Péter Burai

Multi-sensor integration and mapping strategies for the detection and remediation of the red mud spill in Kolontar, Hungary: Estimating the thickness of the spill layer using hyperspectral imaging and Lidar

The village of Kolontár, Hungary had became the site of one of the largest industrial spills in Europe on October 4th, 2010. The primary objective of the hyperspectral remote sensing mission was monitoring needed to estimate environmental damage, the precise size of the polluted area, the rating of substance concentration in the mud, and the overall condition of the flooded district. The secondary objective was aimed to provide geodetic data necessary for the high-resolution visual information from the data obtained with an additional Lidar survey, and for coherent modelling of the event. For quick assessment and remediation purposes, it was deemed important to estimate the thickness of the red mud, particularly the areas where the depths of the layer were more than 3cm. The results showed that some of the existing tools can be readily modified and implemented to get the most out of the available advanced remotely sensed data.

Mapping invasive vegetation using AISA Eagle airborne hyperspectral imagery in the Mid-Ipoly-Valley

The primary goal of this study was to investigate the utility of high ground (1 m) and spectral (253 bands) resolution airborne hyperspectral imagery and several classification approaches for detecting the most problematic invasive plant species in the Mid-Ipoly-Valley. AISA Eagle II airborne sensor was applied to map a NATURA-2000 site on the Hungarian-Slovak cross-border site. This study focused on mapping of goldenrod (Solidago sp.) and milkweed (Asclepias syriaca) species in two selected study areas. During the image analysis, SAM classification method was used by selected dataset. Further classification methods were applied on MNF dataset. Feature selection (SEATH) tool was used to identify relevant feature from original MNF dataset. Maximum likelihood classification applied on the selected MNF dataset provided more accurate result than other methods. The primary outcome of this study was a comparison of different image classification methods to evaluate invasive species.

Application of advanced technologies for the detection of pollution migration

With the use of hyperspectral imagery the distribution of pyritic minerals (sphalerite, pyrite galena) in the area was defined. Both of the mineral formations occur at the same distribution in flotation sludge reservoir. The results showed that hyperspectral remote sensing is an effective tool for the characterization and modeling the distribution of Pb, Zn and Fe containing minerals at the examined heavy metal polluted sites. Therefore this technology could be used for fast environmental assessment since it provides wide range of information in no time. Although, validation (field measurements, soil samples) is necessary to carry out in order to obtain more precise results and nevertheless, the effect of the vegetation and other objects should also be assessed.

Zooming on Aerial Survey

The aim of this chapter is to provide a general overview about the main components of a developed UAS mapping system, the survey, and processing procedure. At first (4.1), a brief introduction is given about basic operational elements and accessories of UAS. Then, recent camera/sensor technologies allowing various survey solutions are going to be discussed. Once these hardware components are presented, the detailed workflow of a basic UAV-based mapping procedure is described (4.2). A further discussion focuses not only on the analytical or planning phases but also on providing useful information on the operational and processing parts as well (4.3). Then, there comes image acquisition and project planning (4.4). The photogrammetry-based image processing requires detailed expertise and attention; Sect. 4.5 maybe helpful to avoid potential mistakes. The last section (4.6) summarizes some aspects of the use of LiDAR technologies in UAV-based surveys.

Pixelwise segmentation of uterine wall in endoscopic video frame using convolutional neural networks.

Though the number of in vitro fertilization (IVF) has been rising continuously from the beginning of the new millennium, however the success rate of the implantations remained low. According to the statistics, the main reason of unsuccessful IVF relates to the woman factors. The aim of our research project is to provide an automatic image processing based decision support system for the gynecologists which tries to help medical experts to determine the most appropriate time for the insemination. In this paper, we present the first component of this tool, which deals with the preprocessing of the videos about the uterus for further examinations. It includes the segmentation of the video frames by fully convolutional neural network (FCNN) to determines the region of interest. The chosen model has been trained on 4000 images acquired during real hysteroscopic surgeries and tested on other 716 ones. We have achieved 92% segmentation accuracy regarding the correct recognition of the fundus.

Classification of energy tree species using support vector machines

This paper investigates whether the combination of airborne hyperspectral imagery (Aisa EAGLE II) and image classification methods (MLC, SVM) using feature extraction can discriminate among species and clones of energy trees. The trees examined have similar morphological traits due to limitation of detection. The image classification was applied on a spectrally selected and transformed (PCA, MNF) dataset. A binary tree SVM classifier was developed in accordance with the principle of SVM, based on the Jeffries-Matusita (JM) separability measure of selected classes. The adaptive binary tree SVM on MNF-transformed dataset provided more accurate results than applied MLC and multiclass SVM methods. The primary outcome of this study was a comparison of support vector machines (SVM) classification methods to evaluate species or clones of energy plants. In this paper, an adaptive binary tree SVM classifier (ABTSVM) is proposed to increase the accuracy of subspecies level.