Tudor Barbu

Tudor Barbu
Romanian Academy - Iasi Branch · Institute of Computer Science

Professor, Habil. PhD
PhD Coordinator at SCOSAAR: PhD positions available at http://iit.academiaromana-is.ro/iit_phd.html

About

187
Publications
23,387
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
926
Citations
Introduction
Dr. habil. Tudor Barbu is Senior Researcher I and compartment coordinator at Institute of Computer Science of the Romanian Academy. He is a SCOSAAR PhD Coordinator and Correspondent Member of Academy of Romanian Scientists. He authored 4 books, 10 chapters ,180 articles in international journals and conference volumes. He has 1250 citations and H-index=20. He is director of 3 projects, organizing committee member of 33 international conferences, editorial board member of 8 international journals
Additional affiliations
August 2017 - September 2017
University of Monastir
Position
  • Researcher
August 2016 - September 2016
University of Pittsburgh
Position
  • Professor
June 2015 - July 2015
University of Bologna
Position
  • Professor (Associate)
Description
  • GNAMPA/InDAM fellowship for university professors and senior researchers
Education
July 2014 - April 2015
Romanian Academy
Field of study
  • Computer Science
October 2000 - October 2004
Polytechnic University of Bucharest
Field of study
  • Computer Engineering
August 1998 - June 1999
University of Jyväskylä
Field of study
  • Information Technology

