Papers by Natasa Glisovic
Fuzzy Metaheuristic Algorithm for Copy - Move Forgery Detection in Image
Lecture Notes in Computer Science, 2020
One of the most common methods for counterfeiting digital images is copy-move forgery detection (... more One of the most common methods for counterfeiting digital images is copy-move forgery detection (CMFD). It implies that part of the image is copied and pasted to another part of the same image. The purpose of such changes is to hide certain image content, or to duplicate image content. The aim of this paper is to propose a new clustering algorithm for edited images. The image is divided into non-overlapping blocks. New fuzzy metric is used to calculate the distance between the blocks. In this research the metaheuristic method of the variable neighbourhood search (VNS) is used for the classification of the block. The aim of the classification is that the division should be on the changed and unchanged blocks. The results of this research are compared with the latest results from the literature dealing with this problem and it is shown that the proposed algorithm gives better results. Publicly available image databases were used. The proposed algorithm was implemented in the Python programming language.
A Hybrid System for Copy-Move Forgery Detection
Springer eBooks, Aug 24, 2021
The Skyline Image Segmentation using Color and Detail Clustering
There are a lot of different segmentation methods used to locate objects or particular scene part... more There are a lot of different segmentation methods used to locate objects or particular scene parts. One of the most interesting scene parts can be horizon or skyline. The proper skyline image segmentation includes proper division into three clusters: sky, skyline region and the rest of the image. This paper analyzes the influence of color for the skyline-based segmentation.
Optimizacija problema upravljanja odnosima koristi i troškova pri rasporedjivanju projekata primenom metaheurističkih algoritama
Оптимизација проблема управљања односима користи и трошкова при распоређивању пројеката применом метахеуристичких алгоритама
Универзитет у Београду, May 11, 2018
Poslovna ekonomija, 2016
Econophysics is a new and growing discipline that strives to connect and establish the interdepen... more Econophysics is a new and growing discipline that strives to connect and establish the interdependence between natural phenomena and human economic activities. A growing number of studies attempts to prove empirical correlation and cause-effect interdependence between sunspots and the behaviour of stock market indices. This relationship is theoretically based on the fact that sunspots cause enhanced geomagnetic activity of the Earth, and it changes human psychology and behaviour, inducing waves of
Application of neural networks in investments: A case of BELEX15 stock index
ABSTRACT The time series represent a line of observations of unintentional variable in different ... more ABSTRACT The time series represent a line of observations of unintentional variable in different time periods. The aim of the time series is, on the basis of those observations, to forecast the values of the variable. Further reactions should be predicted on the basis of the data from the past. The research covers the sample representing the Serbian (BELEX15) stock index. The aim of this study is the survey of the algorithm for predicting the reaction of the problem of this type by using the neural networks in function of prediction of the daily stock index values in the emerging market of the Republic of Serbia. The proposal of the algorithm is implemented by software and the results are significant both to academics and professionals in the subject area, especially having in mind the possibility of prediction in investments.
Fuzzy Clustering in Searching Through Information
ABSTRACT Clustering is most often used method for noticing data groupings. With an increase of a ... more ABSTRACT Clustering is most often used method for noticing data groupings. With an increase of a number of sets for testing there came the development of numerous techniques which are based on different criteria and prior knowledge on the corpus. This work presents the summary of elementary fuzzy partition clustering techniques and their application in searching through information.
Ekonomika Poljoprivrede (1979), 2016
Many profitable businesses can get bankrupt if they grow too slowly or too fast. In order to asse... more Many profitable businesses can get bankrupt if they grow too slowly or too fast. In order to assess whether a company is growing too slowly or too fast, it is necessary to establish a sustainable growth rate, that represents the maximum growth rate which the company can achieve without additional debt growing. The aim of this study is to determine the sustainable growth rate for the enterprises of the agricultural and food sector in Serbia in 2012 and 2013 and to determine whether there are differences between these two interconnected sectors. The construction sector which is not reproductively connected with agriculture and food ones, was introduced in the study as a control sector. The aim is also to establish a real sustainable growth rate of observed sectors in Serbia taking into consideration the inflation rate which in Serbia has visible fluctuations for the observed years. Real sustainable growth rate serves as a landmark for drawing conclusions about the development potential of these branches and as a conclusion of which internal or external incentives are needed to further increase sustainable growth rate, namely potential for growth. Limitation of the work scope focused research on identifying sustainable growth rate and mutual comparison of observed sectors and provided a basis for further research in the direction of the comparative analysis within the agricultural sector, as well as for determining agricultural activities with the greatest growth potential, expressed through the sustainable growth rate.
