Allt fler misshandelsbrott sker i Sverige enligt Brå. För att reducera detta kan det som fångats på övervakningskameror användas i brottsutredningar, för att senare användas som bevismaterial till att döma den eller de skyldiga till brottet. Genom att optimera övervakningen kan företag använda sig av automatiserad igenkänning. Automatisering för igenkänningen av normala kontra onormala beteenden går att lösa med djupinlärning. Syftet med denna undersökning är att finna en lämplig modell som kan identifiera det onormala beteendet (till exempel ett slagsmål). Modell arkitekturen som användes under projektet var 3D ResNet, eftersom den klara av en djupare arkitektur. Ett djupare nätverk, innebär bättre prediktion av problemet. 3DResNet-34 var den modell arkitekturen som gav högst noggrannhet med 93,33%. Implementering av projektet utfördes i ramverket PyTorch. Undersökningen har visat att med           hjälp av överförd inlärning går det att återanvända kunskap från förtränade modeller och applicera dessa kunskaper på det aktuella problemet. Detta bidrar till en mer pålitligare modell med noggrann prediktion på nytt övervaknings           material. / According to Brå, more assault crimes are taking place in Sweden. To reduce this, information that was captured on surveillance cameras can be used in criminal investigations, to convict the perpetrator or perpetrators of the crime. To optimize monitoring, companies can use automation. Automation of the recognition of normal versus abnormal activities can be solved with deep learning. The purpose of this study is to find a suitable model that can identify               the abnormal activity (for example, a fight). The model architecture used during the project was 3D ResNet, because it was capable of handling deeper architectures. Having a deeper network means better prediction of the problem.           3D ResNet-34 was the model architecture that gave the highest accuracy with 93,33%. Implementation of the project was carried out in the framework of PyTorch. The study has shown that with the help of transfer learning it is possible to transfer knowledge from pre-trained models and apply this          knowledge to the current problem. This contributes to a more reliable model with accurate prediction for new surveillance footage.

The main benefit of the work is the optimization of the hardware configuration for the calculation of neural networks. The theoretical part describes neural networks, deep learning frameworks and hardware options. The next part of the thesis deals with implementation of performance tests, which include application of Inception V3 and ResNet models. Network models are applied to various graphics cards and computing hardware. The output of the thesis is the implemented model of the network Inception V3, which examines the graphics cards and their performance, time-consuming calculations and their efficiency. The ResNet model is applied to a section that examines other impacts on neural network computing such as used disk, operating memory, and so on. Each practical part contains a discussion where the knowledge of the given part is explained. In the case of consumption measurement, a mismatch between the declaration by the manufacturer and the measured values was identified.

With the development of artificial neural network (ANN), it has been introduced in more and more computer vision tasks. Convolutional neural networks (CNNs) are widely used in object detection, object tracking, and semantic segmentation, achieving great performance improvement than traditional algorithms. As a classical topic in computer vision, the exploration of applying deep CNNs for depth recovery from monocular images is popular, since the single-view image is more common than stereo image pair and video. However, due to the lack of motion and geometry information, monocular depth estimation is much more difficult. This thesis aims at investigating depth prediction from single images by exploiting state-of-the-art deep CNN models. Two neural networks are studied: the first network uses the idea of a global and local network, and the other one adopts a deeper fully convolutional network by using a pre-trained backbone CNN (ResNet or DenseNet). We compare the performance of the two networks and the result shows that the deeper convolutional neural network with the pre-trained backbone can achieve better performance. The pre-trained model can significantly accelerate the training process. We also find that the amount of training dataset is essential for CNN-based monocular depth prediction. / Utvecklingen av artificiella neurala nätverk (ANN) har gjort att det nu använts i flertal datorseende tekniker för att förbättra prestandan. Convolutional Neural Networks (CNN) används ofta inom objektdetektering, objektspårning och semantisk segmentering, och har en bättre prestanda än de föregående algoritmerna. Användningen av CNNs för djup prediktering för single-image har blivit populärt, på grund av att single-image är vanligare än stereo-image och filmer. På grund av avsaknaden av rörelse och geometrisk information, är det mycket svårare att veta djupet i en bild än för en film. Syftet med masteruppsatsen är att implementera en ny algoritm för djup prediktering, specifikt för bilder genom att använda CNN modeller. Två olika neurala nätverk analyserades; det första använder sig av lokalt och globalt nätverk och det andra består av ett avancerat Convolutional Neural Network som använder en pretrained backbone CNN (ResNet eller DenseNet). Våra analyser visar att avancerat Convolutional Neural Network som använder en pre-trained backbone CNN har en bättre prestanda som påskyndade inlärningsprocessen avsevärt. Vi kom även fram till att mängden data för inlärningsprocessen var avgörande för CNN-baserad monokulär djup prediktering.

