Improved robustness and efficiency for automatic visual site monitoring
نویسنده
چکیده
Knowing who people are, where they are, what they are doing, and how they interact with other people and things is valuable from commercial, security, and space utilization perspectives. Video sensors backed by computer vision algorithms are a natural way to gather this data. Unfortunately, key technical issues persist in extracting features and models that are simultaneously efficient to compute and robust to issues such as adverse lighting conditions, distracting background motions, appearance changes over time, and occlusions. In this thesis, we present a set of techniques and model enhancements to better handle these problems, focusing on contributions in four areas. First, we improve background subtraction so it can better handle temporally irregular dynamic textures. This allows us to achieve a 5.5% drop in false positive rate on the Wallflower waving trees video. Secondly, we adapt the Dalal and Triggs Histogram of Oriented Gradients pedestrian detector to work on large-scale scenes with dense crowds and harsh lighting conditions: challenges which prevent us from easily using a background subtraction solution. These scenes contain hundreds of simultaneously visible people. To make using the algorithm computationally feasible, we have produced a novel implementation that runs on commodity graphics hardware and is up to 76× faster than our CPU-only implementation. We demonstrate the utility of this detector by modeling scene-level activities with a Hierarchical Dirichlet Process. Third, we show how one can improve the quality of pedestrian silhouettes for recognizing individual people. We combine general appearance information from a large population of pedestrians with semi-periodic shape information from individual silhouette sequences. Finally, we show how one can combine a variety of detection and tracking techniques to robustly handle a variety of event detection scenarios such as theft and left-luggage detection. We present the only complete set of results on a standardized collection of very challenging videos.
منابع مشابه
Automatic classification of highly related Malate Dehydrogenase and L-Lactate Dehydrogenase based on 3D-pattern of active sites
Accurate protein function prediction is an important subject in bioinformatics, especially wheresequentially and structurally similar proteins have different functions. Malate dehydrogenaseand L-lactate dehydrogenase are two evolutionary related enzymes, which exist in a widevariety of organisms. These enzymes are sequentially and structurally similar and sharecommon active site residues, spati...
متن کاملDETERMINATION OF SENSOR LOCATIONS FOR MONITORING OF ORCHARDS PARAMETERS USING REMOTE SENSING AND GIS
Optimal management of the farm and increasing production efficiency can be achieved by collecting accurate and appropriate information from the fields. The aim of this study is to determine the location of soil moisture sensors in pistachio orchards. For this purpose, initial information was obtained using satellite image processing. Then, using clustering method the information was clustered t...
متن کاملEfficient Strategy based on Improved Biogeography-based Algorithm for Inventory Routing problem
Researchers and urban administrators have often considered the routing problem as one of the fundamental phases in developing hazard management systems. In this research, a routing problem is investigated and analyzed by proposing an enhanced metaheuristic algorithm based on biogeography. In this problem, the production planning, inventory management, and distribution planning have been conside...
متن کاملSnowWatch: Snow Monitoring through Acquisition and Analysis of User-Generated Content
We present a system for complementing snow phenomena monitoring with virtual measurements extracted from public visual content. The proposed system integrates an automatic acquisition and analysis of photographs and webcam images depicting Alpine mountains. In particular, the technical demonstration consists in a web portal that interfaces the whole system with the population. It acts as an ent...
متن کاملComputerized Addenbrooke’s Cognitive Examination for Neurocognitive Monitoring of Children with Benign and Malignant Brain Cerebellar Tumor
Background: Brain tumors in the cerebellum site, especially in the cranial nuclei, are most commonly associated with cognitive impairment in children, but the differences in dimensions of cognitive impairments in pediatric benign and malignant tumors are rarely studied. Meanwhile, cognitive impairments in this area are required to be widely studied. This study aimed at using Computerized Addenb...
متن کاملWireless Monitoring System Based on the Non-uniform Stratified WSN in Viticulture
Due to the wide coverage area of vineyards, the humid climate and the complex terrain in southern China, we present a viticulture monitoring system based on the non-uniform stratified WSN network and Wi-Fi technology. Firstly, we propose the improved virtual grid system, construct the non-uniform virtual grid by using the dynamic side length and construct a stratified routing structure by using...
متن کامل