Indiana University-Purdue University Indianapolis (IUPUI) / The maintenance of a healthy sewer infrastructure is a major challenge due to the root damages from nearby plants that grow through pipe cracks or loose joints, which may lead to serious pipe blockages and collapse. Traditional inspections based on video surveillance to identify and localize root damages within such complex sewer networks are inefficient, laborious, and error-prone. Therefore, this study aims to develop a robust and efficient approach to automatically detect root damages and localize their circumferential and longitudinal positions in CCTV inspection videos by applying deep neural networks and computer vision techniques. With twenty inspection videos collected from various resources, keyframes were extracted from each video according to the difference in a LUV color space with certain selections of local maxima. To recognize distance information from video subtitles, OCR models such as Tesseract and CRNN-CTC were implemented and led to a 90% of recognition accuracy. In addition, a pre-trained segmentation model was applied to detect root damages, but it also found many false positive predictions. By applying a well-tuned YoloV3 model on the detection of pipe joints leveraging the Convex Hull Overlap (CHO) feature, we were able to achieve a 20% improvement on the reliability and accuracy of damage identifications. Moreover, an end-to-end deep learning pipeline that involved Triangle Similarity Theorem (TST) was successfully designed to predict the longitudinal position of each identified root damage. The prediction error was less than 1.0 feet.

No description available.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

This thesis presents a robust solution to the classification and variable selection problem when the dimension of the data, or number of predictor variables, may greatly exceed the number of observations. When faced with the problem of classifying objects given many measured attributes of the objects, the goal is to build a model that makes the most accurate predictions using only the most meaningful subset of the available measurements. The introduction of [cursive l] 1 regularized model titling has inspired many approaches that simultaneously do model fitting and variable selection. If parametric models are employed, the standard approach is some form of regularized maximum likelihood estimation. While this is an asymptotically efficient procedure under very general conditions, it is not robust. Outliers can negatively impact both estimation and variable selection. Moreover, outliers can be very difficult to identify as the number of predictor variables becomes large. Minimizing the integrated squared error, or L 2 error, while less efficient, has been shown to generate parametric estimators that are robust to a fair amount of contamination in several contexts. In this thesis, we present a novel robust parametric regression model for the binary classification problem based on L 2 distance, the logistic L 2 estimator (L 2 E). To perform simultaneous model fitting and variable selection among correlated predictors in the high dimensional setting, an elastic net penalty is introduced. A fast computational algorithm for minimizing the elastic net penalized logistic L 2 E loss is derived and results on the algorithm's global convergence properties are given. Through simulations we demonstrate the utility of the penalized logistic L 2 E at robustly recovering sparse models from high dimensional data in the presence of outliers and inliers. Results on real genomic data are also presented.

One of the oldest and most fundamental problems in statistics is the analysis of cross-classified data called contingency tables. Analyzing contingency tables is typically a question of association - do the variables represented in the table exhibit special dependencies or lack thereof? The statistical models which best capture these experimental notions of dependence are the categorical conditional independence models; however, until recent discoveries concerning the strongly algebraic nature of the conditional independence models surfaced, the models were widely overlooked due to their unwieldy implicit description. Apart from the inferential question above, this thesis asks the more basic question - suppose such an experimental model of association is known, how can one incorporate this information into the estimation of the joint distribution of the table? In the traditional parametric setting several estimation paradigms have been developed over the past century; however, traditional results are not applicable to arbitrary categorical conditional independence models due to their implicit nature. After laying out the framework for conditional independence and algebraic statistical models, we consider three aspects of estimation in the models using the minimum Euclidean (L2E), minimum Pearson chi-squared, and minimum Neyman modified chi-squared distance paradigms as well as the more ubiquitous maximum likelihood approach (MLE). First, we consider the theoretical properties of the estimators and demonstrate that under general conditions the estimators exist and are asymptotically normal. For small samples, we present the results of large scale simulations to address the estimators' bias and mean squared error (in the Euclidean and Frobenius norms, respectively). Second, we identify the computation of such estimators as an optimization problem and, for the case of the L2E, propose two different methods by which the problem can be solved, one algebraic and one numerical. Finally, we present an R implementation via two novel packages, mpoly for symbolic computing with multivariate polynomials and catcim for fitting categorical conditional independence models. It is found that in general minimum distance estimators in categorical conditional independence models behave as they do in the more traditional parametric setting and can be computed in many practical situations with the implementation provided.

