University of Northern British Columbia Institutional Repository

Search results

Search Results

Displaying (1 - 20 of 88)

Pages

Title: A neural network model of the primary visual cortex.
Contributors: Alan Spara (author), Charles Brown (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: Many problems in modern computing require a visual component. That is to say, it is fairly common for applications to have a need to see their environments. These applications will typically employ techniques designed specifically to solve the particular task needed for the application, and have little or no relation to the human visual system. Humans generally do not have difficulty interpreting the world around us. When traveling through known environments, we can easily recognize particular walls, doors and other objects in our view. We are not confused by the huge number factors that can complicate an image. The generalization and robustness of the human system would provide a huge benefit to any system that requires more advanced vision than is capable with the ad-hoc methods developed previously. If the underlying principles that make the human visual system so powerful can be identified and implemented programmatically, then a machine could reap the benefits obtained by humans. The purpose of this thesis is to demonstrate that a visual system modeled after the human visual system will be robust and accurate enough to solve real world problems - and to be useful in a non-trivial application. By developing neural networks that directly model the most primitive image processing cells of the human visual system, a platform can be built on which advanced vision systems can be developed.
Discipline: Computer Science
Date added: 2017-03-30T17:06:00.019Z

Title: Inverse scale invariant feature transform models for object recognition and image tagging.
Contributors: Md. Kamrul Hasan (author), Liang Chen (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: This thesis presents three novel image models based on Scale Invariant Feature Transform (SIFT) features and the k-Nearest Neighbors (k-NN) machine learning methodology. While SIFT features characterize an image with distinctive keypoints, the k-NN filters away and normalizes the keypoints with a two-fold goal: (i) compressing the image size, and (ii) reducing the bias that is induced by the variance of keypoint numbers among object classes. Object recognition is approached as a supervised machine learning problem, and the models have been formulated using Support Vector Machines (SVMs). These object recognition models have been tested for single and multiple object detection, and for asymmetrical rotational recognition. Finally, a hierarchical probabilistic framework with basic object classification methodology is formulated as a multi-class learning framework. This framework has been tested for automatic image annotation generation. Object recognition models were evaluated using recognition rate (rank 1) whereas the annotation task was evaluated using the well-known Information Retrieval measures: precision, recall, average precision and average recall.
Discipline: Computer Science
Date added: 2017-03-30T17:06:52.164Z

Title: Face recognition using convolutional macropixel comparison approach
Contributors: Yunke Li (author), Liang Chen (thesis advisor), University of Northern British Columbia (Degree granting institution), Jernej Polajnar (committee member), Jianbing Li (committee member)
Abstract: Convolutional Neural Network (CNN) is a widely used deep learning framework and is applied in the field of face recognition achieving outstanding results. Macropixel Comparison Approach is a shallow mathematical approach that recognizes face by comparing original pixel blocks of face images. In this thesis, we are inspired by ideas of the currently popular deep neural network framework and introduce two features into the mathematical approach: deep overlap and weighted filter. The aim is exploring if the idea of deep learning could benefit mathematical method which might extends the scope of face recognition research. Results from our experiments show that the new proposed approach achives markedly better recognition rates than the original macropixel method.
Discipline: Computer Science
Date added: 2019-03-31T20:16:03.958Z

Title: A software framework for simulation studies of interaction models in agent teamwork.
Contributors: Omid Alemi (author), Jernej Polajnar (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: This thesis proposes a new software framework that facilitates the study of agent interaction models in early development stages from a designer's perspective. Its purpose is to help reduced the design decision space through simulation experiments that provide early feedback on comparative performance of alternative solutions. This is achieved through interactive concurrent simulation of multiple teams in a representative microworld context. The generic simulator's architecture accommodates an open class of different microworlds and permits multiple communication mechanisms. It also supports interoperability with other software tools, distributed simulation, and various extensions. The framework was validated in the context of two different research projects on helpful behavior in agent teams: the Mutual Assistance Protocol, based on rational criteria for help, and the Empathic Help Model, based on a concept of empathy for artificial agents. The results show that the framework meets its design objectives and provides the flexibility needed for research experimentation. --Leaf i.
Discipline: Computer Science
Date added: 2017-03-30T17:12:21.252Z

