Hamid Dehghani

Using high-resolution satellite imagery to provide a relief priority map after earthquake

After the earthquake occurrence, collecting correct information about the extent of damage is essential for managing critical conditions and allocating limited resources. The prepared building damage maps sometimes bring about waste of time required for rescuing individuals under the rubble by wrongly conducting rescue teams toward regions with a lower rescue priority. In this research, an algorithm based on using a proposed standard at database level was developed to prioritize damaged buildings by considering five key elements of land use type, the degree of damage to buildings, the land use differentiation index, time of the highest population density in each land use, and time of disaster’s incidence. The steps of the proposed method which was implemented in the MATLAB environment include: detecting buildings on the pre- and post-event imagery, implementing texture features for each candidate building, choosing the optimal features by genetic algorithm, determining the degree of building damage in three classes of negligible damage, substantial damage, and heavy damage by using the difference between chosen features as inputs of the designed neurofuzzy inference system. Data collected from field observations were compared to the output obtained from the proposed algorithm. This comparison presented a general accuracy of 88% and Kappa coefficient of 79% in the classification of buildings into three damage classes. The proposed standard then was used for classifying damaged buildings into relief priorities of high, medium, and low. Findings revealed that the relief priority map could be a basis for correct guidance of relief and rescue teams during crucial times following earthquakes.

A proposed spatial index to prioritize damaged buildings for allocating USAR operations

Abstract A spatial index using fuzzy hierarchical analysis (FAHP) is proposed in this study for prioritizing damaged buildings in the allocation of search and rescue operations after the earthquake disaster. The relevant prioritization criteria have been identified through literature review and interviews with 22 relief managers; the relative importance of these criteria and sub criteria has been computed using the FAHP method. The GIS layers equivalent to the selected criteria were prepared and integrated with one another after normalization in the GIS platform. The proposed method to prioritize the damaged buildings was implemented in the city of Varzeghan in the East Azerbaijan province of Iran. The obtained priority map, with five prioritization classes, is presented. Single-parameter sensitivity analysis method identifies the criteria ‘hazardous facilities’, ‘degree of building blockage’ and ‘chance of survival’ as the most effective criteria for prioritizing damaged buildings.

A multiple classifier template for hyperspectral images classification

Error and uncertainty in remotely sensed data come from several sources, and can be increased or mitigated by the processing to which that data is subjected. For example, the assumed models for probability distribution of the classes in some classifiers are inaccurate and increase error in results of classification. A solution for reducing uncertainty and error and improvement the results of classification process is the use of multiple classifiers. However, previous work clearly showed that such classification systems are effective only if, the errors of their classifiers are independent. In this paper a template is proposed that increases the flexibility for selection the proper classifiers and can form the multiple classifier system that error independence problem is met. In addition to, this scheme is an effective plan to deal with limitation of training samples problem. At the beginning of this method, spectral bands are categorized in several small groups, based on spectral correlation criterion. Information of each group is used as a new source. These sources are reconsidered separately and classified based on one or more models, such as MLC and neural networks. With using of the benefits of all primary classifiers, it is made sure that the results of the fused local decisions are accurate enough. After providing the primary decisions, some rules are used in decision fusion center to determine the final class of pixels. Reported results on the hyperspectral remote sensing images show that this approach allows one to has effective classifier

A Clustering Technique for Remote Sensing Images Using Combination of Watershed Algorithm and Gustafson-Kessel Clustering

In clustering of remote sensing images by using conventional algorithms, not detects the boundaries of image properly. In this paper, an image clustering algorithm based on watershed algorithm and Gustafson-Kessel fuzzy clustering has been proposed. Initially, the watershed algorithm is used for segmentation of the image that is obtained of summing image derivative with the original image. Then, the average of pixels spectrum in each segment is selected as a representative of that area and using combination of the neighboring pixels data of each area with neighbor areas and Gustafson-kessel clustering, the average spectrum of different areas is clustered. Then to improve the clustering results, a new partition matrix is calculated for each area of the image according to the characteristics of neighbor areas. Because the remote sensing images are including a high noise level, proposed algorithm have a greater ability to opposition with noise than watershed algorithm and appears the image edges better. The results of proposed algorithm on a sample of remote sensing image show practicality and efficiency of the proposed algorithm.

