<title>Image quality measures for performance assessment of compresssion transforms</title>

Abstract

Image compression is increasingly employed in the exploitation of limited communication channel bandwidth and maximization of storage system efficiency. Although traditional measures of signal or image quality (e.g., mean-squared error or MSE) are useful in a communication context, few such measures are evident in the open literature that effectively address implementational issues such as visual quality or suitability for automated target recognition (ATh) of a decompressed image. For example, reconstruction errors such as blocking effect, aliasing or suppression of high frequencies, preservation of image or neighborhood statistics, and quantization-induced perturbation of line segments or arcs can degrade visual image quality.In numerous ATh applications, such error is an important source of ATh algorithm performance degradation. Unfortunately, the majority of published image quality measures (IQMs) tend to be based on ad hoc or first-order models of human visual system (HVS) function. Such models do not necessarily correlate with the appearance of image details (e.g., higherorder models) or mathematical descriptions of ATR filters (e.g., non-HVS models). Additionally, published IQMs tend not to address the primary ATR problem of feature visibility or the secondary problem of estimating image or neighborhood error distributions in decompressed imagery.In this paper, a collection of performance measures and IQMs designed specifically for image compression is presented. Beginning with the customary performance measures of MSE and space-time bandwidth product, we progress to spatial measures such as cutoff frequency, effect of greyscale quantization on variance, measures of texture preservation, and statistics of the modulation transfer function (MTF). Texture is estimated by an area-based variance descriptor derived from the Lorentzian distribution, which has been successfully employed in ATh practice [1]. Measures for disruption of linear features such as line segments and arcs are based on analysis of the Hough transform and spatial linear regression. Performance analysis of each measure is couched in terms of computational cost, sensitivity to error, and relevance to ATh scenarios. Examples are given for template-and codebook-based transforms such as JPEG, VPIC, EPIC, and EBLAST [2][3][4][5].

Detection and Characterization of Motion in Video Compression

“…As shown in [31][32][33], the vast majority of error measures for image compression are based in one way or another on MSE, which does not necessarily describe subtleties such as feature orientation errors, or the effect of feature distortions on the syntactic or semantic perception of an object within the context of a scene. Compromises in the complexity/accuracy tradeoff have been reported in [25,34,35], where multispectral image properties are employed to yield a small residual color error and coarse motion information that is sufficiently accurate to satistfy application-specific constraints.…”

<title>Performance analysis of tabular nearest-neighbor encoding for joint image compression and ATR: I. Background and theory</title>

“…The latter is a difficult goal, due to considerations of mathematical rigor in deriving error measures that should also correspond to realistic assessments of visual image quality. In previous research, two of the authors (MSS and FMC) have addressed the development of image quality measures (IQMs) for digital imagery, to which discussion the reader is referred for additional background [13]. Implementationally, two additional issues arise in VQ-based compression.…”