M. Scuturici - Academia.edu (original) (raw)
Papers by M. Scuturici
ArXiv, 2018
Within the big data tsunami, relational databases and SQL are still there and remain mandatory in... more Within the big data tsunami, relational databases and SQL are still there and remain mandatory in most of cases for accessing data. On the one hand, SQL is easy-to-use by non specialists and allows to identify pertinent initial data at the very beginning of the data exploration process. On the other hand, it is not always so easy to formulate SQL queries: nowadays, it is more and more frequent to have several databases available for one application domain, some of them with hundreds of tables and/or attributes. Identifying the pertinent conditions to select the desired data, or even identifying relevant attributes is far from trivial. To make it easier to write SQL queries, we propose the notion of SQL query completion: given a query, it suggests additional conditions to be added to its WHERE clause. This completion is semantic, as it relies on the data from the database, unlike current completion tools that are mostly syntactic. Since the process can be repeated over and over again...
Lecture Notes in Computer Science, 2003
In this paper we propose a topological 1 model for image database query using neighborhood graphs... more In this paper we propose a topological 1 model for image database query using neighborhood graphs. A related neighborhood graph is built from automatically extracted low-level features, which represent images as points of p \ space. Graph exploration correspond to database browsing, the neighbors of a node represent similar images. In order to perform query by example, we define a topological query model. The query image is inserted in the graph by locally updating the neighborhood graph. The topology of an image database is more informative than a similarity measure usually applied in content based image retrieval, as proved by our experiments.
Fourth European Conference on Universal Multiservice Networks (ECUMN'07), 2007
Pervasive systems enable us to have an overview of what digital environments will look like in th... more Pervasive systems enable us to have an overview of what digital environments will look like in the future. Opportunities given by pervasive systems, both in terms of applications and services to the user are manifold and very promising. From the user point of view, privacy and security of her personal data is a real issue, which must be addressed to make pervasive systems accepted. A wide adoption of pervasive systems can not be possible without an integrated approach to security. We propose a model of security and privacy for pervasive environments, integrated with an architecture, namely PerSE, in which privacy is a main concern and is at the core of the conception.
Extraction et Gestion …, 2004
This paper presents a local no-reference blur assessment method in natural macro-like images. The... more This paper presents a local no-reference blur assessment method in natural macro-like images. The purpose is to decide the blurriness of the object of interest. In our case, it represents the first step for a plant recognition system. Blur detection works on small non-overlapping blocks using wavelet decomposition and edge classification. At the block level the number of edges is less than on global images. A new set of rules is obtained by a supervised decision tree algorithm trained on a manually labelled base of 1500 blurred/un-blurred images. Our purpose is to achieve a qualitative decision of the blurriness/sharpness of the object of interest making it the first step towards a segmentation process. Experimental results show this method outperforms two other methods found in literature, even if applied on a block basis. Together with a pre-segmentation step, the method allows to decide if the object of interest (leaf, flower) is sharp in order to extract precise botanical key id...
Lecture Notes in Computer Science, 2013
There are several approaches for performance evaluation of image processing algorithms in video-b... more There are several approaches for performance evaluation of image processing algorithms in video-based surveillance systems: Precision/ Recall, Receiver Operator Characteristics ROC, F-measure, Jaccard Coefficient, etc. These measures can be used to find good values for input parameters of image segmentation algorithms. Different measures can give different values of these parameters, considered as optimal by one criterion, but not by another. Most of the times, the measures are expressed as a compromise between two of the three aspects that are important for a quality assessment: Precision, Recall and Specificity. In this paper, we propose a new 3-dimensional measure D prs , which takes into account all of the three aspects. It can be considered as a 3D generalization of 2D ROC analysis and Precision/Recall curves. To estimate the impact of parameters on the quality of the segmentation, we study the behavior of this measure and compare it with several classical measures. Both objective and subjective evaluations confirm that our new measure allows to determine more stable parameters than classical criteria, and to obtain better segmentations of images.