As the Internet has grown in size and importance to society, it has become increasingly difficult to generate global metrics of interest that can be used to verify proposed algorithms or monitor performance. This dissertation tackles the problem by proposing several novel algorithms designed to perform Internet-wide measurements using existing or inexpensive resources. We initially address distance estimation in the Internet, which is used by many distributed applications. We propose a new end-to-end measurement framework called Turbo King (T-King) that uses the existing DNS infrastructure and, when compared to its predecessor King, obtains delay samples without bias in the presence of distant authoritative servers and forwarders, consumes half the bandwidth, and reduces the impact on caches at remote servers by several orders of magnitude. Motivated by recent interest in the literature and our need to find remote DNS nameservers, we next address Internet-wide service discovery by developing IRLscanner, whose main design objectives have been to maximize politeness at remote networks, allow scanning rates that achieve coverage of the Internet in minutes/hours (rather than weeks/months), and significantly reduce administrator complaints. Using IRLscanner and 24-hour scan durations, we perform 20 Internet-wide experiments using 6 different protocols (i.e., DNS, HTTP, SMTP, EPMAP, ICMP and UDP ECHO). We analyze the feedback generated and suggest novel approaches for reducing the amount of blowback during similar studies, which should enable researchers to collect valuable experimental data in the future with significantly fewer hurdles. We finally turn our attention to Intrusion Detection Systems (IDS), which are often tasked with detecting scans and preventing them; however, it is currently unknown how likely an IDS is to detect a given Internet-wide scan pattern and whether there exist sufficiently fast stealth techniques that can remain virtually undetectable at large-scale. To address these questions, we propose a novel model for the windowexpiration rules of popular IDS tools (i.e., Snort and Bro), derive the probability that existing scan patterns (i.e., uniform and sequential) are detected by each of these tools, and prove the existence of stealth-optimal patterns.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

The World Health Organization claims that there are more than 285 million blind and visually impaired people in the world. In the US, 25 million Americans suffer from total or partial vision loss. As a result of their impairment, they struggle with mobility problems, especially the risk of falling. According to the National Council On Aging, falls are among the primary causes for fatal injury and they are the most common cause of non-fatal trauma-related hospital admissions among older adults. Visibility, an organization that helps visually impaired people, reports that people with visual impairments are twice as likely to fall as their sighted counterparts. The Centers for Disease Control and Prevention reported that 2.5 million American adults were treated for fall-related injuries in 2013, leading to over 800,000 hospitalizations and over 27,000 deaths. The total cost of fall injuries in the United States in 2013 was $31 billion, and the financial total is expected to rise to $67.7 billion by 2020. Reducing the amount of these unexpected hospital visits saves money and expands the quality of life for the affected population. Technology has completely revolutionized how nowadays activities are conducted and how var- ious tasks are accomplished, and mobile devices are at the center of this paradigm shift. According to the Pew Research Center, 64% of American adults own a smartphone currently, and this number is trending upward. Mobile computing devices have evolved to include a plethora of data sensors that can be manipulated to create solutions for humanity, including fall prevention. Fall prevention is an area of research that focuses on strengthening safety in order to prevent falls from occurring. Many fall prevention systems use sensing devices to measure the likelihood of a fall. Sensor data are usually processed using computer vision, data mining, and machine learning techniques. This work pertains to the implementation of a smartphone-based fall prevention system for the elderly and visually impaired. The system consists of two modules: fall prevention and fall detection. Fall prevention is in charge of identifying tripping hazards in the user’s surroundings. Fall detection is in charge of detecting when falls happen and alerting a person of interest. The proposed system is challenged by multiple problems: it has to run in near real time, it has to run efficiently in a smartphone hardware, it has to process structured and unstructured environments, and many more related to image analysis (occlusion, motion blur, computational complexity, etc). The fall prevention module is divided into three parts, floor detection, object-on-floor detection, and distance estimation. The evaluation process of the best approach for floor detection achieved an accuracy of 92%, a precision of 88%, and a recall of 92%. The evaluation process of the best approach for object-on-floor detection achieved an accuracy of 90%, a precision of 56%, and a recall of 78%. The evaluation process of the best approach for distance estimation achieved a MSE error of 0.45 meters. The fall detection module is approached from two perspectives, using inertial measuring units (IMU) embedded in today’s smartphones, and using a 2D camera. The evaluation process of the solution using IMUs achieved an accuracy of 83%, a precision of 89%, and a recall of 58.2%. The evaluation process of the solution that uses a 2D camera achieved an accuracy of 85.37% and a recall of 70.97%.