Ring artifacts are a common problem with the use of photon-counting detectors and commercial deployment rests on being able to compensate for them. Deep learning has been proposed as a candidate for tackling the inefficiency or high cost of traditional techniques. In that spirit, we propose a new approach to ring artifact reduction, namely one that employs Residual Networks in sinogram domain. We train them on data simulated via a realistic photon-counting CT model based on numerical phantoms of real scans acquired by the KiTS19 Challenge dataset. By exploring various architectures we find that shallow ResNets achieve a significant artifact reduction by staying more true to the ground truth in terms of not introducing new artifacts. All networks introduce a smoothing effect which is attributed to the use of MSE as a loss function. An alternative training scheme using patches instead of whole sinograms is tested and it shows a slightly improved model stability. Lastly, we demonstrate via a performance metric study that common metrics are not suitable for quantifying the performance in this problem, save for a potential new approach in the virtual mono-energetic domain. / Ringartefakter är ett vanligt problem vid användning av fotonräknande detektorer och kommersiell introduktion kräver att man kan kompensera för dem. Djupinlärning har föreslagits som en kandidat för att hantera ineffektiviteten eller de höga kostnaderna för traditionella tekniker. I den andan föreslår vi ett nytt tillvägagångssätt för att reducera ringartefakter, nämligen en som använder sig av residualnätverk i sinogramdomänen. Vi tränar dem på data simulerad via en realistisk fotonräkning CT modell baserad på numeriska fantomer av verkliga skanningar från datamängen KiTS19 Challenge. Genom att utforska olika arkitekturer finner vi att grunda ResNet uppnår en betydande minskning av artefakter genom bevara en större likhet med den sanna bilden när det gäller att inte introducera nya artefakter. Alla nätverk introducerar en utsmetningseffekt som tillskrivs användningen av MSE som en förlustfunktion. Ett alternativt träningsschema med utsnitt istället för hela sinogram testas och det visar en något förbättrad modellstabilitet. Slutligen visar vi genom en prestandamåttstudie att vanliga prestandamått inte är lämpliga för att kvantifiera prestandan i detta problem med undantag för ett potentiellt nytt tillvägagångssätt i den virtuella monoenergetiska domänen.

No description available.

The aim of the diploma thesis is to make research of the self-supervised learning in computer vision applications, then to choose a suitable test task with an extensive data set, apply self-supervised methods and evaluate. The theoretical part of the work is focused on the description of methods in computer vision, a detailed description of neural and convolution networks and an extensive explanation and division of self-supervised methods. Conclusion of the theoretical part is devoted to practical applications of the Self-supervised methods in practice. The practical part of the diploma thesis deals with the description of the creation of code for working with datasets and the application of the SSL methods Rotation, SimCLR, MoCo and BYOL in the role of classification and semantic segmentation. Each application of the method is explained in detail and evaluated for various parameters on the large STL10 dataset. Subsequently, the success of the methods is evaluated for different datasets and the limiting conditions in the classification task are named. The practical part concludes with the application of SSL methods for pre-training the encoder in the application of semantic segmentation with the Cityscapes dataset.

The main objective of this thesis is to propose a localization method for a UAV using various computer vision and machine learning techniques. It plays a major role in planning the strategy for the flight, and acts as a navigational contingency method, in event of a GPS failure. The implementation of the algorithms employs high processing capabilities of the graphics processing unit, making it more efficient. The method involves the working of various neural networks, working in synergy to perform the localization. This thesis is a part of a collaborative project between The University of North Texas, Denton, USA, and the University of Windsor, Ontario, Canada. The localization has been divided into three phases namely object detection, recognition, and location estimation. Object detection and position estimation were discussed in this thesis while giving a brief understanding of the recognition. Further, future strategies to aid the UAV to complete the mission, in case of an eventuality, like the introduction of an EDGE server and wireless charging methods, was also given a brief introduction.

A brain tumor is detected and classified by biopsy that is conducted after the brain surgery. Advancement in technology and machine learning techniques could help radiologists in the diagnosis of tumors without any invasive measures. We utilized a deep learning-based approach to detect and classify the tumor into Meningioma, Glioma, Pituitary tumors. We used registration and segmentation-based skull stripping mechanism to remove the skull from the MRI images and the grab cut method to verify whether the skull stripped MRI masks retained the features of the tumor for accurate classification. In this research, we proposed a transfer learning based approach in conjunction with discriminative learning rates to perform the classification of brain tumors. The data set used is a 3064 T MRI images dataset that contains T1 flair MRI images. We achieved a classification accuracy of 98.83%, 96.26%, and 95.18% for training, validation, and test sets and an F1 score of 0.96 on the T1 Flair MRI dataset.

The hard equation of railway safety versus the high commercial profits can only be achieved through the use of new inspection methods supported by modern technologies. The track and its components can have different types of troubles, such as rail surface defects, broken sleepers, missing fasteners, and irregular ballast levels. Each component of the track infrastructure plays a significant role, where the failure or the absence of any of them can pave the way to undesired situations. The rail is designed to carry and direct the train, the sleepers are meant to maintain the level of the rail, and the ballast mission is to keep all components floating on the surface of the ground. The fasteners are used to fasten the rail to the sleepers, and therefore too many missing fasteners can lead to sever unsteady tracks, which can, in turn, result in derailment. Therefore, there is a high demand for advanced inspection methods to monitor the railway track and its components continuously. The presence of such advanced inspection models would help the railway industry avoid obstacles such as high operation and maintenance costs, dangerous accidents, and uncomfortable passenger's experience.   This master thesis aims to present an efficient method to classify the track and its components by combining image processing techniques and deep learning algorithms. This method was able to detect the missing fasteners in the set of images captured by a line camera, continuously monitoring the rail and its associated fasteners. The experimental results obtained in this thesis showed that the proposed method is efficient and robust for detecting the track and its components in complex environments. The thesis also discusses the idea of building one complete model that can process and classify all track components at once. The image processing technique was employed to extract different components of the track, individually: fasteners, rail, ballast, and sleepers. The model was trained and used to classify the state of the fasteners. Classification of other components of the track will be a part of the future work.

Yes / In recent years, automatic facial age estimation has gained popularity due to its numerous applications. Much work has been done on frontal images and lately, minimal estimation errors have been achieved on most of the benchmark databases. However, in reality, images obtained in unconstrained environments are not always frontal. For instance, when conducting a demographic study or crowd analysis, one may get profile images of the face. To the best of our knowledge, no attempt has been made to estimate ages from the side-view of face images. Here we exploit this by using a pre-trained deep residual neural network (ResNet) to extract features. We then utilize a sparse partial least squares regression approach to estimate ages. Despite having less information as compared to frontal images, our results show that the extracted deep features achieve a promising performance.