Mikhail L. Uss

Local Signal-Dependent Noise Variance Estimation From Hyperspectral Textural Images

A maximum-likelihood method for estimating hyperspectral sensors random noise components, both dependent and independent from the signal, is proposed. A hyperspectral image is locally jointly processed in the spatial and spectral dimensions within a multicomponent scanning window (MSW), as small as 7 × 7 × 7 spatial-spectral pixels. Each MSW is regarded as an additive mixture of spectrally correlated fractal Brownian motion (fBm)-samples and random noise. The main advantage of the proposed method is its ability to accurately estimate band noise variances locally by using spatial and spectral texture correlations from a single textural MSW. For each spectral band, both additive and signal-dependent band noise components are estimated by linear fit of local noise variances obtained from many MSWs distributed over the whole band intensity range. CRLB-based analysis of the estimator performance shows that a good compromise is to jointly process seven adjacent spectral bands. The proposed method performance is assessed first on synthetic fBm-data and on real images with synthesized noise. Finally, four different AVIRIS datasets from 1997 flying season are considered. Good coincidence between additive and signal-dependent AVIRIS random noise components estimates obtained by our method and the estimates retrieved from AVIRIS calibration data is demonstrated. These experiments suggest that it is worth taking into account noise signal-dependency hypothesis for processing AVIRIS data.

Methods and automatic procedures for processing images based on blind evaluation of noise type and characteristics

In many modern applications, methods and algorithms used for image processing require a priori knowledge or estimates of noise type and its characteristics. Noise type and basic parameters can be sometimes known in advance or determined in an interactive manner. However, it occurs more and more often that they should be estimated in a blind manner. The results of noise-type blind determination can be false, and the estimates of noise parameters are characterized by certain accuracy. Such false decisions and estimation errors have an impact on performance of image-processing techniques that is based on the obtained information. We address some issues of such a negative influence. Possible structures of automatic procedures are presented and discussed for several typical applications of image processing as remote sensing data preprocessing and compression.

Image Informative Maps for Estimating Noise Standard Deviation and Texture Parameters

The problem of automatic detection of image areas appropriate for accurate estimation of additive noise standard deviation (STD) irrespectively to processed image properties is considered in this paper. For accurate estimation of either image texture or noise STD, we distinguish two complementary informative maps: noise- (NI-) and texture- (TI-) informative ones. The NI map is determined and iteratively upgraded based on the Fisher information on noise STD calculated in scanning window (SW) fashion. Fractional Brownian motion (fBm) model for image texture is used to derive the required Fisher information. To extract final noise STD from NI map, fBm- and DCT-based estimators are implemented. The performance of these two estimators is comparatively assessed on large image database for different noise levels. It is also compared with performance of two competitive state-of-the-art estimators recently published. Utilizing NI map along with DCT-based noise STD estimator has proved to be significantly more efficient.

Lossy compression of hyperspectral images based on noise parameters estimation and variance stabilizing transform

Abstract A problem of lossy compression of hyperspectral images is considered. A specific aspect is that we assume a signal-dependent model of noise for data acquired by new generation sensors. Moreover, a signal-dependent component of the noise is assumed dominant compared to a signal-independent noise component. Sub-band (component-wise) lossy compression is studied first, and it is demonstrated that optimal operation point (OOP) can exist. For such OOP, the mean square error between compressed and noise-free images attains global or, at least, local minimum, i.e., a good effect of noise removal (filtering) is reached. In practice, we show how compression in the neighborhood of OOP can be carried out, when a noise-free image is not available. Two approaches for reaching this goal are studied. First, lossy compression directly applied to the original data is considered. According to another approach, lossy compression is applied to images after direct variance stabilizing transform (VST) with properly adjusted parameters. Inverse VST has to be performed only after data decompression. It is shown that the second approach has certain advantages. One of them is that the quantization step for a coder can be set the same for all sub-band images. This offers favorable prerequisites for applying three-dimensional (3-D) methods of lossy compression for sub-band images combined into groups after VST. Two approaches to 3-D compression, based on the discrete cosine transform, are proposed and studied. A first approach presumes obtaining the reference and “difference” images for each group. A second performs compression directly for sub-images in a group. We show that it is a good choice to have 16 sub-images in each group. The abovementioned approaches are tested for Hyperion hyperspectral data. It is demonstrated that the compression ratio of about 15–20 can be provided for hyperspectral image compression in the neighborhood of OOP for 3-D coders, which is sufficiently larger than for component-wise compression and lossless coding.