Journal of Applied Mathematics, 2014
Project planning, defining the limitations and resources by leveling the resources available, hav... more Project planning, defining the limitations and resources by leveling the resources available, have a great importance for the management projects. All these activities directly affect the duration and the cost of the project. To get a competitive value on the market, the project must be completed at the optimum time. In other to be competitive enough the optimum or near optimum solutions of time cost tradeoff and the resource leveling and resource constrained scheduling problems should be obtained in the planning phase of the project. One important aspect of the project management is activity crashing, that is, reducing activity time by adding more resources such as workers and overtime. It is important to decide the optimal crash plan to complete the project within the desired time period. The comparison of fuzzy simulated annealing and the genetic algorithm based on the crashing method is introduced in this paper to evaluate project networks and determine the optimum crashing configuration that minimizes the average project cost, caused by being late and crashing costs in the presence of vagueness and uncertainty. The evaluation results based on a real case study indicate that the method can be reliably applied to engineering projects.
Industrija, 2017
The subject of this paper is the creation and testing of an enhanced fuzzy neural network backpro... more The subject of this paper is the creation and testing of an enhanced fuzzy neural network backpropagation model for the prediction of stock market indexes, including the comparison with the traditional neural network backpropagation model. The objective of the research is to gather information concerning the possibilities of using the enhanced fuzzy neural network backpropagation model for the prediction of stock market indexes focusing on transitional markets. The methodology used involves the integration of fuzzified weights into the neural network. The research results will be beneficial both for the broader investment community and the academia, in terms of the application of the enhanced model in the investment decisionmaking, as well as in improving the knowledge in this subject matter.
The performance of the investment return prediction models: Theory and evidence
ST Segment Change Detection by Means of Wavelets
Scientific Publications of the State University of Novi Pazar, 2018
Recently, McDougall and Wotherspoon have developed a simple modification to the standard Newton m... more Recently, McDougall and Wotherspoon have developed a simple modification to the standard Newton method for finding simple roots of nonlinear equations. Their method is based on the arithmetic mean and achieves approximately 2.4 convergence order. This research presents a new iteration scheme based on the harmonic mean with the same convergence order and the comparison with the original McDougall-Wotherspoon method. Several test examples have been used and the results show good numerical behavior of the new method.
Support system in deciding paternity test which involves mutations
ABSTRACT There are many different reasons people want to have a paternity test. Sometimes alleged... more ABSTRACT There are many different reasons people want to have a paternity test. Sometimes alleged fathers are told many years later that a previous relationship resulted in the birth of a child, and they want to make sure they are actually the biological father. Some men might suspect their wives or girlfriends of infidelity and are concerned their child may not be theirs biologically and want the peace of mind a DNA test can provide. Children who were separated from their biological father at a very early age, whether because of adoption or other reasons, often want a conclusive DNA test early in the reunion process. Children sometimes learn about infidelity after the death of a parent and are curious about their genetic makeup and their family's medical history, so they have DNA tests to prove their parentage. The aim of this study was to review the system for support during the analysis of determining the paternity index includes the possibility of mutation. System is done in C# programming language that allows analysis of only the father and child in the investigation and analysis of the father, mother and child, and the case involving the testing of more children. The result is possible to include calculations and with varying degrees subpopulation.
