Ángel Sánchez

SUPPORT VECTOR MACHINES VERSUS MULTI-LAYER PERCEPTRONS FOR EFFICIENT OFF-LINE SIGNATURE RECOGNITION

Abstract The problem of automatic signature recognition has received little attention in comparison with the problem of signature verification despite its potential applications for accessing security-sensitive facilities and for processing certain legal and historical documents. This paper presents an efficient off-line human signature recognition system based on support vector machines (SVM) and compares its performance with a traditional classification technique, multi-layer perceptrons (MLP). In both cases we propose two approaches to the problem: (1) construct each feature vector using a set of global geometric and moment-based characteristics from each signature and (2) construct the feature vector using the bitmap of the corresponding signature. We also present a mechanism to capture the intrapersonal variability of each user using just one original signature. Our results empirically show that SVM, which achieves up to 71% correct recognition rate, outperforms MLP.

Breast thermography from an image processing viewpoint: A survey

Abstract Breast cancer is the leading cause of death among women. This fact justifies researches to reach early diagnosis, improving patients’ life expectancy. Moreover, there are other pathologies, such as cysts and benign neoplasms that deserve investigation. In the last ten years, the infrared thermography has shown to be a promising technique to early diagnosis of breast pathologies. Works on this subject presented results that justify the thermography as a complementary exam to detect breast diseases. Several papers on the use of infrared imaging for breast screening can be found in the current medical literature. This survey explores and analyses these works in the light of their applications in computer vision. Consequently, the comments are organized according to the main steps of pattern recognition systems. These include: image acquisition protocols, exams storage, segmentation methods, feature extraction, classification or diagnostic and computer modelling. Main contributions of discussed papers are summarized in tables to provide a structured vision of the aspects involved in breast thermography.

Improving image segmentation quality through effective region merging using a hierarchical social metaheuristic

This paper proposes a new evolutionary region merging method in order to efficiently improve segmentation quality results. Our approach starts from an oversegmented image, which is obtained by applying a standard morphological watershed transformation on the original image. Next, each resulting region is represented by its centroid. The oversegmented image is described by a simplified undirected weighted graph, where each node represents one region and weighted edges measure the dissimilarity between pairs of regions (adjacent and non-adjacent) according to their intensities, spatial locations and original sizes. Finally, the resulting graph is iteratively partitioned in a hierarchical fashion into two subgraphs, corresponding to the two most significant components of the actual image, until a termination condition is met. This graph-partitioning task is solved by a variant of the min-cut problem (normalized cut) using a hierarchical social (HS) metaheuristic. We have efficiently applied the proposed approach to brightness segmentation on different standard test images, with good visual and objective segmentation quality results.

Differential optical flow applied to automatic facial expression recognition

This work compares systematically two optical flow-based facial expression recognition methods. The first one is featural and selects a reduced set of highly discriminant facial points while the second one is holistic and uses much more points that are uniformly distributed on the central face region. Both approaches are referred as feature point tracking and holistic face dense flow tracking, respectively. They compute the displacements of different sets of points along the sequence of frames describing each facial expression (i.e. from neutral to apex). First, we evaluate our algorithms on the Cohn-Kanade database for the six prototypic expressions under two different spatial frame resolutions (original and 40%-reduced). Later, our methods were also tested on the MMI database which presents higher variabilities than the Cohn-Kanade one. The results on the first database show that dense flow tracking method at original resolution slightly outperformed, in average, the recognition rates of feature point tracking method (95.45% against 92.42%) but it requires 68.24% more time to track the points. For the patterns of MMI database, using dense flow tracking at the original resolution, we achieved very similar average success rates.

Automatic Discrimination between Printed and Handwritten Text in Documents

Recognition techniques for printed and handwritten text in scanned documents are significantly different. In this paper we address the problem of identifying each type. We can list at least four steps: digitalization, preprocessing, feature extraction and decision or classification. A new aspect of our approach is the use of data mining techniques on the decision step. A new set of features extracted of each word is proposed as well. Classification rules are mining and used to discern printed text from handwritten. The proposed system was tested in two public image databases. All possible measures of efficiency were computed achieving on every occasion quantities above 80%.

