Raquel Martínez

Detection of stationary foreground objects

Survey of the most relevant approaches for detecting stationary foreground objects.Kinds of stationary foreground objects in the literature.Typical challenges to overcome, and popular datasets to test the strategies.Descriptions of the algorithms applied in the different stages of the strategies. Detection of stationary foreground objects (i.e., moving objects that remain static throughout several frames) has attracted the attention of many researchers over the last decades and, consequently, many new ideas have been recently proposed, trying to achieve high-quality detections in complex scenarios with the lowest misdetections, while keeping real-time constraints. Most of these strategies are focused on detecting abandoned objects. However, there are some approaches that also allow detecting partially-static foreground objects (e.g. people remaining temporarily static) or stolen objects (i.e., objects removed from the background of the scene).This paper provides a complete survey of the most relevant approaches for detecting all kind of stationary foreground objects. The aim of this survey is not to compare the existing methods, but to provide the information needed to get an idea of the state of the art in this field: kinds of stationary foreground objects, main challenges in the field, main datasets for testing the detection of stationary foreground, main stages in the existing approaches and algorithms typically used in such stages.

Detection of Stationary Foreground Objects Using Multiple Nonparametric Background-Foreground Models on a Finite State Machine

There is a huge proliferation of surveillance systems that require strategies for detecting different kinds of stationary foreground objects (e.g., unattended packages or illegally parked vehicles). As these strategies must be able to detect foreground objects remaining static in crowd scenarios, regardless of how long they have not been moving, several algorithms for detecting different kinds of such foreground objects have been developed over the last decades. This paper presents an efficient and high-quality strategy to detect stationary foreground objects, which is able to detect not only completely static objects but also partially static ones. Three parallel nonparametric detectors with different absorption rates are used to detect currently moving foreground objects, short-term stationary foreground objects, and long-term stationary foreground objects. The results of the detectors are fed into a novel finite state machine that classifies the pixels among background, moving foreground objects, stationary foreground objects, occluded stationary foreground objects, and uncovered background. Results show that the proposed detection strategy is not only able to achieve high quality in several challenging situations but it also improves upon previous strategies.

Detection of static moving objects using multiple nonparametric background models

Detection of moving objects remaining static is a fundamental step in many computer vision applications, since it allows to identify potentially dangerous situations (abandoned objects) and people temporally static. Here, we propose a strategy to efficiently detect such static moving objects, which is based on three nonparametric background models (long term, medium term and short term) to detect moving objects and a novel Finite State Machine to identify when a moving object becomes static.