Juan Moreno-Garcia

Real-time moving object segmentation in H.264 compressed domain based on approximate reasoning

This paper presents a real-time segmentation algorithm to obtain moving objects from the H.264 compressed domain. The proposed segmentation works with very little information and is based on two features of the H.264 compressed video: motion vectors associated to the macroblocks and decision modes. The algorithm uses fuzzy logic and allows to describe position, velocity and size of the detected regions in a comprehensive way, so the proposed approach works with low level information but manages highly comprehensive linguistic concepts. The performance of the algorithm is improved using dynamic design of fuzzy sets that avoids merge and split problems. Experimental results for several traffic scenes demonstrate the real-time performance and the encouraging results in diverse situations.

Fuzzy numbers from raw discrete data using linear regression

This paper focuses on modelling fuzzy numbers with meaningful membership functions. More precisely, it proposes a method to construct trapezoidal fuzzy number approximations from raw discrete data. In many applications, input information is numerical, and therefore, particular fuzzy sets, such as fuzzy numbers, hold great interest and relevance in managing data imprecision and vagueness. The proposed technique provides an efficient way to obtain trapezoidal numbers using linear regression. The technique is simple, fast, and effective. Preliminary tests are performed using different types of input data: a Gaussian function, a Sigmoidal function, three datasets of synthetic discrete data, and an histogram obtained from a colour satellite image.

A direct linguistic induction method for systems

The aim of this paper is to present a method of obtaining a linguistic model that reflects the behavior of a combined well-known data. The method is based on the technique of successive division of the input space as CART (Classification and Regression Tree, Wadsworth, Monterey, CA, 1984) and ID3 (Mach. Learning 1 (1986) 81). The obtained rules have linguistic variables such as antecedent and consequent. The methods ID3 and CART are generalized to work directly with linguistic variables defined a priori, incorporating the concept of linguistic interval in order to represent the disjunctions of the linguistic labels.

Automatic objects behaviour recognition from compressed video domain

In this paper we present a system that, directly from compressed video domain, establishes a correspondence between objects in motion in a video scene and a concrete behaviour. This behaviour is expressed by using linguistic variables. Besides, with this fuzzy logic-based approach, the imprecision and vagueness of our primary source of information, MPEG motion vectors, is reduced. Proposed algorithms for segmentation and tracking are based on fuzzification of MPEG motion data. Once the tracking phase has finished, a linguistic model for each objective in the scene is generated and compared with each one of the behaviour models previously described in a linguistic manner. Finally, a practical application of this system for detection, tracking and behaviour analysis of vehicles in complex traffic scenes is presented.

A Fuzzy Inductive Algorithm for Modeling Dynamical Systems in a Comprehensible Way

In this paper, we propose the use of temporal fuzzy chains for the modeling of dynamical systems in a way that is comprehensible. We are interested in helping the overall understanding of the system execution, over and during a precise and finite time. To this end, we model its input/output behavior and how this has changed in the past. There is a double goal in mind: accuracy and interpretability. An inductive algorithm for analyzing finite continuous multivariate time series will be achieved, in which the use of fuzzy logic has been taken into account. The aim of the algorithm is to help us to find changes in a system, as well as to identify the causes of these changes in a linguistic form. The causes will be specified by means of a set of fuzzy transitions between consecutive states, which consist of fuzzy rules that model the system. The method suggested has been applied on a real life case, human walk modeling.

Approximate reasoning and finite state machines to the detection of actions in video sequences

In this paper a novel approach for recognizing actions in video sequences is presented, where the information obtained from the segmentation and tracking algorithms is used as input data. First of all, the fuzzification of input data is done and this process allows to successfully manage the uncertainty inherent to the information obtained from low-level and medium-level vision tasks, to unify the information obtained from different vision algorithms into a homogeneous representation and to aggregate the characteristics of the analyzed scenario and the objects in motion. Another contribution is the novelty of representing actions by means of an automaton and the generation of input symbols for the finite automaton depending on the comparison process between objects and actions, i.e., the main reasoning process is based on the operation of automata with capability to manage fuzzy representations of all video data. The experiments on several real traffic video sequences demonstrate encouraging results, especially when no training algorithms to obtain predefined actions to be identified are required.