Title: Experimental study of multi-level regional voting scheme and its application in human face recognition.
Contributors: Jian Zhang (author), Liang Chen (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: Based on previous regional voting scheme model studies and assumptions, we extend studies from two-level voting scheme models to multi-level voting scheme models. Due to mathematical complexity, we use the Monte Carlo approach in studying the stability of the multi-level regional voting scheme with respect to region sizes and levels. Using this model, we are able to obtain the stability characteristics of the national voting scheme, the two-level regional voting scheme, and the multi-level regional voting scheme, and apply it in FERET human face database. According to our study, we verify again that the regional voting scheme (including the two-level regional voting scheme and multi-level regional voting scheme) is always more stable than the national voting scheme. We find that the stability of the multi-level regional voting scheme is not as good as the stability of the two-level regional voting scheme when region size is within a certain range. Out of this range, the multi-level regional voting scheme may compete with the two-level voting scheme. We conclude that the multi-level regional voting scheme may be comparable to the two-level regional voting scheme and prove our conclusion in the face recognition application.
Discipline: Computer Science
Date added: 2017-03-30T17:03:44.07Z

Title: A security architecture for IPv6 enabled wireless medical sensor networks.
Contributors: M. Abdul Alim (author), Liang Chen (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: We present the design of an IPv6 enabled wireless sensor network based on the IEEE 802.15.4 standard for medical monitoring. We design a routing mechanism for efficient flooding, a hop-by-hop error recovery and congestion control mechanism for reliable packet delivery and a lightweight security architecture for the medical monitoring system. We extend the widely used Extensible Authentication Protocol (EAP) to employ the Generalized Pre-shared Key (GPSK) authentication method with some optimizations for securing the system. We use the 3-party EAP model with the Personal Area Network Coordinator (PAN coordinator) of IEEE 802.15.4 standard as the EAP authenticator for authenticating sensor nodes within the radio range of the PAN coordinator. In order to use EAP authentication for a sensor node several hops away from the PAN coordinator, we define a new role (relay authenticator) for its coordinator which tunnels EAP messages to the PAN coordinator securely. We define EAP message encapsulation for IEEE 802.15.4 networks and a key hierarchy for the security architecture. We have simulated the system and shown that EAP based authentication is feasible in wireless sensor networks.
Discipline: Computer Science
Date added: 2017-03-30T17:06:02.1Z

Title: Parallelism in multiple sequence alignment.
Contributors: Qiong Bai (author), Siamak Rezaei (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: No abstract available.
Discipline: Computer Science
Date added: 2017-03-30T17:01:07.266Z

Title: Connectionless routing protocols for mobile ad-hoc networks.
Contributors: Yong Sun (author), Kuppuchamy Alagarsamy (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: No abstract available.
Discipline: Computer Science
Date added: 2017-03-30T17:00:40.608Z

Title: Efficient enumeration of small graphlets and orbits
Contributors: Apratim Das (author), Alex Aravind (thesis advisor), Fan Jiang (committee member), Samuel Walters (committee member), Mark Dale (committee member), University of Northern British Columbia (Degree granting institution)
Abstract: As the world is flooded with data, the demand for mining data for useful purposes is increasing. An effective techniques is to model the data as networks (graphs) and then apply graph mining techniques for analysis. As on date, the algorithms available to count graphlets and orbits for various types of graphs and their generalizations are limited. The thesis aims to fill the gap by presenting a simple and efficient algorithm for 3-node graphlet and orbit counting that is generic enough to work for both undirected and directed graphs. Our algorithm is compared with the state-of-art algorithms and we show that in most cases our algorithm performs better. We demonstrate our algorithm in three case studies related to (i) enzyme and metabolite correlation network in corn, (ii) watershed governance networks, and (iii) patterns exhibited by co-expression networks of healthy and cancerous stomach cells.
Discipline: Computer Science
Date added: 2020-05-28T21:03:27.802Z

