Todt, E.; Torras, C. International journal of pattern recognition and artificial intelligence Vol. 27, num. 3, p. 1355004-1-1355004-20 DOI: 10.1142/S0218001413550045 Data de publicació: 2013 Article en revista
Vision-based robot localization outdoors has remained more elusive than its indoors counterpart. Drastic illumination changes and the scarceness of suitable landmarks are the main difficulties. This paper attempts to surmount them by deviating from the main trend of using local features. Instead, a global descriptor called landmark-view is defined, which aggregates the most visually-salient landmarks present in each scene. Thus, landmark co-occurrence and spatial and saliency relationships between them are added to the single landmark characterization, based on saliency and color distribution. A suitable framework to compare landmark-views is developed, and it is shown how this remarkably enhances the recognition performance, compared against single landmark recognition. A view-matching model is constructed using logistic regression. Experimentation using 45 views, acquired outdoors, containing 273 landmarks, yielded good recognition results. The overall percentage of correct view classification obtained was 80.6%, indicating the adequacy of the approach.
Solé, A.; Serratosa, F.; Sanfeliu, A. International journal of pattern recognition and artificial intelligence Vol. 26, num. 5, p. 1-21 DOI: 10.1142/S021800141260004X Data de publicació: 2012-08 Article en revista
We model the edit distance as a function in a labeling space. A labeling space is an Euclidean space where coordinates are the edit costs. Through this model, we de¯ne a class of cost. A class of cost is a region in the labeling space that all the edit costs have the same optimal labeling.
Moreover, we characterize the distance value through the labeling space. This new point of view of the edit distance gives us the opportunity of de¯ning some interesting properties that are useful for a better understanding of the edit distance. Finally, we show the usefulness of these properties through some applications.
This paper presents an efficient IrisCode classifier, built from phase features which uses AdaBoost for the selection of Gabor wavelets bandwidths. The final iris classifier consists of a weighted contribution of weak classifiers. As weak classifiers we use three-split decision trees that identify a candidate based on the Levenshtein distance between phase vectors of the respective iris images. Our experiments show that the Levenshtein distance has better discrimination in comparing IrisCodes than the Hamming distance. Our process also differs from existing methods because the wavelengths of the Gabor filters used, and their final weights in the decision function, are chosen from the robust final classifier, instead of being fixed and/or limited by the programmer, thus yielding higher iris recognition rates. A pyramidal strategy for cascading filters with increasing complexity makes the system suitable for real-time operation. We have designed a processor array to accelerate the computation of the Levenshtein distance. The processing elements are simple basic cells, interconnected by relatively short paths, which makes it suitable for a VLSI implementation.
Serratosa, F.; Sanfeliu, A.; Alquezar, R. International journal of pattern recognition and artificial intelligence Vol. 18, num. 3, p. 375-396 DOI: 10.1142/S0218001404003253 Data de publicació: 2004-05 Article en revista
Andrade-Cetto, J.; Sanfeliu, A. International journal of pattern recognition and artificial intelligence Vol. 16, num. 3, p. 361-374 DOI: 10.1142/S0218001402001745 Data de publicació: 2002-05 Article en revista
Cardoner, R.; Thomas, F. International journal of pattern recognition and artificial intelligence Vol. 11, num. 6, p. 947-960 DOI: 10.1142/S0218001497000433 Data de publicació: 1997-09 Article en revista
Image compression techniques have been recently used not only for reducing storage requirements, but also computational costs when processing images on low cost computers. This approach might be also of interest for processing large engineering drawings, where feature extraction techniques must be intensively applied for their segmentation into regions of interest for subsequent analysis. This paper explores this alternative using a simple run-length compression, leading to excellent results.
Although this approach is not new and can be classified within the decomposition paradigm used since the early stages of line drawing image processing, the developed formalism allows directional morphological set transformations to be performed, on a low cost personal computer, faster than on costly parallel computers for the same, but uncompressed, images. This good performance is proved in two different applications: the generation of homotopic skeletons through thinning processes, and the extraction of linear features through serializing multiangle parallelism operations.