Publications

Publications (187)
Cover Page
Full-text available
Partial differential equations (PDEs) have been applied successfully to formulate some dynamical phenomena in many engineering domains, since these equations model continuous change. Thus, in the last 35 years, PDEs have been used to solve many challenges in various image and video processing and analysis and computer vision areas, including image...
Conference Paper
Full-text available
We consider here a novel automatic unsupervised speech-dependent speaker recognition technique. The proposed approach clusters a set of speech sequences in voice classes corresponding to the generating speakers, whose number is unknown. The speech feature vectors of these utterances are constructed by using the mel-cepstral analysis. Then, an autom...
Preprint
Full-text available
We consider here a novel automatic unsupervised speech-dependent speaker recognition technique. The proposed approach clusters a set of speech sequences in voice classes corresponding to the generating speakers, whose number is unknown. The speech feature vectors of these utterances are constructed by using the mel-cepstral analysis. Then, an autom...
Article
Full-text available
A novel content-based color image retrieval framework is proposed in this article. It uses a querying by example technique and a cluster-based image database indexing approach. A multi-scale deep learning-based image analysis is performed for this retrieval task. The scale-space representation is constructed by applying a nonlinear second-order rea...
Article
Full-text available
This research paper represents a survey describing some high-level mathematical models for digital image segmentation. They are divided into parametric and geometric partial differential equation (PDE)-based segmentation models. So, various parametric Active Contour models , which are based on PDE variational schemes, are surveyed first. The geomet...
Article
Full-text available
A novel automatic vehicle detection and tracking framework is proposed in this research work. First, the moving vehicles are detected in the frames of the video sequence by combining some deep learning and Gaussian Mixture Model-based object detection techniques. Then, the correspondence between the video objects detected in successive frames is de...
Article
Full-text available
A novel partial differential equation (PDE)—based mixed image noise removal technique is proposed in this article. The considered approach filters successfully the images corrupted by combined Poisson–Gaussian noise by using a nonlinear fourth-order PDE-based denoising model that is proposed and carefully investigated here. Thus, a rigorous mathema...
Conference Paper
Full-text available
The main objective of this work is to develop a novel multi-scale analysis framework for coronavirus disease diagnosis from Chest X-Ray images. A scale-space representation is created by applying the numerical approximation algorithm that solves a new well-posed nonlinear anisotropic diffusion-based model. A 2D circular filter-based feature extract...
Chapter
The incidence of colorectal cancers nowadays, with a high rate of mortality, justifies the effort of scientists from different disciplines in solving this problem. Early detection of pre-cancerous signs considerably increases the survival rate. Inter and intra-observer variability might be stated in lesions’ diagnosis due to the patient’s personal...
Conference Paper
A variational framework for mixed noise removal is proposed in this article. The considered variational model is based on an energy functional minimization that leads to a nonlinear second-order anisotropic diffusion-based model. This PDE (partial differential equation) model is then solved numerically by applying an approximation algorithm that is...
Conference Paper
A novel automatic deep learning-based temporal video segmentation technique is proposed in this research paper. The considered approach uses machine and deep learning solutions to detect the video shot transitions in a movie sequence. First, a high-level image feature extraction is performed on all the video frames, by applying a pre-trained convol...
Article
Full-text available
An automatic multi-scale image edge detection technique is proposed in this research work. The multiscale image analysis used by the proposed approach is based on a nonlinear anisotropic diffusion-based scale-space that is constructed by applying the finite difference method-based numerical approximation scheme of a novel well-posed second-order an...
Article
Full-text available
An overview of the state of the art mathematical models for photon-limited image restoration is provided in this research article. The most important partial differential equation (PDE)-based Poisson de-noising techniques are presented here. Thus, quantum noise filtering algorithms using Total Variation (TV) regularization-based models are describe...
Article
Full-text available
A novel unsupervised texture classification technique is proposed in this research work. The proposed method clusters automatically the textures of an image collection in similarity classes whose number is not a priori known. A nonlinear diffusion-based multi-scale texture analysis approach is introduced first. It creates an effective scale-space b...
Article
A novel multi-scale image analysis framework for rotation-invariant texture recognition is proposed in this article. A well-posed second-order nonlinear anisotropic diffusion-based model is proposed and a scale-space representation is then constructed by applying the finite difference-based numerical approximation algorithm of this PDE-based model...
Article
Full-text available
A digital image compression framework based on nonlinear partial differential equations (PDEs) is proposed in this research article. First, a feature keypoint-based sparsification algorithm is proposed for the image coding stage. The interest keypoints corresponding to various scale-invariant image feature descriptors, such as SIFT, SURF, MSER, ORB...
Article
Full-text available
A comprehensive survey on structure-based inpainting methods based on equations of engineering physics is provided in this work. Diffusion equations, describing the random motion of the micro-particles in physics, have been successfully applied in the image processing areas in the last decades. Nonlinear anisotropic diffusion-based structural inpai...
Preprint
Full-text available
This work represents a survey of the main image denoising and inpainting techniques based on variational (enery-based) schemes that lead to nonlinear partial differential equation (PDE)-based filtering models. The variational denoising and restoration approaches are described first. Then, the energy-based structural image reconstruction models are...
Article
This work represents a survey of the main image denoising and inpainting techniques based on variational (enery-based) schemes that lead to nonlinear partial differential equation (PDE)-based filtering models. The variational denoising and restoration approaches are described first. Then, the energy-based structural image reconstruction models are...