2006 ACS/IEEE International Conference on Pervasive Services
In order to implement reactive and proactive functionalities in a pervasive environment, contextu... more In order to implement reactive and proactive functionalities in a pervasive environment, contextual data must be processed. One of the most important features of the context is the position of the users and the devices. In this paper, we describe a method to determine the position of a WiFi enabled device. The prediction is based on the signal strength of the available access points. The prediction model is built from a database containing the signal strength measured in some known locations. The result is the name of the room/office where the device is localised. We also present a usage scenario, in which the user/device position is used to start proactive actions in our pervasive service environment called PerSE.
2008 19th International Conference on Pattern Recognition, 2008
This article introduces new similarity measures between trajectories, in order to detect uncommon... more This article introduces new similarity measures between trajectories, in order to detect uncommon behaviors. These measures are used to find the most common trajectories in a sequence, using an implicit aggregation method. They may be applied to trajectories of objects tracked in real time. Moreover, by combining one or more measures, it is possible to variate the impact of the
Fifth Annual IEEE International Conference on Pervasive Computing and Communications Workshops (PerComW'07), 2007
A growing interest for pervasive applications and an increasing diversity of pervasive computing ... more A growing interest for pervasive applications and an increasing diversity of pervasive computing devices integrated in our surroundings demand incorporating context-awareness in such applications in order to protect users from being disturbed by such services while on their regular duty. The behavior of these applications should depend not only on their internal state and user interactions but also on the context sensed during their execution. Context and context awareness, therefore, are the key components of pervasive computing so as to perform tasks on behalf of users. In this paper, we propose ontology based reusable context model. The model facilitates the context reasoning by providing structure for contexts, rules and their semantics. Initial prototype of the use of the model in a multi-domain platform is created and the result obtained is promising.
2007 2nd International Conference on Digital Information Management, 2007
Dejene Ejigu Laboratoire LIRIS-UMR-CNRS 5205, INSA de Lyon, 7 avenue Jean Capelle, 69621 Villeurb... more Dejene Ejigu Laboratoire LIRIS-UMR-CNRS 5205, INSA de Lyon, 7 avenue Jean Capelle, 69621 Villeurbanne cedex, France dejene.ejigu@insa-lyon.fr ... Marian Scuturici Laboratoire LIRIS-UMR-CNRS 5205, INSA de Lyon, 7 avenue Jean Capelle, 69621 Villeurbanne ...
Fourth International Conference on Information Technology (ITNG'07), 2007
Abstract A growing interest for pervasive applications and the concern that users should not be d... more Abstract A growing interest for pervasive applications and the concern that users should not be disturbed by such services while they are on their regular duty have introduced several new challenges. The behavior of these applications should depend not only on their internal ...
Lecture Notes in Computer Science, 2013
The problem we address in this paper is the segmentation and hierarchical grouping in digital ima... more The problem we address in this paper is the segmentation and hierarchical grouping in digital images. In terms of image acquisition protocol, no constraints are posed to the user. At first, a histogram thresholding provides numerous segments where a homogeneity criterion is respected. Segments are merged together using similarity properties and aggregated in a hierarchy based on spatial inclusions. Shape and color features are extracted on the produced segments. Tests performed on Oxford Flower 17 [8] show that our method outperforms a similar one and allow the relevant object selection from the hierarchy. In our case, this approach represents the first stage towards flower variety identification.
Journal of Computers, 2008
The behavior of pervasive applications depend not only on their internal state and user interacti... more The behavior of pervasive applications depend not only on their internal state and user interactions but also on contexts sensed during their execution. In this work, we present a collaborative context-aware service platform based on hybrid context management model (enhanced CoCA). It performs reasoning and decisions based on context data, context semantics, and related rules and policies. Such data is organized into a hybrid context management model (HCoM). The platform also introduces a neighborhood collaboration mechanism to facilitate peer collaboration in order to share their resources. We have developed an initial prototype of the platform. Our preliminary test shows that the platform is a promising data independent development environment for pervasive context-aware applications and it possesses good standard of scalability.