Nova metoda za detekciju copy move promena na slikama [Not available in English]
Zahvajuci stalnom razvoju tehnologija, mobilnih telefona i digitalnih kamera, upotreba digitalnih... more Zahvajuci stalnom razvoju tehnologija, mobilnih telefona i digitalnih kamera, upotreba digitalnih fotografija je zastupljena u svakodnevnom životu. Koriste se kao izvori informacija, ali eesto i kao potrekpljujuci dokazi u forenziekim istraživanjima, kao dokazi na sudu, novinarstvu i slieno. Izmena sadržaja takvih slika je postala jako jednostavna, preko odreeenih softverskih alata, tako da ne možemo biti sigurni u originalnost slike. Promene nastaju tako sto se doda ili ukloni odreeeni sadržaj slike, ili se od vise slika pravi jedna slika, a sve u cilju pogresnog prenosa informacija. Iz tih razloga je znaeajno razvijati nove metode za detekciju takvih promena. Jedna od najvise zastupljenih promena jeste tkz. CMFD (Copy Move Forgery Detection). Promene nastaju tako sto se deo slike kopira i nalepi na neki drugi deo unutar iste slike. Takve promene se naknadno mogu skalirati, rotirati, visestruko kopirati i slieno. Obzirom da su kopirani i nalepljeni delovi sa iste slike, njihova detekcija postaje dodatno složenija, jer su karakteristike kopiranih i nalepljenih delova jako sliene (komponente suma, temperature boja, sjajnost, i slieno). Multifraktali kao i multifraktalni spektar su se pokazali dobro u samoj analizi slika, segmentaciji i izdvajanju odreeenih karakteristika slike. Mogu da opisiju signal lokalno, preko vrednosti odgovarajuceg Holder-ovog eksponenta, i globalno, preko vrednosti raspodele tih koeficijenata – multifraktalnog spektra. Stoga je predložena nova metoda za detekciju CMFD promena na slikama, zasnovana na odreeenim parametrima multifraktalnog spekra, kao i nekim statistiekim parametrima. Slika od interesa se deli na nepreklapajuce blokove fiksnih dimenzija. Svaki blok se opisuje karakteristienim vektorom, sastavljenim od parametara MF spektra i statistiekih paramatera. Za uparivanje blokova, pronalaženje duplikata i klasterovanje blokova u izmenjene i neizmenjene, predložena je nova semimetrika, zasnovana na metaheuristici. Preciznost predložene metode je potvreena eksperimentalno, testiranjem slika iz javno dostupnih baza slika, kao i testiranjem slika generisanih za potrebe istraživanja. Parametri preciznosti detekcije jesu procenat taene kao i procenat lažne detekcije, kao i preciznost i opoziv. Dobijeni rezultati pokazuju da predložena metoda daje znaeajno bolje rezultate u odnosu na rezultate koje su predstavili drugi autori. Takoee, predložena metoda ima manju kompleksnost raeunanja, u odnosu na ostale metode (karakteristieni vektor se sastoji od samo 7 deskriptora)..
Conversion of texts by applying the memory based learning
ABSTRACT Machine learning presents the often used technique in natural language processing. The c... more ABSTRACT Machine learning presents the often used technique in natural language processing. The choices of learning algorithms, their realizations and practical applications are more than heterogeneous. This paper presents the review of memory based learning organization, the concepts of well-known k-nearest neighbor algorithm and its application in the conversion of texts.
The comparative analyses of the nonparametric methods for investment return prediction
The financial market is complex, evolving and dynamic system, which has an extremely non-linear m... more The financial market is complex, evolving and dynamic system, which has an extremely non-linear movement. Thus, investment return prediction represents a significant challenge, especially because of its great diversity, unsteadiness and unstructured data with a high degree of instability and pronounced hidden connections. It is known that accurate prediction of the stock market indexes is very important for the development of effective trading strategies in investments. The main objective of the research is to perform the comparative analyses of different nonparametric methods, that is, fuzzy artificial neural networks (fuzzyANN) and genetic algorithm artificial neural networks (GAANN) for predicting the movements of the stock market indexes. The survey is conducted on the BELEX15, SBITOP, BUX and CROBEX stock market indexes. Model estimates were carried out through the prediction error MAE, MAPE and RMSE. The research results point to the adequacy of the nonparametric methods application in investments.
The system to support the decision-making in detecting genetic mutations
ABSTRACT The aim of this paper is to present a system for detection of point and frame shift muta... more ABSTRACT The aim of this paper is to present a system for detection of point and frame shift mutations of a DNA sequence as well as a reconstruction of original sequence. Entire system is implemented in the programming language C#.
Resources Conservation and Recycling, May 1, 2014
Problem of waste treatment is very important for environment and the share of tires in this scope... more Problem of waste treatment is very important for environment and the share of tires in this scope is high. The aim of this paper is to analyse retreading as one very interesting way of tires' treatment. In practice, firms involved with commercial vehicles exploitation have to decide whether to retread used tires or not, depending on the number of retreadings of used tires and travelled distances after each retreading. An approach based on Bayesian networks is proposed as a tool for decision making support in tire retreading process. Analysis is performed on database of tires' exploitation from a company of public passenger transportation and the statistical results are used as inputs to the proposed model. The results obtained according to the proposed model provide a good basis when it comes to making a decision whether to retread or not a used tire.
Uploads
Papers by Natasa Glisovic