Article
Full-text available
An overview of nonlinear partial differential equation (PDE)-based structural inpainting approaches is provided in this paper. The main image interpolation techniques based on variational models are described first. Next, the state of the art inpainting methods using PDE models of various orders that do not follow variational principles are present...
Article
A partial differential equation (PDE) - based technique for filtering the Poisson noise from digital images is proposed in this work. It is based on a nonlinear second-order anisotropic diffusion-based model that is adapted for the Poisson distribution. The considered PDE model is well-posed and its unique and weak solution is computed using a fini...
Article
The paper is concerned with a qualitative analysis for a nonlinear second-order parabolic problem, subject to non-homogeneous Cauchy–Neumann boundary conditions, extending the types already studied. Under some certain assumptions, we prove the existence, estimate, regularity and uniqueness of a classical solution. The considered nonlinear second-or...
Chapter
The aim of this chapter is to describe the linear diffusion-based image filtering domain. The diffusion process theory and the existing linear partial differential equation (PDE)—based denoising techniques are discussed in the first section. Then, our main contributions to this image processing field are described in the following sections. Thus, a...
Chapter
The nonlinear diffusion-based image denoising and restoration field is addressed in this chapter. The state of the art of this image processing domain is described in the first section. Then, our major contributions in this area are detailed in the following sections. So, the second section describes the anisotropic diffusion models for image resto...
Chapter
The nonlinear PDE-based image reconstruction domain is approached in this chapter. The state of the art inpainting approaches based on differential models are described in the first section. The interpolation techniques based on variational models are described first and the PDE-based inpainting methods that do not follow variational principles are...
Chapter
An overview of additive noise reduction algorithms using nonlinear second-order partial differential equations (PDEs) is provided here. First, the state-of-the-art anisotropic diffusion models that are used for image restoration are described here. The second-order PDE-based denoising approaches using variational schemes are addressed next. Our mos...
Chapter
A state of the art on the actual methods for automatic video processing is necessary, toward further development of research methods in computer assisted diagnosis and image analysis regarding the pathology of the lower gastrointestinal tract. Automatic analysis of these types of video frames might be useful in the evaluation of the correctness of...
Conference Paper
Full-text available
An image restoration technique for halftone pattern printed pictures is provided in this paper. Our approach recovers the original analogue halftoning of the old pictures into a digital clean halftoning. It starts from a thresholded version of the scanned images from old original books, and applies several black and white spots detection and replac...
Book
The aim of this book is to describe the image restoration and inpanting techniques using partial differential equations, but focusing mainly on the novel and recent contributions of the author in these strongly related domains. Thus, some of the most important theoretical and practical results of our research in these areas are described here. The...
Conference Paper
An effective fourth-order PDE-based scheme for image restoration is proposed in this article. First a novel PDE variational model is described. Then, a nonlinear fourth-order diffusion model is obtained from it. A robust explicit numerical approximation scheme based on the finite-difference method and converging fast to the solution of this PDE is...
Article
A novel structure-based image interpolation technique is proposed in this paper. It is based on a nonlinear anisotropic diffusion model that is properly constructed for the reconstruction process. A rigorous mathematical investigation of this partial differential equation (PDE) - based scheme is then performed, its well-posedness being treated. An...
Article
Full-text available
A novel anisotropic diffusion-based image denoising and restoration approach is proposed in this paper. A variational model for image restoration is introduced first, then the corresponding Euler-Lagrange equation being determined. A nonlinear parabolic PDE model is then obtained from this equation. It is based on a novel edge-stopping function and...
Article
A nonlinear PDE-based image restoration approach is described in this article. The proposed filtering technique is based on a novel fourth-order diffusion model. Unlike many other fourth-order PDE denoising schemes, this nonlinear model provides an optimal trade-off between noise removal, image detail preservation and avoiding of undesired effects....
Chapter
An easy to compute and small colour feature vector is introduced in this paper, as a tool to be used in the process of retrieval or classification of similarly coloured digital images from very large databases. A particular set of “ab” planes from the LAB colour system is used, along with a specific configuration of colour regions within them. The...
Conference Paper
A hybrid nonlinear diffusion-based image restoration technique is proposed in this article. The novel compound PDE denoising model combines nonlinear second-order and fourth-order diffusions to achieve a more effective image enhancement. The weak solution of the combined PDE, representing the restored digital image, is determined by developing a ro...
Conference Paper
This article is concerned with hybrid image denoising and restoration using partial differential equations (PDE). A compound filtering model combining a nonlinear fourth-order PDE to a two-dimension Gaussian filter is proposed first. The obtained fourth-order diffusion model is discretized by constructing an explicit finite difference method-based...
Article
Full-text available
A second-order nonlinear parabolic PDE-based restoration model is provided in this article. The proposed anisotropic diffusion-based denoising approach is based on some robust versions of the edge-stopping function and of the conductance parameter. Two stable and consistent approximation schemes are then developed for this differential model. Our P...
Article
Full-text available
The most used approach for iris identification is based on analyzing its texture by methods introduced and refined by John Daugman. In order to obtain very good rates of correct identification, a relatively good iris image is required, i.e. the iris must have at least a radius of 70 pixels in a nearinfrared monochrome image. In latest years, we hav...