2010 6th International Conference on Emerging Technologies (ICET), 2010
The aim of this paper is to present an algorithm for multiple object tracking and video summariza... more The aim of this paper is to present an algorithm for multiple object tracking and video summarization in a scene filmed by one or several cameras. We propose a computationally efficient real time human tracking algorithm, which can 1) track objects inside the field of view (FOV) of a camera even in case of occlusions; 2) recognize objects that quit and then return on a camera's FOV; 3) recognize objects passing through different cameras FOV. We propose a simple 1-D appearance model, called vertical feature (VF), view and size invariant, which is stored in a database in order to help object recognition. We combine it with other motion features like position and velocity for real-time tracking. We find the k closest matches of current object and select the one whose predicted position is closest to the current object position. Our algorithm shows good capabilities for objects tracking even with the change of object view angle and also with the partial change of shape. We compare our algorithm with appearance based and motion based algorithms and show the advantage of a combined approach.
Lecture Notes in Computer Science, 2012
Life Cycle Assessment provides a well-accepted methodology for modelling environmental impacts of... more Life Cycle Assessment provides a well-accepted methodology for modelling environmental impacts of human activities. This methodology relies on the decomposition of a studied system into interdependent processes. Several organisations provide processes databases containing several thousands of processes with their interdependency links. The usual work-flow to manage those databases is based on the manipulation of individual processes which turns out to be a very harnessing work. We propose a new work-flow for LCA inventory databases maintenance based on the addition of semantic information to the processes they contained. This method eases considerably the modelling process and also offers a more understandable model of the dependencies links. In this paper, we explain our approach and some key parts of the implementation. We also present a case study based on the U.S. electricity production and an experiment on the scalability of our implementation.
Journal of Experimental & Theoretical Artificial Intelligence, 2005
... Publication Hors Liris. Topological Representation Model for Image Database Query. Mihaela Sc... more ... Publication Hors Liris. Topological Representation Model for Image Database Query. Mihaela Scuturici[LIRIS] , J Clech , Vasile-Marian Scuturici[LIRIS] , Djamel Zighed. 1/2005. Journal of Experimental & Theoretical Artificial Intelligence, Vol. 17, No. 1-2, January-June 2005. ...
ArXiv, 2018
Within the big data tsunami, relational databases and SQL are still there and remain mandatory in... more Within the big data tsunami, relational databases and SQL are still there and remain mandatory in most of cases for accessing data. On the one hand, SQL is easy-to-use by non specialists and allows to identify pertinent initial data at the very beginning of the data exploration process. On the other hand, it is not always so easy to formulate SQL queries: nowadays, it is more and more frequent to have several databases available for one application domain, some of them with hundreds of tables and/or attributes. Identifying the pertinent conditions to select the desired data, or even identifying relevant attributes is far from trivial. To make it easier to write SQL queries, we propose the notion of SQL query completion: given a query, it suggests additional conditions to be added to its WHERE clause. This completion is semantic, as it relies on the data from the database, unlike current completion tools that are mostly syntactic. Since the process can be repeated over and over again...
Lecture Notes in Computer Science, 2003
In this paper we propose a topological 1 model for image database query using neighborhood graphs... more In this paper we propose a topological 1 model for image database query using neighborhood graphs. A related neighborhood graph is built from automatically extracted low-level features, which represent images as points of p \ space. Graph exploration correspond to database browsing, the neighbors of a node represent similar images. In order to perform query by example, we define a topological query model. The query image is inserted in the graph by locally updating the neighborhood graph. The topology of an image database is more informative than a similarity measure usually applied in content based image retrieval, as proved by our experiments.
Fourth European Conference on Universal Multiservice Networks (ECUMN'07), 2007
Pervasive systems enable us to have an overview of what digital environments will look like in th... more Pervasive systems enable us to have an overview of what digital environments will look like in the future. Opportunities given by pervasive systems, both in terms of applications and services to the user are manifold and very promising. From the user point of view, privacy and security of her personal data is a real issue, which must be addressed to make pervasive systems accepted. A wide adoption of pervasive systems can not be possible without an integrated approach to security. We propose a model of security and privacy for pervasive environments, integrated with an architecture, namely PerSE, in which privacy is a main concern and is at the core of the conception.