Robust off-line signature verification using compression networks and positional cuttings

A novel robust technique for the off-line signature verification problem in practical real conditions is presented. The technique is based on the use of compression neural networks, and in the automatic generation of the training set from only one signature for each writer. Our proposal incorporates a new kind of acceptance/rejection rule, which is based on the similarity between subimages or positional cuttings of a test signature and the corresponding representation stored in the class compression network. Experimental results show that the proposed technique reduces significantly the false acceptation rate (FAR).

Particle filter on GPUs for real-time tracking

Efficient object tracking is required by many Computer Vision application areas like surveillance or robotics. It deals with statespace variables estimation of interesting features in image sequences and their future prediction. Probabilistic algorithms has been widely applied to tracking. These methods take advantage of knowledge about previous states of the system reducing the computational cost of an exhaustive search over the whole image. In this framework, posterior probability density function (pdf) of the state is estimated in two stages: prediction and update. General particle filters are based on discrete representations of probability densities and can be applied to any state-space model [Arulampalam et al. 2002]. Discrete particles j of a set (Xt ,Πt) = {(x0 t ,π0 t )...(xN t ,πN t )} in time step t, contains information about one possible state of the system x j t and its importance weight π j t . In a practical approach, particle weights computation is the most expensive stage of the particle filter algorithm, and it has to be executed at each time step for every particle [Deutscher et al. 2000].

Automatic linguistic report of traffic evolution in roads

In the field of intelligent transportation systems, one important challenge consists of maintaining updated the electronic panels installed in roads with relevant information expressed in natural language. Currently, these messages are produced by human experts. However, the amount of data to analyze in real time and the number of available experts are imbalanced and new computational tools are required to assist them in this work. Moreover, the same problem appears when we deal with automatically generating linguistic reports to assist traffic managers that must take their decisions based on large amounts of quickly evolving information. In this paper, we contribute to solve this problem by designing a computational application based on our research in the field of computational theory of perceptions. Here, we present an application where we generate linguistic descriptions of the traffic behavior evolving in time and changing between different levels of service. We include some results obtained with both, simulated and real data.

A low-level hybridization between memetic algorithm and VNS for the max-cut problem

The Max-Cut problem consists of finding a partition of the graph nodes into two subsets, such that the sum of the edge weights having endpoints in different subsets is maximized. This NP-hard problem for non planar graphs has different applications in areas such as VLSI and ASIC design. This paper proposes an evolutionary hybrid algorithm based on low-level hybridization between Memetic Algorithms and Variable Neighborhood Search. This algorithm is tested and compared with the results, found in the bibliography, obtained by other hybrid metaheuristics for the same problem. Achieved experimental results show the suitability of the approach, and that the proposed hybrid evolutionary algorithm finds near-optimal solutions. Moreover, on a set of standard test problems, new best known solutions were produced for several instances.

Multiple and variable target visual tracking for video-surveillance applications

Visual detection and tracking are interdisciplinary tasks which are oriented at estimating the state of one or multiple moving objects in a video sequence. This is one of the first tasks in processing video systems which try to describe human behaviour in different contexts, such as video-surveillance, sport technique analysis. This work presents a multiple object tracking system which properly hybridizes particle filters and memetic algorithms to produce a more reliable and efficient tracking algorithm. The system has been tested on synthetic and real image sequences, with the aim of describing their performance for different levels of noise, occlusions, a variable number of objects, etc. Experimental results demonstrate that the proposed system accurately tracks multiple objects in the scene, by grouping and ungrouping them when necessary, while keeping their identities during the sequence of images. Moreover, the performance of the proposed system is not strongly affected by the increase in the number of objects, maintaining computational load and precision in proper balance.