Article
We propose a technique for image denoising by solving a nonconvex optimal control problem with the state and controller connected on a manifold described by a nonlinear elliptic equation. The existence of a control able to denoise an initial blurred image is proved and the optimality conditions are determined by a passing to the limit technique in...
Article
This article describes a novel nonlinear second-order PDE model for image smoothing. The proposed denoising model is based on a second-order hyperbolic equation and provides effective detail-preserving image noise removal results. The well-posedness of this nonlinear PDE scheme is then investigated in this paper. A finite-difference based explicit...
Article
We consider a novel variational model for image reconstruction, based on second-order partial differential equations, in this paper. This inpainting approach is inspired by the anisotropic diffusion-based denoising solutions. A stable variational scheme is proposed first. Then, a nonlinear differential model is derived from it, by determining the c...
Article
A stochastic diffusion-based image restoration technique is proposed in this paper. The stochastic differential model is described first. Then, the SDE-based denoising scheme is transformed into a parabolic PDE model. Next, a rigorous mathematical investigation is performed on it. A numerical approximation scheme is then developed for this PDE-base...
Conference Paper
A novel PDE-based image restoration approach is proposed in this article. The provided PDE model is based on a linear second-order hyperbolic diffusion equation. The well-posedness of the proposed differential model and some nonlinear PDE schemes derived from it are also discussed. A consistent and fast-converging numerical approximation scheme usi...
Conference Paper
An effective nonlinear hyperbolic diffusion-based image restoration technique is proposed in this paper. The developed PDE denoising model is based on a nonlinear fourth-order hyperbolic equation. A consistent numerical discretization scheme is also constructed for this continuous model. The proposed PDE-based approach provides a satisfactory image...
Conference Paper
A multimodal biometric system is proposed in this paper. Our multimodal biometric authentication solution is based on three identifiers: iris, voice and faces. A mel-cepstral voice recognition is described first. Then, face authentication is approached by using SIFT-based features. Next, LAB-based iris recognition approach is proposed. The biometri...
Article
We consider a novel nonlinear PDE-based image denoising technique. The proposed restoration model uses second-order hyperbolic diffusion equations. It represents an improved nonlinear version of a linear hyperbolic PDE model developed recently by the author, providing more effective noise removal results whilst preserving the edges and other image...
Chapter
Full-text available
Proiectul se opreşte asupra uneia dintre problemele sensibile ale societăţii contemporane: aceea a siguranţei informatice şi a importanţei protecţiei datelor, cu precădere în domenii strategice, precum apărarea şi securitatea naţională, sănătatea, educaţia, descoperirile din domeniile ştiinţei şi tehnicii, sistemul bancar şi cel financiar etc., pre...
Conference Paper
An effective nonlinear second-order PDE image denoising and restoration model is described in this paper. The proposed anisotropic diffusion-based filtering scheme is based on some novel versions of the edge-stopping function and the conductance parameter. A consistent numerical approximation scheme is constructed for this continuous model. Our PDE...
Conference Paper
A content-based indexing approach that facilitates the iris image database retrieval process is provided in this article. Our indexing model is based on a HOG-based image feature extraction producing high-dimension feature vectors that are accessed via SAM (Spatial Access Methods). These content-based feature vectors are indexed by using a K-D-tree...
Article
In this article we propose a novel image denoising and restoration system based on a combination of second and fourth order partial differential equations. We consider a variational scheme that minimizes a energy functional composed of functions of gradient magnitude and of absolute value of Laplacian of image intensity. Then the corresponding Eule...
Conference Paper
A PDE-based image restoration model is proposed in this paper. It aims to restore degraded images that are affected by both noise and missing zones. The considered restoration approach is based on two PDE variational techniques. The first variational method performs an efficient noise reduction, while the second variational model provides the image...
Conference Paper
The paper is focused on an experimental study on positive and negative emotion vocal recognition. After some considerations about the positive and negative emotions, the paper gives a short description of the three corpuses used in the work we have accomplished. The paper describes three sets of coefficients used, the statistic features used to gen...
Article
Full-text available
An anisotropic diffusion-based image enhancement method is proposed in this article. The provided PDE denoising approach is derived from the well-known Perona-Malik nonlinear diffusion model, representing an improved version of it. Thus, we model a novel diffusivity function and explain the mathematical reasons behind it. The conductance parameter...
Article
Full-text available
A nonlinear diffusion based image denoising technique is introduced in this paper. The proposed PDE denoising and restoration scheme is based on a novel diffusivity function that uses an automatically detected conductance parameter. A robust mathematical treatment is also provided for our anisotropic diffusion model. We demonstrate that edge-st...
Conference Paper
A color-based image indexing technique that facilitates the image database search process is provided in this paper. Our indexing approach is based on a LAB color-based image feature extraction producing highdimensional feature vectors that are accessed via Spatial Access Methods. These content-based feature vectors are indexed by using a K-D-tree...

Network

Cited By

Projects

Projects (6)
Project
The main objective of this Research Topic is to disseminate valuable original research in these image and video processing and analysis domains, providing novel PDE-based approaches in these areas and bringing together the achievements of the researchers in these fields. Therefore, we encourage the authors to contribute original and high-quality submissions that present new theoretical and practical results in this field, and consider unaddressed aspects of it, as well as survey articles describing the state of the art methods.
Project
Special Issue of "Advances in Mathematical Physics", Hindawi Call for Papers: https://www.hindawi.com/journals/amp/si/626404/cfp