We describe a technique for image encoding in which local operators of many scales but identical shape serve as the basis functions. The representation differs from established techniques in that the code elements are localized in spatial frequency as well as in space. Pixel-to-pixel correlations are first removed by subtracting a lowpass filtered copy of the image from the image itself. The result is a net data compression since the difference, or error, image has low variance and entropy, and the low-pass filtered image may represented at reduced sample density. Further data compression is achieved by quantizing the difference image. These steps are then repeated to compress the low-pass image. Iteration of the process at appropriately expanded scales generates a pyramid data structure. The encoding process is equivalent to sampling the image with Laplacian operators of many scales. Thus, the code tends to enhance salient image features. A further advantage of the present code is that it is well suited for many image analysis tasks as well as for image compression. Fast algorithms are described for coding and decoding.

/pdf/the-laplacian-pyramid-as-a-compact-image-code-3pr3qcg6c5.pdf

The Laplacian Pyramid as a Compact Image Code

http://library.utia.cas.cz/prace/20030125.pdf

Image registration methods: a survey

The structural similarity image quality paradigm is based on the assumption that the human visual system is highly adapted for extracting structural information from the scene, and therefore a measure of structural similarity can provide a good approximation to perceived image quality. This paper proposes a multiscale structural similarity method, which supplies more flexibility than previous single-scale methods in incorporating the variations of viewing conditions. We develop an image synthesis method to calibrate the parameters that define the relative importance of different scales. Experimental comparisons demonstrate the effectiveness of the proposed method.

/pdf/multiscale-structural-similarity-for-image-quality-5a7vl3othe.pdf

Multiscale structural similarity for image quality assessment

Brains, it has recently been argued, are essentially prediction machines. They are bundles of cells that support perception and action by constantly attempting to match incoming sensory inputs with top-down expectations or predictions. This is achieved using a hierarchical generative model that aims to minimize prediction error within a bidirectional cascade of cortical processing. Such accounts offer a unifying model of perception and action, illuminate the functional role of attention, and may neatly capture the special contribution of cortical processing to adaptive success. This target article critically examines this "hierarchical prediction machine" approach, concluding that it offers the best clue yet to the shape of a unified science of mind and action. Sections 1 and 2 lay out the key elements and implications of the approach. Section 3 explores a variety of pitfalls and challenges, spanning the evidential, the methodological, and the more properly conceptual. The paper ends (sections 4 and 5) by asking how such approaches might impact our more general vision of mind, experience, and agency.

/pdf/whatever-next-predictive-brains-situated-agents-and-the-epj56cs868.pdf

Whatever next? Predictive brains, situated agents, and the future of cognitive science

Image processing techniques find applications in many areas, chief among which are image enhancement, pattern recognition, and efficient picture coding. Some aspects of image processing are discussed--specifically: the mathematical operations one is likely to encounter, and ways of implementing them by optics and on digital computers; image description; and image quality evaluation. Many old results are reviewed, some new ones presented, and several open questions are posed.

Image processing

The introduction of the fast Fourier transform algorithm has led to the development of the Fourier transform image coding technique whereby the two-dimensional Fourier transform of an image is transmitted over a channel rather than the image itself. This devlopement has further led to a related image coding technique in which an image is transformed by a Hadamard matrix operator. The Hadamard matrix is a square array of plus and minus ones whose rows and columns are orthogonal to one another. A high-speed computational algorithm, similar to the fast Fourier transform algorithm, which performs the Hadamard transformation has been developed. Since only real number additions and subtractions are required with the Hadamard transform, an order of magnitude speed advantage is possible compared to the complex number Fourier transform. Transmitting the Hadamard transform of an image rather than the spatial representation of the image provides a potential toleration to channel errors and the possibility of reduced bandwidth transmission.

Hadamard transform image coding

From the Publisher:
Today, image processing has become a vital, cost-effective technology in a host of applications. This Second Edition covers these new developments in this dynamic field. Five sections explore the characterization of continuous images, image sampling and quantization techniques along with the mathematical representation of discrete images, two-dimensional signal processing techniques, image enhancement and restoration techniques and image analysis.

Digital Image Processing: Piks Inside

An extension to the basic concept of correlation detection as a means of image registration is developed. The technique involves linear spatial preprocessing of the inages to be registered prior to the application of a correlation measure. This preprocessing operation utilizes the spatial correlation within each image and greatly improves the detectability of image misregistration. An analysis of the computational aspects of the algorithm is given. Also, results of a computer simulation to evaluate the technique are given.

