Pei-hsiu Suen

The analysis and recognition of real-world textures in three dimensions

The observed image texture for a rough surface has a complex dependence on the illumination and viewing angles due to effects such as foreshortening, local shading, interreflections, and the shadowing and occlusion of surface elements. We introduce the dimensionality surface as a representation for the visual complexity of a material sample. The dimensionality surface defines the number of basis textures that are required to represent the observed textures for a sample as a function of ranges of illumination and viewing angles. Basis textures are represented using multiband correlation functions that consider both within and between color band correlations. We examine properties of the dimensionality surface for real materials using the Columbia Utrecht Reflectance and Texture (CUReT) database. The analysis shows that the dependence of the dimensionality surface on ranges of illumination and viewing angles is approximately linear with a slope that depends on the complexity of the sample. We extend the analysis to consider the problem of recognizing rough surfaces in color images obtained under unknown illumination and viewing geometry. We show, using a set of 12,505 images from 61 material samples, that the information captured by the multiband correlation model allows surfaces to be recognized over a wide range of conditions. We also show that the use of color information provides significant advantages for three-dimensional texture recognition.

Modeling and classifying color textures using random fields in a random environment

Abstract We present a random environment model for color textures. This model generalizes traditional random field models by allowing the spatial interaction parameters of the field to be random variables. We use this new model to define a compact color feature vector which captures within and between color band information. A set of color textures is used to show that this feature vector contains most of the information in the model. We show experimentally that the new model improves on the performance of multiband Markov fields for texture classification using small samples.

Analyzing the bidirectional texture function

The observed image texture for a rough surface has a complex dependence on the illumination and viewing angles due to effects such as local shading, interreflections, and the shadowing and occlusion of surface elements. We introduce the dimensionality surface as a representation for the visual complexity of a material sample. The dimensionality surface defines the number of basis features that are required to represent the space of observed textures for a surface as a function of ranges of illumination and viewing angles. Basis textures are represented using multiband correlation functions. We study properties of the dimensionality surface for real materials using the Columbia Utrecht Reflectance and Texture (CUReT) database. The analysis shows that the dependence of the dimensionality surface on ranges of illumination and viewing angles is approximately linear with a slope dependent on the complexity of the sample.

The impact of viewing geometry on material discriminability in hyperspectral images

An increase in the off-nadir viewing angle for an airborne visible/near-infrared through short-wave infrared (VNIR/SWIR) imaging spectrometer leads to a decrease in upward atmospheric transmittance and an increase in line-of-sight scattered path radiance. These effects combine to reduce the spectral contrast between different materials in the sensed signal. The authors analyze the impact of viewing angle on material discriminability for 237 materials over a wide range of conditions. Material discriminability is quantified using a statistical algorithm that employs a subspace model to represent the set of spectra for a material as conditions vary. The authors show that reliable material discrimination is possible over a range of conditions even for large off-nadir viewing angles. They illustrate the performance of material identification over different viewing angles using simulated forest and desert hyperspectral digital imagery collection experiment (HYDICE) images.

Invariant mixture recognition in hyperspectral images

We present an algorithm for identifying linear mixtures of a specified set of materials in 0.4-2.5 /spl mu/m airborne imaging spectrometer data. The algorithm is invariant to the illumination and atmospheric conditions and the relative amounts of the specified materials within a pixel. Only the spectral reflectance functions for the specified materials are required by the algorithm. Invariance over illumination and atmosphere conditions is achieved by incorporating a physical model for scene variability in the constrained optimization formulation. The algorithm also computes estimates of the amounts of the specified materials in identified mixtures. We demonstrate the effectiveness of the algorithm using real and synthetic HYDICE imagery acquired over a range of conditions and altitudes.

The impact of viewing geometry on vision through the atmosphere

An increase in the off-nadir viewing angle for an airborne visible/near-infrared through short-wave infrared (VNIR/SWIR) imaging spectrometer leads to a decrease in upward atmospheric transmittance and an increase in line-of-sight scattered path radiance. These effects combine to reduce the spectral contrast between different materials in the sensed signal. We analyze the impact of viewing angle on material discriminability for 237 materials over a wide range of conditions. Material discriminability is quantified using a statistical algorithm that employs a subspace model to represent the set of spectra for a material as conditions vary. We show that reliable material discrimination is possible over a range of conditions even for large off-nadir viewing angles. We illustrate the performance of material identification over different viewing angles using simulated forest hyperspectral images.

Invariant identification of material mixtures in airborne spectrometer data

We present an algorithm for identifying linear mixtures of a specified set of materials in 0.4-2.5 microm airborne imaging spectrometer data. The algorithm is invariant to the illumination and atmospheric conditions and the relative amounts of the specified materials within a pixel. Only the spectral reflectance functions for the specified materials are required by the algorithm. Invariance over illumination and atmospheric conditions is achieved by incorporating a physical model for scene variability in the constrained optimization formulation. The algorithm also computes estimates of the amounts of the specified materials in identified mixtures. We demonstrate the effectiveness of the algorithm by using real and synthetic Hyperspectral Digital Imaging Collection Experiment imagery acquired over a range of conditions and altitudes.

Material identification over variation of scene conditions and viewing geometry

As the viewing angle and scene conditions change, the spectral appearance of a material also changes. We present a material identification algorithm for hyperspectral images that is invariant to these changes. Only the solar zenith angle, the viewing angle and sensor altitude, and the spectral reflectance function for the material are required by the algorithm. A material subspace model allows the algorithm to compute an error measure for a given pixel that indicates its similarity to the material. Classification results using USGS mineral reflectance functions and MODTRAN atmospheric functions are presented to demonstrate the performance of the algorithm. Recognition experiments using simulated off-nadir HYDICE images are also presented to demonstrate the use of the algorithm.

A new spatial interaction model for color texture

We present a random environment model for color textures. This model generalizes traditional random field models by allowing the spatial interaction parameters of the field to be random variables. We use this new model to define a color feature vector which captures within and between color band information. We show experimentally that this model improves on the performance of multiband Markov fields for texture classification using small samples.

Modeling and recognizing hyperspectral textures under unknown conditions

We present a method for identifying hyperspectral textures composed of a set of given materials. The algorithm is invariant to illumination and atmospheric conditions as well as the spatial sampling of the texture. Only the spectral reflectance functions for the materials in the texture are required by the algorithm. A texture analysis method based on minimizing the squared error between a pixel spectrum and a synthetic spectral mixture allows pixels to be ranked according to consistency with the texture model. Experimental results using HYDICE imagery demonstrate the use of the method to identify hyperspectral textures under different conditions.