In recent years, deep learning based computer vision technology has developed rapidly. This is not only due to the improvement of computing power, but also due to the emergence of high-quality datasets. The combination of object detectors and drones has great potential in the field of rescue and disaster relief. We created an image dataset specifically for vision applications on drone platforms. The dataset contains 5000 images, and each image is carefully labeled according to the PASCAL VOC standard. This specific dataset will be very important for developing deep learning algorithms for drone applications. In object detection models, loss function plays a vital role. Considering the uneven distribution of large and small objects in the dataset, we propose adjustment coefficients based on the frequencies of objects of different sizes to adjust the loss function, and finally improve the accuracy of the model.

It is known that quantum strategies change the range of possible payoffs for the players in the prisoner's dilemma. In this paper, we examine the effect of the degree of entanglement in determining the payoffs. When both players play quantum strategies, we show that the payoff for both players is unaffected by the entanglement value and it leads to a new Nash equilibrium.

Coarse-Grained Reconfigurable Architectures (CGRAs) are promising in developing high performance low-power portable applications. In this research, we crowdsource a mapping problem using gamification to harnass human intelligence. A scientific puzzle game, Untangled, was developed to solve a mapping problem by encapsulating architectural characteristics. The primary motive of this research is to draw insights from the mapping solutions of players who possess innate abilities like decision-making, creative problem-solving, recognizing patterns, and learning from experience. In this dissertation, an extensive analysis was conducted to investigate how players' computational skills help to solve an open-ended problem with different constraints. From this analysis, we discovered a few common strategies among players, and subsequently, a library of dictionaries containing identified patterns from players' solutions was developed. The findings help to propose a better version of the game that incorporates these techniques recognized from the experience of players. In the future, an updated version of the game that can be developed may help low-performance players to provide better solutions for a mapping problem. Eventually, these solutions may help to develop efficient mapping algorithms, In addition, this research can be an exemplar for future researchers who want to crowdsource such electrical engineering problems and this approach can …

The purpose of this research is not to research new technology but how to improve existing technology and understand how the manufacturing process works. Reliability Engineering fall under the category of Quality Control and uses predictions through statistical measurements and life testing to figure out if a specific manufacturing technique will meet customer satisfaction. The research also answers choice of materials and choice of manufacturing process to provide a device that will not only meet but exceed customer demand. Reliability Engineering is one of the final testing phases of any new product development or redesign.

This thesis proposes a design of a rapidly deployable cellular network prototype that provides voice and data communications and it is interoperable with legacy devices and the existing network infrastructure. The prototype is based on software defined radio and makes use of IEEE 802.11 unlicensed wireless radio frequency (RF) band for backhaul link and an open source GSM implementation software. The prototype is also evaluated in environments where there is limited control of the radio frequency landscape, and using Voice Over Internet Protocol (VoIP) performance metrics to measure the quality of service. It is observed that in environments where the IEEE 802.11 band is not heavily utilized, a large number of calls are supported with good quality of service. However, when this band is heavily utilized only a few calls can be supported as the quality of service rapidly degrades with increasing number of calls, which is due to interference. It is concluded that in order to achieve tolerable voice quality, unused licensed spectrum is needed for backhaul communication between base stations.

This thesis targets at providing an algebraic method to indicate network behaviors. Furthermore, for a signed-average consensus problem of the system behaviors, event-triggering signed-average algorithms are designed to reduce the communication overheads. In Chapter 1, the background is introduced, and the problem is formulated. In Chapter 2, notations and basics of graph theory are presented. It is known that the terminal value of the system state is determined by the initial state, left eigenvector and right eigenvector associated with zero eigenvalue of the Laplacian matrix. Since there is no mathematical expression of right eigenvector, in Chapter 3, mathematical expression of right eigenvector is given. In Chapter 4, algebraic trait for structurally balanced property of a node is proposed. In Chapter 5, a method for characterization of collective behaviors under directed signed networks is developed. In Chapter 6, dynamic event-triggering signed-average algorithms are proposed and proved for the purpose of relieving the communication burden between agents. Chapter 7 summarizes the thesis and gives future directions.

By organically fusing the methods of convolutional neural network (CNN) and recurrent neural network (RNN), this dissertation focuses on the application of optical character recognition and image classification processing. The first part of this dissertation presents an end-to-end novel receipt recognition system for capturing effective information from receipts (CEIR). The main contributions of this research part are divided into three parts. First, this research develops a preprocessing method for receipt images. Second, the modified connectionist text proposal network is introduced to execute text detection. Third, the CEIR combines the convolutional recurrent neural network with the connectionist temporal classification with maximum entropy regularization as a loss function to update the weights in networks and extract the characters from receipt. The CEIR system is validated with the scanned receipts optical character recognition and information extraction (SROIE) database. Furthermore, the CEIR system has strong robustness and can be extended to a variety of different scenarios beyond receipts. For the convolutional recurrent structure application of land use image classification, this dissertation comes up with a novel deep learning model for land use classification, the convolutional recurrent land use classifier (CRLUC), which further improves the accuracy in classifying remote sensing land use images. Besides, the …