Title: Computing ethics: the case for codes of ethics and privacy policies
Contributors: Laura Pauline Nyanchama Kombo (author), Todd Whitcombe (Thesis advisor), University of Northern British Columbia College of Science and Management (Degree granting institution), Chris Opio (Committee member), Iliya Bluskov (Committee member)
Abstract: Ethics and privacy are integral to life although limited research has been conducted relative to global codes of ethics and privacy policies of corporations. This piqued the interest on this research where the first contribution examines codes of ethics worldwide. It compares codes of different societies to IEEE and proposes changes which address issues of diversity, culture, and sociopolitical differences. Four countries have adopted the IEEE codes of ethics, while 28 countries have some variations. A global code of ethics would be useful in a world without borders. The second contribution introduces new guidelines for Canadian corporations regarding privacy policies. It examines the compatibility and compliance of corporate privacy policies with PIPEDA. An examination of the corporations revealed only 1,017 have public-facing privacy policies on their websites and some do not seem to satisfy all PIPEDA principles. New guidelines will help to ensure a better compliance with PIPEDA by corporations.
Discipline: Computer Science
Date added: 2017-05-15T20:51:03.798Z

Title: A software system for agent-assisted ontology building
Contributors: Denish Mumbaiwala (author), Jernej Polajnar (Thesis advisor), Desanka Polajnar (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: This thesis investigates how one can design a team of intelligent software agents that helps its human partner develop a formal ontology from a relational database and enhance it with higher-level abstractions. The resulting efficiency of ontology development could facilitate the building of intelligent decision support systems that allow: high-level semantic queries on legacy relational databases autonomous implementation within a host organization and incremental deployment without affecting the underlying database or its conventional use. We introduce a set of design principles, formulate the prototype system requirements and architecture, elaborate agent roles and interactions, develop suitable design techniques, and test the approach through practical implementation of selected features. We endow each agent with model meta-ontology, which enables it to reason and communicate about ontology, and planning meta-ontology, which captures the role-specific know-how of the ontology building method. We also assess the maturity of development tools for a larger-scale implementation. --Leaf i.
Discipline: Computer Science
Date added: 2017-03-29T17:30:31.878Z

Title: Design and simulation of an adaptive concurrency control protocol for distributed real-time database systems.
Contributors: Paul R. Stokes (author), Waqar Haque (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: No abstract available.
Discipline: Computer Science
Date added: 2017-03-30T17:03:30.16Z

Title: Entropy of printed Bengali language texts.
Contributors: Subrata Pramanik (author), Saif Zahir (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: One of the most important sources of information is written and spoken human language. The language that is spoken, written, or signed by humans for general-purpose communication is referred to as natural language. Determining the entropy of natural language text is a fundamentally important problem in natural language processing. The study and analysis of the entropy of a language can be a meaningful resource for researchers in linguistics and communication theory. For the purpose of this research we have taken printed Bengali language text as our source of natural language. We have collected a sufficient number of printed Bengali language text samples and divided them into two classes, newspaper and literature. We have studied each class in order to come up with specific entropy for each category and analyzed their characteristics. As a separate study, we collected printed religious Bengali language texts, divided them into two classes, Islamic and Hindu, found their entropy and studied and analyzed their characteristics. From our research, we have found the zero and first-order entropy of Bengali language to be 5.52 and 4.55 respectively. The language uncertainty and redundancy are 0.8242 and 17.58% respectively. These entropy and redundancy results of the language will be useful to researchers to help find a better text compression method for Bengali language.
Discipline: Computer Science
Date added: 2017-03-30T17:05:33.062Z

