Local Smoothness in terms of Variance: the Adaptive GaussianFilter

Gomez, G.

doi:10.5244/c.14.82

Cited by 22 publications

(16 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The occurrence of this effect however is diminishable, in the case of local Gaussian filtering [48]. In our approach, memristive dynamics are employed for achieving this performance intrinsically.…”

Section: Memristive Components the Recent Nanoscale Implementation Omentioning

confidence: 99%

Biomimetic model of the outer plexiform layer by incorporating memristive devices

et al. 2012

View full text Add to dashboard Cite

In this paper we present a biorealistic model for the first part of the early vision processing by incorporating memristive nanodevices. The architecture of the proposed network is based on the organisation and functioning of the outer plexiform layer (OPL) in the vertebrate retina. We demonstrate that memristive devices are indeed a valuable building block for neuromorphic architectures, as their highly non-linear and adaptive response could be exploited for establishing ultra-dense networks with similar dynamics to their biological counterparts. We particularly show that hexagonal memristive grids can be employed for faithfully emulating the smoothing-effect occurring at the OPL for enhancing the dynamic range of the system. In addition, we employ a memristor-based thresholding scheme for detecting the edges of grayscale images, while the proposed system is also evaluated for its adaptation and fault tolerance capacity against different light or noise conditions as well as distinct device yields.Over the past years, the performance and efficiency of biological systems has inspired many researchers and engineers, giving birth to the emerging fields of biomimetics [1] and bioinspiration [2]. The human retina is anything but a simple passive relay station, as it preprocesses and compresses all sensed information through an immensely complex neuronal network that on average contains 4.6 million cones, 92 million rods [3] and 1 million ganglion [4]. If we take into account the remaining bipolar, horizontal and amacrine cells and the fact that they are highly interconnected we get a parallel network of grand complexity. This complexity is further elucidated in recent studies where evidence is provided that at least ten parallel signals arise from a single visual point [5] and we get a much more accurate view from the functioning of the vertebrate retina. In fact it is nowadays believed that although biological systems are

show abstract

Section: Memristive Components the Recent Nanoscale Implementation Omentioning

confidence: 99%

Biomimetic model of the outer plexiform layer by incorporating memristive devices

et al. 2012

View full text Add to dashboard Cite

show abstract

“…Using a twofold similarity measure like the geodesic time, defined on either the spatial gradient or the image variation, enables to account for these correlations. A smoothing technique described in (Gomez, 2000) is based on adaptive Gaussian filters. It adjusts locally the smoothing scale (i.e.…”

Section: Experiments and Comparison With Other Techniquesmentioning

confidence: 99%

“…Nonlinear smoothing has been developed to overcome these shortcomings, which tend to preserve important features along with noise removal during smoothing (Narendra, 1981;Brownrigg, 1984;Perona and Malik, 1990;Nitzberg and Shiota, 1992). In this context, the need for an adaptive approach to cope with inhomogeneities in images is well recognised (Saint-Marc and Medioni, 1991;Gomez, 2000;Grazzini et al, 2004). Strong relations have been established between a number of widely-used adaptive filters for digital image processing (Mrázek et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

Edge-Preserving Smoothing of Natural Images Based on Geodesic Time Functions

2008

Proceedings of the Third International Conference on Computer Vision Theory and Applications

View full text Add to dashboard Cite

In this paper, we address the problem of edge-preserving smoothing of natural images. We introduce a novel adaptive approach derived from mathematical morphology as a preprocessing stage in feature extraction and/or image segmentation. Like other filtering methods, it assumes that the local neighbourhood of a pixel contains the essential information required for the estimation of local image properties. It performs a weighted averaging by combining both spatial and tonal information in a single similarity measure based on the local calculation of geodesic time functions from the estimated pixel. By designing relevant geodesic masks, it can deal with specific situation and type of images. We describe in the following two possible strategies and we show their capabilities at smoothing heterogeneous areas while preserving relevant structures in naturalgreyscale or multispectral-images displaying different features.

show abstract

“…Gomez [12] gave an adaptive Gaussian filter in which local variance σ for each kernel is selected automatically by working with a scale-space framework and minimal description length criterion (MDL). The filtering procedure can be understood as [12]:…”

Section: Adaptive Gaussian Filter For Local Smoothnessmentioning

confidence: 99%

“…where I σ (x, y) is the Low-Pass part, and ε σ (x, y) is the residual part. Rewrite Equation 4as description length (dl), [12] has following equation:…”

Section: Adaptive Gaussian Filter For Local Smoothnessmentioning

confidence: 99%

Mesh Smoothing via Adaptive Bilateral Filtering

Hou

Bai

Wang

2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

In this paper, we present an adaptive bilateral filtering algorithm that can be used to remove unavoidable noise from 3D mesh data generated by initial stages. Selecting the parameters for bilateral filters automatically, this algorithm smoothes meshes in the normal field using anisotropic character of local neighborhood triangles. Experimental results demonstrate that the proposed method remove light noise from meshes and reserve fine features of meshes as good as best results of other methods, with the advantage of none user-assisted parameters setting. Visual comparisons display that the method proposed in this paper performs better than other smoothing method for heavy noisy mesh.

show abstract

Local Smoothness in terms of Variance: the Adaptive GaussianFilter

Cited by 22 publications

References 13 publications

Biomimetic model of the outer plexiform layer by incorporating memristive devices

Biomimetic model of the outer plexiform layer by incorporating memristive devices

Edge-Preserving Smoothing of Natural Images Based on Geodesic Time Functions

Mesh Smoothing via Adaptive Bilateral Filtering

Contact Info

Product

Resources

About