Extraction et Gestion …, 2004
This paper presents a local no-reference blur assessment method in natural macro-like images. The... more This paper presents a local no-reference blur assessment method in natural macro-like images. The purpose is to decide the blurriness of the object of interest. In our case, it represents the first step for a plant recognition system. Blur detection works on small non-overlapping blocks using wavelet decomposition and edge classification. At the block level the number of edges is less than on global images. A new set of rules is obtained by a supervised decision tree algorithm trained on a manually labelled base of 1500 blurred/un-blurred images. Our purpose is to achieve a qualitative decision of the blurriness/sharpness of the object of interest making it the first step towards a segmentation process. Experimental results show this method outperforms two other methods found in literature, even if applied on a block basis. Together with a pre-segmentation step, the method allows to decide if the object of interest (leaf, flower) is sharp in order to extract precise botanical key id...
Lecture Notes in Computer Science, 2013
There are several approaches for performance evaluation of image processing algorithms in video-b... more There are several approaches for performance evaluation of image processing algorithms in video-based surveillance systems: Precision/ Recall, Receiver Operator Characteristics ROC, F-measure, Jaccard Coefficient, etc. These measures can be used to find good values for input parameters of image segmentation algorithms. Different measures can give different values of these parameters, considered as optimal by one criterion, but not by another. Most of the times, the measures are expressed as a compromise between two of the three aspects that are important for a quality assessment: Precision, Recall and Specificity. In this paper, we propose a new 3-dimensional measure D prs , which takes into account all of the three aspects. It can be considered as a 3D generalization of 2D ROC analysis and Precision/Recall curves. To estimate the impact of parameters on the quality of the segmentation, we study the behavior of this measure and compare it with several classical measures. Both objective and subjective evaluations confirm that our new measure allows to determine more stable parameters than classical criteria, and to obtain better segmentations of images.
2006 ACS/IEEE International Conference on Pervasive Services
In order to implement reactive and proactive functionalities in a pervasive environment, contextu... more In order to implement reactive and proactive functionalities in a pervasive environment, contextual data must be processed. One of the most important features of the context is the position of the users and the devices. In this paper, we describe a method to determine the position of a WiFi enabled device. The prediction is based on the signal strength of the available access points. The prediction model is built from a database containing the signal strength measured in some known locations. The result is the name of the room/office where the device is localised. We also present a usage scenario, in which the user/device position is used to start proactive actions in our pervasive service environment called PerSE.
2008 19th International Conference on Pattern Recognition, 2008
This article introduces new similarity measures between trajectories, in order to detect uncommon... more This article introduces new similarity measures between trajectories, in order to detect uncommon behaviors. These measures are used to find the most common trajectories in a sequence, using an implicit aggregation method. They may be applied to trajectories of objects tracked in real time. Moreover, by combining one or more measures, it is possible to variate the impact of the
Fifth Annual IEEE International Conference on Pervasive Computing and Communications Workshops (PerComW'07), 2007
A growing interest for pervasive applications and an increasing diversity of pervasive computing ... more A growing interest for pervasive applications and an increasing diversity of pervasive computing devices integrated in our surroundings demand incorporating context-awareness in such applications in order to protect users from being disturbed by such services while on their regular duty. The behavior of these applications should depend not only on their internal state and user interactions but also on the context sensed during their execution. Context and context awareness, therefore, are the key components of pervasive computing so as to perform tasks on behalf of users. In this paper, we propose ontology based reusable context model. The model facilitates the context reasoning by providing structure for contexts, rules and their semantics. Initial prototype of the use of the model in a multi-domain platform is created and the result obtained is promising.
2007 2nd International Conference on Digital Information Management, 2007
Dejene Ejigu Laboratoire LIRIS-UMR-CNRS 5205, INSA de Lyon, 7 avenue Jean Capelle, 69621 Villeurb... more Dejene Ejigu Laboratoire LIRIS-UMR-CNRS 5205, INSA de Lyon, 7 avenue Jean Capelle, 69621 Villeurbanne cedex, France dejene.ejigu@insa-lyon.fr ... Marian Scuturici Laboratoire LIRIS-UMR-CNRS 5205, INSA de Lyon, 7 avenue Jean Capelle, 69621 Villeurbanne ...