Title: Lumping of atmospheric organic chemical species by machine learning.
Contributors: Pruthvi Polam (author), Margot Mandy (Thesis advisor), Charles Brown (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: No abstract available.
Discipline: Chemistry and Computer Science
Date added: 2017-03-30T17:01:32.938Z

Title: New watermarking methods for digital images.
Contributors: Md. Wahedul Islam (author), Saif Zahir (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: The phenomenal spread of the Internet places an enormous demand on content-ownership-validation. In this thesis, four new image-watermarking methods are presented. One method is based on discrete-wavelet-transformation (DWT) only while the rest are based on DWT and singular-value-decomposition (SVD) ensemble. The main target for this thesis is to reach a new blind-watermarking-method. Method IV presents such watermark using QR-codes. The use of QR-codes in watermarking is novel. The choice of such application is based on the fact that QR-Codes have errors self-correction-capability of 5% or higher which satisfies the nature of digital-image-processing. Results show that the proposed-methods introduced minimal distortion to the watermarked images as compared to other methods and are robust against JPEG, resizing and other attacks. Moreover, watermarking-method-II provides a solution to the detection of false watermark in the literature. Finally, method IV presents a new QR-code guided watermarking-approach that can be used as a steganography as well. --Leaf ii.
Discipline: Computer Science
Date added: 2017-03-30T17:12:33.168Z

Title: A basis for pronominal anaphora resolution using a model of working memory and long-term memory
Contributors: Clifford James Thompson (author), Charles Brown (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: This thesis presents a new theory of information modelling in natural language processing that attempts to resolve anaphoric references, while also addressing the problem of knowledge complexity. A modular model of semantic representation is introduced that addresses the deficiencies of existing representations, as well as the drawbacks associated with expanding these semantic representations. Rather than using a single semantic representation to model human knowledge and the knowledge within a sentence, the theory proposes a modular, multi-level model which is based around a semantic network. The behaviour of the model uses theories on the nature of working and long-term memory from cognitive psychology. Two methods of artificial neuron activation and decay were implemented - the ACT-R model and the Thompson model. Maximum success rates of 54.10% and 83.61% were achieved for The Three Brothers corpus, and maximum success rates of 56.00% and 86.67% were achieved for the Rumpelstiltskin corpus.
Discipline: Computer Science
Date added: 2017-04-10T22:16:32.018Z

Title: Improving Sonar Sensor Fidelity in a Robot Simulator.
Contributors: Allan Edward Kranz (author), Charles Brown (Thesis advisor), Liang Chen (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: It is slow and expensive to develop robot control systems using real robots. Simulation can provide the benefits of lowering the time and cost. In order to take advantage of the benefits of development in a simulator we need high fidelity representations of actual sensors. Sensors do not provide perfect data and simulations that use either perfect models or models that are too simple will not translate well into the real world. This research introduces a sensor model that overcomes some of the existing limitations in current simulations and provides a methodology for developing both new models and corresponding testing regimes. An actual sensor is used in realistic situations to create authentic models that more closely match the performance of the robot in the real world. A simple sonar sensor is tested against three generic obstacles and a realistic software simulation model of its capabilities is created. The Simbad robot simulator is modified to use this model, a testing regime is created to validate the results, and improved performance over the existing model is achieved. --Leaf iii.
Discipline: Computer Science
Date added: 2017-04-11T21:14:05.531Z

Title: Face recognition using Dual Linear Regression based Classification
Contributors: Yuan Wang (author), Liang Chen (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: This project discusses a face recognition algorithm that uses Dual Linear Regression based Classification (DLRC) and a voting approach. Each face image used is first converted into a cluster of images each image in the cluster is obtained by shifting the original image a few pixels. The similarity of a pair of face images can be measured by comparing the distance between the corresponding image clusters, which is calculated using DLRC approach. To further improve performance, each cluster of images, representing a single face image, is then partitioned into a union of clusters of sub images. DLRC is then used to measure similarities between corresponding sub-image clusters to provide temporary identity decisions a voting approach is applied to make final conclusions. We have carried out experiments on a benchmark dataset for face recognition. The result demonstrates that the proposed approach works best in certain simple situations, while its performance is also comparable to known algorithms in complicated situations. --Leaf ii.
Discipline: Computer Science
Date added: 2017-03-29T17:30:05.64Z