Correlation Techniques of Image Registration

Preface Acknowledgments PART 1 CONTINUOUS IMAGE CHARACTERIZATION 1 Continuous Image Mathematical Characterization 11 Image Representation 12 Two-Dimensional Systems 13 Two-Dimensional Fourier Transform 14 Image Stochastic Characterization 2 Psychophysical Vision Properties 21 Light Perception 22 Eye Physiology 23 Visual Phenomena 24 Monochrome Vision Model 25 Color Vision Model 3 Photometry and Colorimetry 31 Photometry 32 Color Matching 33 Colorimetry Concepts 34 Tristimulus Value Transformation 35 Color Spaces PART 2 DIGITAL IMAGE CHARACTERIZATION 4 Image Sampling and Reconstruction 41 Image Sampling and Reconstruction Concepts 42 Monochrome Image Sampling Systems 43 Monochrome Image Reconstruction Systems 44 Color Image Sampling Systems 5 Image Quantization 51 Scalar Quantization 52 Processing Quantized Variables 53 Monochrome and Color Image Quantization PART 3 DISCRETE TWO-DIMENSIONAL PROCESSING 6 Discrete Image Mathematical Characterization 61 Vector-Space Image Representation 62 Generalized Two-Dimensional Linear Operator 63 Image Statistical Characterization 64 Image Probability Density Models 65 Linear Operator Statistical Representation 7 Superposition and Convolution 71 Finite-Area Superposition and Convolution 72 Sampled Image Superposition and Convolution 73 Circulant Superposition and Convolution 74 Superposition and Convolution Operator Relationships 8 Unitary Transforms 81 General Unitary Transforms 82 Fourier Transform 83 Cosine, Sine and Hartley Transforms 84 Hadamard, Haar and Daubechies Transforms 85 Karhunen-Loeve Transform 9 Linear Processing Techniques 91 Transform Domain Processing 92 Transform Domain Superposition 93 Fast Fourier Transform Convolution 94 Fourier Transform Filtering 95 Small Generating Kernel Convolution PART 4 IMAGE IMPROVEMENT 10 Image Enhancement 101 Contrast Manipulation 102 Histogram Modification 103 Noise Cleaning 104 Edge Crispening 105 Color Image Enhancement 106 Multispectral Image Enhancement 11 Image Restoration Models 111 General Image Restoration Models 112 Optical Systems Models 113 Photographic Process Models 114 Discrete Image Restoration Models 12 Image Restoration Techniques 121 Sensor and Display Point Nonlinearity Correction 122 Continuous Image Spatial Filtering Restoration 123 Pseudoinverse Spatial Image Restoration 124 SVD Pseudoinverse Spatial Image Restoration 125 Statistical Estimation Spatial Image Restoration 126 Constrained Image Restoration 127 Blind Image Restoration 128 Multi-Plane Image Restoration 13 Geometrical Image Modification 131 Basic Geometrical Methods 132 Spatial Warping 133 Perspective Transformation 134 Camera Imaging Model 135 Geometrical Image Resampling PART 5 IMAGE ANALYSIS 14 Morphological Image Processing 141 Binary Image Connectivity 142 Binary Image Hit or Miss Transformations 143 Binary Image Shrinking, Thinning, Skeletonizing and Thickening 144 Binary Image Generalized Dilation and Erosion 145 Binary Image Close and Open Operations 146 Gray Scale Image Morphological Operations 15 Edge Detection 151 Edge, Line and Spot Models 152 First-Order Derivative Edge Detection 153 Second-Order Derivative Edge Detection 154 Edge-Fitting Edge Detection 155 Luminance Edge Detector Performance 156 Color Edge Detection 157 Line and Spot Detection 16 Image Feature Extraction 161 Image Feature Evaluation 162 Amplitude Features 163 Transform Coefficient Features 164 Texture Definition 165 Visual Texture Discrimination 166 Texture Features 17 Image Segmentation 171 Amplitude Segmentation 172 Clustering Segmentation 173 Region Segmentation 174 Boundary Segmentation 175 Texture Segmentation 176 Segment Labeling 18 Shape Analysis 181 Topological Attributes 182 Distance, Perimeter and Area Measurements 183 Spatial Moments 184 Shape Orientation Descriptors 185 Fourier Descriptors 186 Thinning and Skeletonizing 19 Image Detection and Registration 191 Template Matching 192 Matched Filtering of Continuous Images 193 Matched Filtering of Discrete Images 194 Image Registration PART 6 IMAGE PROCESSING SOFTWARE 20 PIKS Image Processing Software 201 PIKS Functional Overview 202 PIKS Scientific Overview 21 PIKS Image Processing Programming Exercises 211 Program Generation Exercises 212 Image Manipulation Exercises 213 Color Space Exercises 214 Region-of-Interest Exercises 215 Image Measurement Exercises 216 Quantization Exercises 217 Convolution Exercises 218 Unitary Transform Exercises 219 Linear Processing Exercises 2110 Image Enhancement Exercises 2111 Image Restoration Models Exercises 2112 Image Restoration Exercises 2113 Geometrical Image Modification Exercises 2114 Morphological Image Processing Exercises 2115 Edge Detection Exercises 2116 Image Feature Extraction Exercises 2117 Image Segmentation Exercises 2118 Shape Analysis Exercises 2119 Image Detection and Registration Exercises Appendix 1 Vector-Space Algebra Concepts Appendix 2 Color Coordinate Conversion Appendix 3 Image Error Measures Appendix 4 PIKS Compact Disk Bibliography Index

William K. Pratt

Papers

Hadamard transform image coding

Digital Image Processing: Piks Inside

Correlation Techniques of Image Registration

Digital Image Processing: Piks Scientific Inside

Computer Techniques in Image Processing