Gabor space and the development of preattentive similarity

“…The attention mechanism is driven by the visual saliency of the scene elements, which refers to the idea that certain parts of a scene are distinctive and that they create some form of significant visual arousal at the early visual stages [16]. This mechanism is essentially data-driven, which is particularly useful in those situations where the semantics of the contents of the image is not known and models of the perceived objects are not available [30].Light intensity contrast appears to be the primary variable on which humans base visual saliency computation, although other features participate in defining visual saliency at higher processing levels in the visual cortex [12,25]. Among these are edge or line orientation, color, motion, and stereo.…”

mentioning

confidence: 99%

Outdoor View Recognition Based on Landmark Grouping and Logistic Regression

Todt

Torras

2013

Int. J. Patt. Recogn. Artif. Intell.

View full text Add to dashboard Cite

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.Keywords: visual landmarks, visual saliency, robot navigation, autonomous robot IntroductionThe extraction of reliable visual landmarks for mobile robot localization in unknown outdoor unstructured environments is still an open research problem. One of the key factors that makes the detection and recognition of visual landmarks in outdoor environments, as well as indoors without dominant artificial illumination, a challenging task is that acquired visual information is strongly dependent on lighting geometry (direction and intensity of light source) and illuminant color (spectral power distribution), which change with sun position and atmospheric conditions.Most feature extraction approaches are not adequate for this type of environments, since they rely on either structured information from non-deformable objects [7], or on a priori knowledge about the landmarks [3]. Several recent works achieved interesting results using SIFT features to match pairs of images [18,22,32], which can be extended to the landmark recognition problem. Since mobile robot navigation tasks require real-time execution, some efforts have been made to reduce the considerable computational cost necessary to evaluate SIFT features for a whole image [17,28]. Also it was reported that SIFT features fail to consider global context to resolve ambiguities that can occur locally in images, motivating solutions that improve the amount of global information used in the descriptors [10,24] A suitable framework to compare views is developed, and it is shown how this remarkably enhances the recognition performance.The remainder of the paper is organized as follows. Section 2 presents the concept of landmark and visual saliency. In this context, landmarks are the visual elements that are used to provide the features for the recognition of places, and they are found or defined based on classical visual saliency criteria, inspired in a biological model of visual opponency. Section 3 describes how to ref...

show abstract

“…Entre as técnicas para extração de características de textura estão os filtros de Gabor [Santini e Jain, 1996] e as transformadas de wavelets [Daubechies, 1990, Santini e Gupta, 2001, Arivazhagan e Ganesan, 2003]. Esses métodos tentam capturar partes da imagem com relaçãoà mudança de direção e escala, e são muitoúteis para imagens ou regiões com texturas homogêneas.…”

Section: Texturaunclassified

“…Em [Schilham et al, 2006]é utilizada a segmentação de imagens de raios-x de tórax para detecção de nódulos pulmonares para auxílio ao diagnóstico de câncer de pulmão e em [Dy et al, 2003] características de cor, textura e forma são combinadas em um mesmo vetor para serem usadas na recuperação de imagens de tomografia computadorizada de tórax. Em [Antani et al, 2006] são utilizados imagens de raios-x da coluna vertebral para auxílio ao diagnóstico de degeneração de vértebras.…”

Section: Métodos Específicos Para Imagens De Exames Médicosunclassified

Adequando consultas por similaridade para reduzir a descontinuidade semântica na recuperação de imagens por conteúdo

Razente¹

View full text Add to dashboard Cite

A Deus por estar sempre comigo. Aos meus pais, por terem me incentivado no caminho das ciências e por sempre terem feito tudo o que estava ao alcance para que eu pudesse chegar até esta tese. A minha querida esposa Camila, por estar sempre ao meu lado, pela sua dedicação e apoio durante todos esses anos, e pela grande contribuição nos vários trabalhos desenvolvidos durante a realização desta tese. Ao meu orientador, professor Caetano, pela orientação durante o mestrado e o doutorado, agradeço seu apoio, incentivo e confiança, seus conselhos foram essenciais para a minha formação como pesquisador. A professora Agma, pela co-orientação e pelos conselhos preciosos. A professora Franklina M. B. Toledo pelo auxílio com os métodos de programação linear e não-linear e com os métodos de avaliação de problemas de otimização. Aos professores Josiane M. Bueno (in memorian), Eduardo R. Hruschka e Cristina D. A. Ciferri pela orientação no Programa de Aperfeicoamento de Ensino. A professora Cláudia A. Martins (UFMT), pela amizade, incentivo e pelas valiosas cartas de recomendação. Aos professores Mauro Biajiz e Ricardo Ciferri (UFSCAR) pelas dicas nas reuniões do grupo de pesquisa. Ao professor Christos Faloutsos (Carnegie Mellon) pelas sugestões. Ao professor Vassilis Tsotras (University of California em Riverside) por me receber em seu grupo de pesquisa e me orientar durante meu estágio sanduíche. Ao amigo Marcos R. Vieira pela grande ajuda durante minha estadia em Riverside. Aos amigos que me receberam em seus lares (mais conhecidos como repúblicas), Caio e Guilherme

show abstract

Gabor space and the development of preattentive similarity

Cited by 13 publications

References 6 publications

Detecting salient cues through illumination-invariant color ratios

Detecting salient cues through illumination-invariant color ratios

Outdoor View Recognition Based on Landmark Grouping and Logistic Regression

Adequando consultas por similaridade para reduzir a descontinuidade semântica na recuperação de imagens por conteúdo

Contact Info

Product

Resources

About