Title: Prediction and preemptive control of network congestion in distributed real-time environment
Contributors: Ramandeep Dhanoa (author), Haque Waqar (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: Due to ever increasing demand for network capacity, the congestion problem is inflating. Congestion results in queuing within the network, packet loss and increased delays. It should be controlled to increase the system throughput and quality of service. The existing congestion control approaches such as source throttling and re-routing focus on controlling congestion after it has already happened. However, it is much more desirable to predict future congestion based on the current state and historical data, so that efficient controlling techniques can be applied to prevent congestion from happening in future. We have proposed a Neural Network Prediction-based routing (NNPR) protocol to predict as well as control the network traffic in distributed real time environment. A distributed real time transaction processing simulator (DRTTPS) has been used as the test-bed. For predictions, multi-step neural network model is developed in SPSS Modeler, which predicts congestion in future. ADAPA (Adaptive Decision and Predictive Analytics) scoring engine has been used for real-time scoring. An ADAPA wrapper calls the prediction model through web services and predicts the congestion in real-time. Once predicted results are obtained, messages are re-routed to prevent congestion. To compare our proposed work with existing techniques, two routing protocols are also implements "" Dijkstra's Shortest Path (DSP) and Routing Information Protocol (RIP). The main metric used to analyze the performance of our protocol is the percentage of transactions which complete before their deadline. The NNPR protocol is analyzed with various simulation runs having parameters both inside and outside the neural network input training range. Various parameters which can cause congestion were studied. These include bandwidth, worksize, latency, max active transactions, mean arrival time and update percentage. Through experimentation, it is observed that NNPR consistently outperforms DSP and RIP for all congestion loads. --Leaves [i]-ii.
Discipline: Computer Science
Date added: 2017-03-29T17:30:41.883Z

Title: Local binary pattern network: a deep learning approach for face recognition
Contributors: Meng Xi (author), Liang Chen (Thesis advisor), University of Northern British Columbia (Degree granting institution)
Abstract: Deep learning is well known as a method to extract hierarchical representations of data. This method has been widely implemented in many fields, including image classification, speech recognition, natural language processing, etc. Over the past decade, deep learning has made a great progress in solving face recognition problems due to its effectiveness. In this thesis a novel deep learning multilayer hierarchy based methodology, named Local Binary Pattern Network (LBPNet), is proposed. Unlike the shallow LBP method, LBPNet performs multi-scale analysis and gains high-level representations from low-level overlapped features in a systematic manner. The LBPNet deep learning network is generated by retaining the topology of Convolutional Neural Network (CNN) and replacing its trainable kernel with the off-the-shelf computer vision descriptor, the LBP descriptor. This enables LBPNet to achieve a high recognition accuracy without requiring costly model learning approach on massive data. LBPNet progressively extracts features from input images from test and training data through multiple processing layers, pairwisely measures the similarity of extracted features in regional level, and then performs the classification based on the aggregated similarity values. Through extensive numerical experiments using the popular benchmarks (i.e., FERET, LFW and YTF), LBPNet has shown the promising results. Its results out-perform (on FERET) or are comparable (on LFW and FERET) to other methods in the same categories, which are single descriptor based unsupervised learning methods on FERET and LFW, and single descriptor based supervise learning methods with image-restricted no outside data settings on LFW and YTF, respectively. --Leaves i-ii.
Discipline: Computer Science
Date added: 2017-03-29T17:29:47.121Z

Pages

Usage statistics