrous mate-rial with a purely vision based system. The thesis also shows that it is possible to automatically create annotated training data. It becomes evident that it is possi-ble to create better training data, tailored for the task at hand, by merging image data. A segmentation network trained on more than two classes yields lowerprediction confidence for objects unknown to the classifier.Substituting manual sorting with a purely vision based system seems like aviable approach. Before a substitution is considered, the automatic system needsto be evaluated in comparison to the manual sorting.

Additive Manufacturing (AM) is a fast-growing technology in manufacturing industries. Applications of AM are spread across a wide range of fields. The aerospace industry is one of the industries that use AM because of its ability to produce light-weighted components and design freedom. Since the aerospace industry is conservative, quality control and quality assurance are essential. The quality of the welding is one of the factors that determine the quality of the AM components, hence, detecting faults in the welding is crucial. In this thesis, an automated system for detecting the faults in the welding process is presented. For this, three methods are proposed to find the anomalies in the process. The process videos that contain weld melt-pool behaviour are used in the methods. The three methods are 1) Autoencoder method, 2) Variational Autoencoder method, and 3) Image Classification method. Methods 1 and 2 are implemented using Convolutional-Long Short Term Memory (LSTM) networks to capture anomalies that occur over a span of time. For this, instead of a single image, a sequence of images is used as input to track abnormal behaviour by identifying the dependencies among the images. The method training to detect anomalies is unsupervised. Method 3 is implemented using Convolutional Neural Networks, and it takes a single image as input and predicts the process image as stable or unstable. The method learning is supervised. The results show that among the three models, the Variational Autoencoder model performed best in our case for detecting the anomalies. In addition, it is observed that in methods 1 and 2, the sequence length and frames retrieved per second from process videos has effect on model performance. Furthermore, it is observed that considering the time dependencies in our case is very beneficial as the difference between the anomalous and the non anomalous process is very small

An intelligent sensor system has the potential of providing its operator with relevant information, lowering the risk of human errors, and easing the operator's workload. One way of creating such a system is by using reinforcement learning, and this thesis studies how reinforcement learning can be applied to a simple sensor control task within a detailed 3D rendered environment. The studied agent controls a stationary camera (pan, tilt, zoom) and has the task of finding stationary targets in its surrounding environment. The agent is end-to-end, meaning that it only uses its sensory input, in this case images, to derive its actions. The aim was to study how an agent using a simple neural network performs on the given task and whether behavior cloning can be used to improve the agent's performance. The best-performing agents in this thesis developed a behavior of rotating until a target came into their view. Then they directed their camera to place the target at the image center. The performance of these agents was not perfect, their movement contained quite a bit of randomness and sometimes they failed their task. But even though the performance was not perfect, the results were positive since the developed behavior would be able to solve the task efficiently given that it is refined. This indicates that the problem is solvable using methods similar to ours. The best agent using behavior cloning performed on par with the best agent that did not use behavior cloning. Therefore, behavior cloning did not lead to improved performance.

This study examines how Artificial Intelligence (AI) and autonomous systems can change the conditions for a decision-making process using John Boyd's OODA loop. The study is conducted through a qualitative text research method. The purpose is to analyze the systems based on a Swedish context by using empirical material from Swedish Defence Research Agency (FOI) and ”Perspektivstudien 2016-2018”. John Boyd's OODA loop is used as a theoretical framework for the analysis where AI and autonomous systems will be studied based on each phase. The results indicate that AI and autonomous systems can provide opportunities to improve intelligence report capabilities by gathering larger amounts of information from the battlefield. The military effect will be increased through a better situation awareness, processing larger amounts of information that will improve the management for decision making, analyzing and offering more course of action at a faster pace. Although it indicates that these systems will not be used as a decisive decision-maker, they will however create good conditions for those who make the decisions and provide a central system support.