Maja Braovic

Vision based wildfire and natural risk observers

Wildfires are natural risk phenomena that cause significant economic and environmental damage. In wildfire fighting strategy it is important to detect the wildfire in its initial stage and to apply, as soon as possible, the most appropriate fire fighting action. In both cases wildfire monitoring and surveillance systems are of great importance, so in the last decade the interest for various wildfire monitoring and surveillance systems has increased, both on the research and the implementation level. This paper describes one such system named iForestFire. It is an example of advanced terrestrial vision based wildfire monitoring and surveillance system, today widely used in various Croatian National and Nature Parks and regions, but it is also a system in constant development and improvement, both on theoretical and practical level. This paper describes its last improvements in video detection part that are based on notation of observer, cogent confabulation theory and mechanism of thought. Inclusion of cogent confabulation theory allows us to expend the use of existing wildfire observers to more general natural risk observers.

Bayesian proprioceptor for forest fire observer network

The paper describes design and implementation of Bayesian proprioceptor for forest fire observer network. The proprioceptor, sometimes also referred to as network observer has task of syintactical and semantical sensor and data validation in advanced sensor network Multi agent Bayesian network is used for cooperative data analysis and data understanding having false alarm reduction as final goal. A multi agent system for data sampling and data analysis is described. The proprioceptor is deployed as a part of intelligent forest fire monitoring system (iForestFire).

Non-parametric context-based object classification in images

Segmentation and classification of objects in images is one of the most important and yet one of the most complex problems in computer vision. Many different things contribute to the complexity of this problem, but amongst those that are most significant are variations in object appearance due to various perspectives, changes in scene illumination and typically a large number of partially occluded objects in a scene. Humans have the ability to successfully deal with these complications and to recognize objects in scenes without seemingly much effort. However, to aid their recognition process, humans use different sources of information such as previous experiences, context of objects placed in a scene and rules learned in the past that suggest how the physical world should appear and behave. Computer vision is trying to replicate the same strategies used by human vision and apply them to computer systems with the aim of improving typical tasks of detection, localization and classification of image objects. In this work we propose a new model for natural image object classification using contextual information at the level of image segments. Context modeling is largely independent of appearance-based classification and proposed model enables simple upgrade of existing systems with information from global and/or local context. Context modeling is based on non-parametric use of appearance-based classification results which is a novel approach compared to previous systems that model context on a limited number of rules expressed with a fixed set of parameters. Model implementation resulted in a system that, in our simulations, showed stable improvement of the appearance-based object classification. DOI: http://dx.doi.org/10.5755/j01.itc.46.1.13610

Cogent Confabulation based Expert System for Segmentation and Classification of Natural Landscape Images

Ever since there has been an increase in the number of automatic wildfire monitoring and surveillance systems in the last few years, natural landscape images have been of great importa ...

Holonic Multi Agent System for Data Fusion in Vehicle Classification

In this paper we describe holonic organization of a multi agent system for automatic vehicle classification in a road toll system. Classification of vehicles in road toll systems is based on physical vehicle features and in this paper we focus on axle counting as the first discriminant feature for class determination. Our system relies on two main sensors—video camera and depth sensor. Video image and depth image processing is performed in several holons. The results from individual holons are fused into the final decision on a number of axles of a passing vehicle. We show that fusion of results from individual holons gives more precise results than individual holons. Holonic organization of the system aids scalability and simplifies inclusion of new sensors and new algorithms.

Augmented reality based segmentation of outdoor landscape images

The segmentation and classification of image regions are very important tasks in the field of computer vision, and yet they remain one of its greatest challenges. These challenges arise from the fact that the same objects can come in different colors, shapes and sizes, and can appear in different contexts and under different illumination. In an attempt to overcome these obstacles, in this paper we propose a system for segmentation and classification of image regions on outdoor landscape images based on augmented reality and CORINE land cover (CLC) classification. We compare the results obtained by the proposed system with the results obtained by the k-NN algorithm, and show that the proposed algorithm outperforms the k-NN one, and generally gives favorable segmentation and classification results.