Context: On airport runways, monitoring is done by Precision Runway Monitor (PRM) method with the help of radar. Most of the airports are built near the forests so there is a greater chance of mam-mal intrusion onto the runways leading to massive accidents. At many airports, there are applied old traditional, mostly manual methods in detecting mammals on the runway. Accidents caused by wildlife strikes between aircraft and mammals are increasing day to day, and this is approximately 3%-10% of all reported collisions [1]. We propose a system that monitors the airport runway by detecting mammals. Objectives: The main objective of this project is to investigate and evaluate the possibility of using thermal vision methods to detect the obstacles encountered on the runways. The system should work in real time. Methods: Mammals detection can be done by using a thermal camera with a thermal sensitivity of less than 50mK and a resolution of 640 x 480 pixels. The thermal camera uses an uncooled microbolometer sensor which is lighter, consumes less power and can see through almost all weather conditions like mist, fog, snow etc. Machine Learning based algorithms like background subtraction are used in detecting the mammal, and contours are used to estimate the size and distance. Results: As a result, the mammals moving on the runway can be detected at a distance of up to 400 m. The system estimates a distance of a moving animal and its size with an accuracy of around 90%. Conclusions: A runway monitoring system is needed to prevent wildlife strikes in airports. The proposed system prevents accidents to some extent. However, further tests are required before its commercialisation. There is a need for further quantitative and qualitative validation of the models in full-scale industry trials.

Context: Windmills became one of main sources of energy. Sincethey are placed in open areas, there are many chances that birds mayenter the wind farms and get killed or damaged. Some wind farms usepulse radar systems for saving the birds from windmills. In this pulseradar technology, the turbines are turned off automatically when a birdis detected. Another technology is ultrasonic "boom boxes", which areattached to turbines and produce high-frequency noises continuouslyto repel birds. The system we are going to propose detects the birdsentering the farm using Bluetooth technology and alerts the windmill farm operator. Using Bluetooth technology can be power efficient, ac-curate, and mainly useful for avifauna method of protection. Objectives: The main objective of the Bluetooth bird detection sys-tem is to make distance estimation possible with the help of signal strength that is measured between two Bluetooth devices where oneis placed at the wind farm and another on bird. Methods: Bird detection and distance measurement is done using a BGX13P Bluetooth transmitter and receiver. According to the distance to the bird, further steps can be taken to protect it. Simplicity Studio application is used to take the readings of the Bluetooth signalstrength of the transmitter and receiver. Results: As a result, the birds are detected at two distances from awindmill, the first distance is 250 m and the second is 175 m from thewindmill. The windmill operator is alerted when the bird is detectedat either of these distances. Conclusion: A bird detection system is built with the help of Blue-tooth technology. This system helps saving the birds from collisions with windmills. However, there is a need for further quantitative andqualitative validation of the models in full-scale industry trials.