Fourth International Conference on Information Technology (ITNG'07), 2007
Abstract A growing interest for pervasive applications and the concern that users should not be d... more Abstract A growing interest for pervasive applications and the concern that users should not be disturbed by such services while they are on their regular duty have introduced several new challenges. The behavior of these applications should depend not only on their internal ...
Lecture Notes in Computer Science, 2013
The problem we address in this paper is the segmentation and hierarchical grouping in digital ima... more The problem we address in this paper is the segmentation and hierarchical grouping in digital images. In terms of image acquisition protocol, no constraints are posed to the user. At first, a histogram thresholding provides numerous segments where a homogeneity criterion is respected. Segments are merged together using similarity properties and aggregated in a hierarchy based on spatial inclusions. Shape and color features are extracted on the produced segments. Tests performed on Oxford Flower 17 [8] show that our method outperforms a similar one and allow the relevant object selection from the hierarchy. In our case, this approach represents the first stage towards flower variety identification.
Journal of Computers, 2008
The behavior of pervasive applications depend not only on their internal state and user interacti... more The behavior of pervasive applications depend not only on their internal state and user interactions but also on contexts sensed during their execution. In this work, we present a collaborative context-aware service platform based on hybrid context management model (enhanced CoCA). It performs reasoning and decisions based on context data, context semantics, and related rules and policies. Such data is organized into a hybrid context management model (HCoM). The platform also introduces a neighborhood collaboration mechanism to facilitate peer collaboration in order to share their resources. We have developed an initial prototype of the platform. Our preliminary test shows that the platform is a promising data independent development environment for pervasive context-aware applications and it possesses good standard of scalability.
2010 6th International Conference on Emerging Technologies (ICET), 2010
The aim of this paper is to present an algorithm for multiple object tracking and video summariza... more The aim of this paper is to present an algorithm for multiple object tracking and video summarization in a scene filmed by one or several cameras. We propose a computationally efficient real time human tracking algorithm, which can 1) track objects inside the field of view (FOV) of a camera even in case of occlusions; 2) recognize objects that quit and then return on a camera's FOV; 3) recognize objects passing through different cameras FOV. We propose a simple 1-D appearance model, called vertical feature (VF), view and size invariant, which is stored in a database in order to help object recognition. We combine it with other motion features like position and velocity for real-time tracking. We find the k closest matches of current object and select the one whose predicted position is closest to the current object position. Our algorithm shows good capabilities for objects tracking even with the change of object view angle and also with the partial change of shape. We compare our algorithm with appearance based and motion based algorithms and show the advantage of a combined approach.
Lecture Notes in Computer Science, 2012
Life Cycle Assessment provides a well-accepted methodology for modelling environmental impacts of... more Life Cycle Assessment provides a well-accepted methodology for modelling environmental impacts of human activities. This methodology relies on the decomposition of a studied system into interdependent processes. Several organisations provide processes databases containing several thousands of processes with their interdependency links. The usual work-flow to manage those databases is based on the manipulation of individual processes which turns out to be a very harnessing work. We propose a new work-flow for LCA inventory databases maintenance based on the addition of semantic information to the processes they contained. This method eases considerably the modelling process and also offers a more understandable model of the dependencies links. In this paper, we explain our approach and some key parts of the implementation. We also present a case study based on the U.S. electricity production and an experiment on the scalability of our implementation.
Journal of Experimental & Theoretical Artificial Intelligence, 2005
... Publication Hors Liris. Topological Representation Model for Image Database Query. Mihaela Sc... more ... Publication Hors Liris. Topological Representation Model for Image Database Query. Mihaela Scuturici[LIRIS] , J Clech , Vasile-Marian Scuturici[LIRIS] , Djamel Zighed. 1/2005. Journal of Experimental & Theoretical Artificial Intelligence, Vol. 17, No. 1-2, January-June 2005. ...