A Precise Lower Bound on Image Subpixel Registration Accuracy

A new performance bound is proposed for analyzing parametric image registration methods objectively. This original bound is derived from the Cramer-Rao lower bound on the estimation error of parameters involved in a geometric transformation assumed between reference and template images (pure translation in this work) and parameters describing the texture of these images. For describing local fragments of both the reference and the template images, the parametric fractional Brownian motion (fBm) model has been chosen. Experimental results, obtained first on pure fBm data with full matching of the data to the texture model assumption, give evidence that the proposed bound describes more adequately the performance of conventional estimators than two other bounds previously proposed in the literature. This holds with respect to the signal-to-noise ratio value of both images, the roughness of their texture, their correlation, and the actual value of translation parameters between their grids. Then, one real Hyperion hyperspectral data set is considered to test the proposed bound behavior on real data. The proposed bound is demonstrated to describe more adequately the estimation accuracy of the translation parameters between different bands of this data set.

Efficiency of texture image enhancement by DCT-based filtering

Textures or high-detailed structures as well as image object shapes contain information that is widely exploited in pattern recognition and image classification. Noise can deteriorate these features and has to be removed. In this paper, we consider the influence of textural properties on efficiency of image enhancement by noise suppression for the posterior treatment. Among possible variants of denoising, filters based on discrete cosine transform known to be effective in removing additive white Gaussian noise are considered. It is shown that noise removal in texture images using the considered techniques can distort fine texture details. To detect such situations and to avoid texture degradation due to filtering, filtering efficiency predictors, including neural network based predictor, applicable to a wide class of images are proposed. These predictors use simple statistical parameters to estimate performance of the considered filters. Image enhancement is analysed in terms of both standard criteria and metrics of image visual quality for various scenarios of texture roughness and noise characteristics. The discrete cosine transform based filters are compared to several counterparts. Problems of noise removal in texture images are demonstrated for all of them. A special case of spatially correlated noise is considered as well. Potential efficiency of filtering is analysed for both studied noise models. It is shown that studied filters are close to the potential limits.

Processing of images based on blind evaluation of noise type and characteristics

Most modern methods of image processing exploit a priori knowledge or estimates of noise type and its characteristics obtained in blind or interactive manner. However, the results of noise type blind determination can be false with some hopefully rather small probability. Similarly, the obtained estimates of noise parameters are characterized by certain accuracy. Clearly, false decisions and errors of estimates influence performance of image processing techniques that exploit the information on noise properties obtained in a blind manner. In this paper, we consider some aspects of such influence for several typical applications.

Performance analys of segmentation-based method for blind evaluation of additive noise in images

There is a practical need in blind and automatic evaluation of noise characteristics in remote sensing images, especially multi- and hyperspectral ones. It is both desirable to evaluate noise type and variance [1]. This is needed for image filtering, classification, compression, edge detection and other operations of image processing. In the paper we address the task of noise variance evaluation supposing that noise type is known a priori or determined in advance [1]. The main focus is on estimation of additive noise variance although the obtained results can be easily generalized to other types of noise (signal-dependent, multiplicative).

Joint Estimation of Additive and Impulsive Noise Parameters in Remote Sensing Images with Fractal Structure

A novel approach to joint blind estimation of additive noise variance and probability of impulsive noise occurrence in images is put forward. It is based on using a fractal Brownian model for description of real life image characteristics. As shown, this approach allows rather accurate estimation of mixed noise parameters even for images containing a large percentage of texture regions.

Multimodal Remote Sensing Image Registration With Accuracy Estimation at Local and Global Scales

This paper focuses on the potential accuracy of remote sensing (RS) image registration. We investigate how this accuracy can be estimated without ground truth available and used to improve registration quality of mono- and multimodal pair of images. At the local scale of image fragments, the Cramér-Rao lower bound (CRLB) on registration error is estimated for each local correspondence between coarsely registered pair of images. This CRLB is defined by local image texture and noise properties. Opposite to the standard approach, where registration accuracy is only evaluated at the output of the registration process, such valuable information is used by us as an additional input knowledge. It greatly helps in detecting and discarding outliers and refining the estimation of geometrical transformation model parameters. Based on these ideas, a new area-based registration method called registration with accuracy estimation (RAE) is proposed. In addition to its ability to automatically register very complex multimodal image pairs with high accuracy, the RAE method is able to provide registration accuracy at the global scale as a covariance matrix of estimation error of geometrical transformation model parameters or as pointwise registration standard deviation. This accuracy does not depend on any ground truth availability and characterizes each pair of registered images individually. Thus, the RAE method can identify image areas for which a predefined registration accuracy is guaranteed. This is essential for RS applications imposing strict constraints on registration accuracy such as change detection, image fusion, and disaster management. The RAE method is proved successful with reaching subpixel accuracy while registering eight complex mono-/multimodal and multitemporal image pairs including optical-to-optical, optical-to-radar, optical-to-digital elevation model (DEM) images, and DEM-to-radar cases. Other methods employed in comparisons fail to provide in a stable manner accurate results on the same test cases.