No description available.

No description available.

No description available.

No description available.

The importance of detecting anomalies in surveillance camera data cannot be overemphasized. With the increasing availability of surveillance cameras in public and private locations, the need for reliable and effective methods to detect anomalous behavior has become critical to public safety. Anomaly detection algorithms can help identify potential threats in real time, allowing for rapid intervention and prevention of criminal activity. The examples of anomalies that can be detected by analyzing surveillance camera data include suspicious loitering or lingering, unattended bags or packages, crowd gatherings or dispersals, trespassing or unauthorized access, vandalism or property damage, violence or aggressive behavior, abnormal traffic patterns, missing or abducted persons, unusual pedestrian behavior, environmental anomalies. Detecting these anomalies in surveillance camera data can enable law enforcement, security personnel, and other relevant authorities to respond quickly and effectively to potential threats, ultimately contributing to a safer environment for all.  Surveillance camera data contains a large amount of information that is difficult for humans to analyze in real time. In addition, the sheer volume of data generated by surveillance cameras makes manual analysis impractical. Therefore, the development of automated anomaly detection algorithms is crucial for effective and efficient surveillance. The goal of this master's thesis is to detect anomalies using video cameras with an embedded machine learning processor and video analytics, such as human behavior. For this purpose, the most appropriate machine learning techniques will be selected and after comparing the results of these techniques, the best anomaly detection technique for the given circumstances will be identified.  To gather the evidence needed to answer the research questions, I will use a combination of methods appropriate to the study design. The study will follow a mixed-methods approach, combining a systematic literature review (SLR) and a formal experiment.  In this study, we investigated the effectiveness of various machine learning algorithms in detecting anomalous human behavior in video surveillance data.

This project aims to classify different units on a wireless network with the use of their frequency response. This is in purpose to increase security when communicating over WiFi. We use a convolution neural network for finding symmetries in the frequency responses recorded from two different units. We used two pre-recorded sets of data which contained the same units but from two different locations. The project achieve an accuracy of 99.987%, with a 5 hidden layers CNN, when training and testing on one dataset. When training the neural network on one set and testing it on a second set, we achieve results below 54.12% for identifying the units. At the end we conclude that the amount of data needed, for achieving high enough accuracy, is to large for this method to be a practical solution for non-stationary units.

Classification of physical exercises is an important task in many applications, particularly within health services. Innowearable AB has developed a device called Inno-X that collects data using an accelerometer and sEMG sensors. To optimizeInno-X, a Machine Learning AI must be implemented for real-time exercise classification, balancing simplicity and flexibility for maximum market impact. This enhances efficiency and accuracy in analysis. This thesis investigates how raw data from Inno-X can be used to implement a pipeline and a machine-learning AI with the purpose of classifying physical exercises in real time. Starting from implementing a protocol for collecting data to a finished end-to-end pipeline and AI that can perform the classification, this thesis includes all the steps in between. Comparison of different machine learning algorithms and the execution of transitioning from a training environment to a real-time environment has led to the obtained result. The highest accuracy achieved in the training and real-time environment was 96.98% and 90.00%, respectively. This thesis concludes that the more complex machine-learning algorithms perform better in the training environment, and the less complex algorithms perform better in the real-time environment.

<p>Unmanned Aerial Systems (UAS) have been an essential tool for field scouting, nutrient applications, and farm management. However, assessing the aerial images captured by UAS is labor-intensive, and human assessment can be misleading, introducing bias. Deep learning based image segmentation has been proposed to assist in segmenting different areas of interest in the field, but it usually requires significant pixel-level annotated data. To address this, we propose a semi-supervised learning algorithm, AgSemSeg, to train a robust image segmentation</p> <p>model with less annotated data. Semi-supervised semantic segmentation aims to predict accurate pixel-level segmentation results via incorporating unlabeled images. Existing methods rely on computing the consistency loss on the output predictions between pseudo-labels and unlabeled images. In AgSemSeg, we exploit the intermediate feature representations rather than only using the output predictions to improve the overall performance of the</p> <p>model. Specifically, we add a projection layer on the output of the backbone encoder, and inject consistency loss between intermediate feature representations with Sliced-Wasserstein distance. We evaluate AgSemSeg using Agriculture-Vision dataset and outperform the supervised baseline by up to 9.71%. We also evaluate AgSemSeg on benchmark datasets such as PASCAL VOC 2012 and Cityscapes datasets, and it outperforms supervised baselines by up to 24.6% and 7.5% mIoU, respectively. We also perform extensive ablation studies to show that our proposed components are key to the performance improvements of our method. </p>

This master thesis investigates the possibility to automatically label and classify radio waves for presence detection, where the objective is to obtain information about the number of people in a room based on channel estimates. Labeling data for machine learning is time consuming and tedious process. To address this two approaches are evaluated. One was to develop a framework to generate labels with the aid of computer vision AI. The other relies on unsupervised learning classifiers complemented with heuristics to generate the labels. The investigation also studies the performance of the classifiers as a function of the TX/RX configuration, SNR, number of consecutive samples in a feature vector, bandwidth and frequency band. When someone moves in a room the propagation environment changes and induces variations in the channel estimates, compared to when the room is empty. These variations are the fundamental concept that is exploited in this thesis. Two methods are suggested to perform classification without the need of training data. The first uses random trees embeddings to construct a random forest without labels and the second using statistical bootstrapping with a random forest classifier. The labels used for annotation indicate whether were zero, one or two people in the room. The performance of binary and non-binary classification is evaluated both for the two blind detection models, as well as the performance of the unsupervised learning techniques Kmeans and self-organizing maps. For classification both supervised and unsupervised learning use random forest classifiers. Results show that random forest classifiers perform well for this kind of problem, and that random tree embeddings are able to extract relational data that could be used for automatic labeling of the data.

Background: Customer churn prediction is one of the most popular part of bigbusinesses and often help the companies in customer retention and revenue generation.Customer churn may lead to huge loss of revenue and is important to analyzeand determine the cause for churn. Moreover, it is easier to retain an existing customerrather than acquiring new clients.Therefore, to get a better understanding onchurn prediction, this research work focuses on finding the best performing machinelearning model after effective comparision among four machine learning models. Theresearch also gives a brief report of latest literature work done in churn analysis ofmusic streaming services. Objectives: In this thesis work, we aim to research about churn prediction done inmusic streaming services. We focus on two main objectives, first one includes literaturereview on the latest research work done in churn prediction of music streamingservices. Secondly, we aim in comparing the performance of four supervised machinelearning algorithms, to find out the best performing algorithm for churn prediction. Methods: This thesis involves two methods literature review and experimentationto answer our research questions. We chose to use literature review for RQ1 soit can give a better understanding on our selected problem and works as base workfor our research and helps in clear and better comprehension. Experimentation ischosen for RQ2 to to build and train the selected machine learning model to validatethe performance of algorithms. Experimentation is chosen because it gives betterresults and prediction compared to surveys and reviews. Results: We have selected four classification supervised machine learning algorithmsnamely, Logistic regression, Naive Bayes, KNN, and RF in this research.Upon experimentation and training the models using the algorithms with a preprocessingthe KKBox’s dataset, RF achieved highest accuracy of 97% compared toother models. Conclusions: We have trained four models using the four machine learning algorithmsfor the prediction of churn in music streaming service domain. Upon trainingthe models with the KKBox’s dataset and upon experimentation, we came to a conclusionthat RF has the best performance with better accuracy and AUC score.