A Fuzzy Logic Based Approach to Improve Cataloguing and Searching in e-Commerce Portals

The business achievement among consumers via e-commerce is getting more and more importance at the present time. In this paper, we propose to make use of fuzzy logic with the aim to improve the search and cataloguing of goods and services in Consumer-to-Consumer electronic commerce (E-commerce) portals (e.g. ebay). These portals are the media through most of the electronic transactions among consumers are conducted today. We suggest a method that tries to adapt to users’ real needs. It allows buyers to carry out searches in an imprecise way and sellers to deal with catalogues of items (goods or services) described also with a lack of precision.

Lane mark segmentation and identification using statistical criteria on compressed video

The detection and localization of road lane marks are relevant to many applications of driving assistance and road traffic surveillance. Usually, these techniques work by processing all the pixels in every image, making the computational cost too high. In these situations, the implementation of real-time detection applications is impossible. Processing the video directly in the compressed domain avoids this limitation because the data rate is much reduced and full decoding of the compressed images is unnecessary. The development of a real-time detection systems then becomes possible, even for resource-limited systems like mobile devices. In this paper an approach to the segmentation and recognition of lane marks using only H264/AVC motion vectors is proposed. A new representation of motion vectors is defined in order to detect efficiently the regions or blobs of interest in complex videos captured by moving cameras. Then, a set of mathematical filters are applied removing progressively the blobs detected, depending on their position in the scene, their size, and their shape; and obtaining finally the regions corresponding to the lane marks. The proposed method shows encouraging results in different road traffic video sequences.

From Training to Organizational Behavior: A Mediation Model through Absorptive and Innovative Capacities

The training of human resources improves business performance: myth or reality? While the literature has extensively addressed this issue, the transfer that occurs from training to performance still remains unresolved. The present study suggests an empirical solution to this gap, through a multiple mediation model of dynamic capabilities. Accordingly, the study makes a major contribution to the effectiveness of an organizational-level training: the “true” relationship between training and performance is mediated by absorptive and innovative capacities. It is difficult from training to directly affect the results: it must be done through a chain of intermediate variables. Training can be argued to be indirectly related to performance, through absorptive capacity in the first place, and innovative capacity in the second, sequentially in this order (three-path mediated effect). Of all immediate relationships received by performance, its explained variance is achieved partly via absorptive capacity and partly via innovation. The direct relationship through training is not significant and only explains a small percentage of the variance in performance. These results have been corroborated by combining two methods of analysis: PLS-SEM and fsQCA, using data from an online survey. This dual methodology in the study of the same phenomenon allows overcoming the limitations of each method, which would not have been possible with a single methodological approach, and confirming the findings obtained by any of them.

Description of multivariate time series by means of trends characterization in the fuzzy domain

This paper presents a new method to automatically obtain linguistic descriptions from a multivariate time series. The first step consists of the induction from the multivariate time series of a fuzzy model called the temporal fuzzy model. Later this fuzzy model is analyzed in order to obtain trends based on the output variable. With this information, the trends for the input variables are also obtained and together with the output variable trends are represented in such a way that the temporal evolution of the trend is stored within the own-trends. In a trend there are always some elements or points of interest that are more relevant than others for generating a relevant final description. These elements are also extracted and stored, making the final description generation process more efficient. An event search process generates the final linguistic description. The events are identified in the trends or in the structure containing the points of interest previously selected. Once an event is identified, new text is added at the end of the linguistic description. The design of the events and the text related to each event is made with the cooperation of experts in the field of application. The presented approach is checked in sports, more concretely, in countermovement jumping.