In game based learning, motivating the players to learn by providing them a desirable gameplay experience is extremely important. However, it's not an easy task considering the quality of today's commercial non-educational games. Throughout the gameplay, the player should neither get overwhelmed nor under-challenged. The best way to do so is to monitor the player's actions in the game because these actions can tell the reason behind the player's performance. They can also tell about the player's lacking competencies or knowledge. Based on this information, in-game educational interventions in the form of hints can be provided to the player. The success of such games depends on their interactivity, motivational outlook and thus player retention. UNTANGLED is an online mapping game based on crowd-sourcing, developed by Reconfigurable Computing Lab, UNT for the mapping problem of CGRAs. It is also an educational game for teaching the concepts of reconfigurable computing. This thesis performs qualitative comparative analysis on gameplays of low performing players of UNTANGLED. And the implications of this analysis are used to provide recommendations for improving the gameplay experience for these players by guiding them. The recommendations include strategies to reach a high score and a compact solution, hints in the …

Water scarcity for agriculture is one of the most important challenges to improve food security worldwide. In this thesis we study the potential to develop a low-cost controller for a small scale brackish desalination plant that consists of proven water treatment technologies, reverse osmosis, cation exchange, and nanofiltration to treat groundwater into two final products: drinking water and irrigation water. The plant is powered by a combination of wind and solar power systems. The low-cost controller uses Arduino Mega, and Arduino DUE, which consist of ATmega2560 and Atmel SAM3X8E ARM Cortex-M3 CPU microcontrollers. These are widely used systems characterized for good performance and low cost. However, Arduino also requires drivers and interfaces to allow the control and monitoring of sensors and actuators. The thesis explains the process, as well as the hardware and software implemented.

Deep reinforcement learning is an area of research that has blossomed tremendously in recent years and has shown remarkable potential in computer games. Real-time strategy game has become an important field of artificial intelligence in game for several years. This paper is about to introduce a kind of algorithm that used to train agents to fight against computer bots. Not only because games are excellent tools to test deep reinforcement learning algorithms for their valuable insight into how well an algorithm can perform in isolated environments without the real-life consequences, but also real-time strategy games are a very complex genre that challenges artificial intelligence agents in both short-term or long-term planning. In this paper, we introduce some history of deep learning and reinforcement learning. Then we combine them with StarCraft. PPO is the algorithm which have some of the benefits of trust region policy optimization (TRPO), but it is much simpler to implement, more general for environment, and have better sample complexity. The StarCraft environment: Blood War Application Programming Interface (BWAPI) is open source to test. The results show that PPO can work well in BWAPI and train units to defeat the opponents. The algorithm presented in the thesis is …

Cognitive Radio (CR) technology creates the opportunity for unlicensed users to make use of the spectral band provided it does not interfere with any licensed user. It is a prominent tool with spectrum sensing functionality to identify idle channels and let the unlicensed users avail them. Thus, the CR technology provides the consumers access to a very large spectrum, quality spectral utilization, and energy efficiency due to spectral load balancing. However, the full potential of the CR technology can be realized only with CRs equipped with accurate mechanisms to predict/sense the spectral holes and vacant spectral bands without any prior knowledge about the characteristics of traffic in a real-time environment. Multi-layered perception (MLP), the popular neural network trained with the back-propagation (BP) learning algorithm, is a keen tool for classification of the spectral bands into "busy" or "idle" states without any a priori knowledge about the user system features. In this dissertation, we proposed the use of an evolutionary algorithm, Bacterial Foraging Optimization Algorithm (BFOA), for the training of the MLP NN. We have compared the performance of the proposed system with the traditional algorithm and with the Hybrid GA-PSO method. With the results of a simulation experiment that this …

The goal of the proposed device and software architecture is to apply the functionality of mixed reality (MR) in order to make a virtual environment that is more accessible to the visually-impaired. We propose a glove-based system for MR that will use finger and hand movement tracking along with tactile feedback so that the visually-impaired can interact with and obtain a more detailed sense of virtual objects and potentially even virtual environments. The software architecture makes current MR frameworks more accessible by augmenting the existing software and extensive 3D model libraries with both the interfacing of the glove-based system and the audibly navigable user interface (UI) of a virtual environment we have developed. We implemented a circuit with finger flexion/extension tracking for all 5 fingers of a single hand and variable vibration intensities for the vibromotors on all 5 fingertips of a single hand. The virtual environment can be hosted on a Windows 10 application. The virtual hand and its fingers can be moved with the system's input and the virtual fingertips touching the virtual objects trigger vibration motors (vibromotors) to vibrate while the virtual objects are being touched. A rudimentary implementation of picking up and moving virtual objects inside …