Mariana Rusu

Followers

Following

Public Views

Related Authors

Jaime Gallego

Lenny Ridel

Filip Rooms

Rafael Marcos Luque Baena

Universidad de Extremadura

Ezequiel López-Rubio

Universidad de Málaga

Pedro Rodolfo Melgarejo Trujillo

Nicolas Delanoue

Gerhard B Van Huyssteen

North-West University

Imtnan Qazi

Thierry BOUWMANS

Université De La Rochelle

Interests

Uploads

Papers by Mariana Rusu

Improved Heterogeneous Gaussian and Uniform Mixed Models (GU-MM) and Their Use in Image Segmentation

Recently, the combined Gauss Mixture and Uniform Distributions Mixture Model, shortly Gauss-Unifo... more Recently, the combined Gauss Mixture and Uniform Distributions Mixture Model, shortly Gauss-Uniform Mixture Model (G-U-MM) was proposed to better relate to the nature of a complex distribution and to simplify the characterization of processes that need too many Gauss functions in a standard Gauss Mixed Model (GMM). For a reasonably large class of images, the Gauss-Uniform distribution mixed models are easier to apply than the GMM models because the former ones produce significantly smaller numbers of elements in the mixture. The method has solid mathematical foundation and might be better related to the processes of image segmentation performed by humans. In addition, while computationally simple, it produces remarkable results. We discuss supplementary reasons for the use of the G-U-MM heterogeneous models in image segmentation and improve the previously presented algorithm of segmentation by removing the possible confusion between sections of Gaussian distributions and intervals of uniform distribution. Consequently, the approximation precision of the histogram and the segmentation are improved. Several examples illustrate the algorithm performance.

DownloadEdit

Fuzzy Rule-based System for Pattern Recognition and Automated Classification

Comput. Sci. J. Moldova, 2017

The sorting method contains a fuzzy system that classifies waste into three classes: dangerous (t... more The sorting method contains a fuzzy system that classifies waste into three classes: dangerous (toxic, flammable, harmful to the environment), recyclable and undetermined. To increase the decision weight since some waste is perilous (eg. batteries, spray bottle under pressure, oil drums, etc.), we implement a shapesignature identification algorithm.The extraction of objects contour in order to obtain the signature is done with contour tracking algorithm. The appropriate determination of contour also enables efficient calculation of shape and size features used in fuzzy system. The elaborated system aims to detect dangerous wastes that arrive among the recyclable waste because of people ignorance.

DownloadEdit

Computerized visual inspection applied to identification and classification of labeled chemicals

2015 7th International Conference on Electronics, Computers and Artificial Intelligence (ECAI), 2015

The algorithms for computerized identification are used in different applications: from quality i... more The algorithms for computerized identification are used in different applications: from quality inspection up to complex security systems. The purpose of this paper is to investigate and present the results of applying methods for automated identification and classification of labeled chemicals. Also, a new method based on the histogram matching is investigated and compared to the known ones.

Edit

Yet Another Method for Image Segmentation based on Histograms and Heuristics

We introduce a method for image segmentation that requires little computations, yet providing com... more We introduce a method for image segmentation that requires little computations, yet providing comparable results to other methods. While the proposed method resembles to the known ones based on histograms, it is still different in the use of the gray level distribution. When to the basic procedure we add several heuristic rules, the method produces results that, in some cases, may outperform the results produced by the known methods. The paper reports preliminary results. More details on the method, improvements, and results will be presented in a future paper.

DownloadEdit

Automated identification of objects based on normalized cross-correlation and genetic algorithm

2015 E-Health and Bioengineering Conference (EHB), 2015

The algorithms for the identification and classification of shape/object are widely used in appli... more The algorithms for the identification and classification of shape/object are widely used in applications for defects quality control up to complex security systems for persons detection, face recognition or even identification of individuals. The purpose of this paper is to find a solution to complete sorting of dangerous labeled packages with an automated method for identifying danger symbol. The pattern matching method with the normalized cross-correlation (NCC) for the identification and automat classification of tagged packages was implemented. The NCC is combined with genetic algorithm (GA) in order to improve performance of matching. The normalized correlation coefficient calculates the probable position of template in the scene image. The genetic algorithm calculates scaling and rotation of the pattern for another template matching in the scene image.

Edit

Mariana Rusu

Uploads

Papers by Mariana Rusu

Log In