Frequency and orientation sensitive texture measures using linear symmetry
نویسنده
چکیده
An efficient method for computing texture features based on dominant local orientation is introduced. The features are computed as a Laplacian pyramid is built. At each level of the Laplacian pyramid, the linear symmetry feature is computed. This feature is anisotropic and estimates the optimal local orientation in the Least Square Error (LSE) sense. It is complex valued and hence consists of two components, the local orientation estimate and its confidence measure based on the error. The algorithm is based on convolutions with simple separable filters and pixel-wise non-linear arithmetic operations. These properties allow highly parallel implementation, for example on a pyramid machine, yielding real time applications. Comparative experimental results are presented using the feature for unsupervised segmentation on test images of natural aerial image textures. Zusammenfassung. Es wird eine effiziente Methode zur Berechnung von Textur-Merkmalen eingefiihrt, die auf einer lokalen Orientierung basiert. Die Merkmale werden entsprechend dem Bau einer Laplace-Pyramide berechnet. Auf jeder Stufe der Laplace-Pyramide werden die Iinearen Symmetrie-Merkmale ermittelt. Solche Merkmale sind anisotrop und schiitzen die optimale lokale Orientierung im Sinne des kleinsten quadratischen Fehlers (LSE). Sie sind komplexwertig und bestehen daher aus zwei Komponenten, dem Schiitzwert der lokalen Orientierung und seinem auf dem Pehler basierenden Vertrauensmafl. Der Algorithm us basiert auf Faltungen mit Hilfe separierbarer Filter und pixelweiser nichlinearer arithmetischer Operiltionen. Diese Eigenschaften erlauben ein hohes Mafl paralleler Implementierung, zum Beispiel auf einer Pyramide-Maschine, und eroffnen die Moglichkeit der Echtzeit-Verarbeitung. Vergleichende experimentelle Resultate werden wiedergegeben, die von den Eigenschaften der 'unsupervised' Segmentierung von Testbildern und natiirlichen Texturen von Luftaufnahmen Gebrauch machen. Resume. Une methode efficace de calcul d'attributs de texture basee sur !'orientation locale dominante est introduite. Les attributs sont calcules pendant q'une pyramide laplacienne est construite. A chaque niveau de la pyramide laplacienne, I'attribut de symetrie lineaire est calcule. Cet attribut est anisotrope et estime !'orientation locale optimale au sens de la plus petite erreur quadratique (LSE). C'est une valeur complexe qui comprend done deux composantes, l'estimee d'orientation locale et sa mesure de fiabilite basee sur l'erreur. L'algorithme est base sur des convolutions avec des filtres simples separables et sur des operations arithmetiques non-lineaires au niveau des pixels. Ces proprietes permettent une implantation hautement parallele, par exemple sur une machine de type pyramidal, engendrant des applications temps reel. Des resultats experimentaux comparatifs sont presentes sur des attributs utilises pour Ia segmentation non supervisee d'images de test realisees a partir d'images aeriennes texturees.
منابع مشابه
Planar surface orientation from texture spatial frequencies
-This paper presents a computational model and a practical algorithm for determining the three-dimensional orientation of a planar surface from visual texture information. The model consists of three parts: (1) a local spatial frequency based texture representation; (2) a model describing the projection of surface texture to image texture; and (3) a solution of the texture projection model for ...
متن کاملMeasuring the spatial frequency selectivity of second-order texture mechanisms
Recent investigations of texture and motion perception suggest two early filtering stages: an initial stage of selective linear filtering followed by rectification and a second stage of linear filtering. Here we demonstrate that there are differently scaled second-stage filters, and we measure their contrast modulation sensitivity as a function of spatial frequency. Our stimuli are Gabor modula...
متن کاملPattern Orientation and Texture Symmetry1
Human texture perception relies on a few basic high-level features including directionality and symmetry. Today, research on oriented patterns finds its applications in various areas of applied machine vision. In this study, we present and investigate a new method for assessing pattern anisotropy via texture symmetry. Pattern orientation is viewed as the direction of persistent statistical text...
متن کاملPrioritization of Erosion Prone Sub-Watersheds using MCDM Methods in Roudzard Watershed, Khuzestan Province
Soil erosion has been one of the most important problems of watersheds in the world and is considered one of the main obstacles to achieving sustainable development in agriculture and natural resources. Identifying and prioritizing regions sensitive to soil erosion is essential for water and soil conservation and natural resource management in watersheds. The present research was performed in 2...
متن کاملLand Cover Change Detection Using Texture Analysis
Problem statement: It is an important task to detect land cover changes from remotely sensed data for environmental monitoring. Although there are some applications of visual textures to the land use, they are limited to a few land cover categories with the application of one texture measure. Since land cover types are complex and often the integration of various objects, applying one texture m...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Signal Processing
دوره 29 شماره
صفحات -
تاریخ انتشار 1992