SAR target recognition using behaviour library of different shapes in different incidence angles and polarisations

ABSTRACT Target recognition is one of the most important issues in the interpretation of the synthetic aperture radar (SAR) images. Modelling, analysis, and recognition of the effects of influential parameters in the SAR can provide a better understanding of the SAR imaging systems, and therefore facilitates the interpretation of the produced images. Influential parameters in SAR images can be divided into five general categories of radar, radar platform, channel, imaging region, and processing section, each of which has different physical, structural, hardware, and software sub-parameters with clear roles in the finally formed images. In this paper, for the first time, a behaviour library that includes the effects of polarisation, incidence angle, and shape of targets, as radar and imaging region sub-parameters, in the SAR images are extracted. This library shows that the created pattern for each of cylindrical, conical, and cubic shapes is unique, and due to their unique properties these types of shapes can be recognised in the SAR images. This capability is applied to data acquired with the Canadian RADARSAT1 satellite.

Analytical modelling and software implementation of synthetic aperture radars

ABSTRACT Synthetic aperture radar (SAR) imaging systems are a complement to passive remote sensing. However, the process of image formation is so complex that the final image in the system is formed only after three basic steps: raw data acquisition, forming the signal and the image space. In addition, there are various factors that are involved in the information recorded by SAR within the system and outside the system, such as radar, platform, processing algorithm, imaging region and channel. Each of these factors has been formed by many parameters and this issue adds to the complexity of the behaviour of SAR. Therefore, due to the complexity, providing a model that describes how the SAR imaging system works is highly important. In this paper, the performance of SAR in the image formation section is analytically modelled at first, and then implemented as software. Raw data acquisition is performed in CST software and the signal and image formation are performed in MATLAB software. This implementation provides many abilities, such as better interpretation of SAR images, simulating the effect of the important parameters in SAR images, etc.

Evaluation of camouflage effectiveness using hyperspectral images

Abstract. Recent advances in camouflage engineering have made it more difficult to detect targets. Assessing the effectiveness of camouflage against different target detection methods leads to identifying the strengths and weaknesses of camouflage designs. One of the target detection methods is to analyze the content of the scene using remote sensing hyperspectral images. In the process of evaluating camouflage designs, there must be comprehensive and efficient evaluation criteria. Three parameters were considered as the main factors affecting the target detection and based on these factors, camouflage effectiveness assessment criteria were proposed. To combine the criteria in the form of a single equation, the equation used in target visual search models was employed and for determining the criteria, a model was presented based on the structure of the computational visual attention systems. Also, in software implementations on the HyMap hyperspectral image, a variety of camouflage levels were created for the real targets in the image. Assessing the camouflage levels using the proposed criteria, comparing and analyzing the results can show that the provided criteria and model are effective for the evaluation of camouflage designs using hyperspectral images.

Automatic modulation classification using an 8PSK demodulator for variants of QPSK

In this paper a feature-based modulation classification algorithm is developed for discriminating PSK signals. The candidate modulation types are assumed to be QPSK, OQPSK, π/4 DQOSK and 8PSK. The proposed method applies an 8PSK baseband demodulator in order to extract required features from observed symbols. The received signal with unknown modulation type is demodulated by an 8PSK demodulator whose output is considered as a finite state machine with different states and transitions for each candidate modulation. Estimated probabilities of particular transitions constitute the discriminating features. The obtained features are given to a classifier which decides on the modulation type of the received signal, according to the patterns of candidate modulations. In order to evaluate the performance of the proposed method, the probability of correct classification is computed with different number of observed symbols and SNR conditions by carrying out several simulations. The results show that the proposed method offers more accurate classification compared to previous methods for classifying variants of QPSK.