探索HDR成像的艺术与科技：融合传统与未来

图像处理

图像质量

需积分: 11 77 浏览量更新于2024-07-17 1 收藏 92.06MB PDF 举报

身份认证购VIP最低享 7 折!

领优惠券(最高得80元）

资源详情

资源推荐

xvi Contents

30.3.4 Rod Retinex plus L-cone Color Appearance Demonstration 276

30.3.5 Rod Retinex plus L-cone Retinex Color Space 278

30.3.6 Rod Retinex plus L-cone Retinex 278

30.4 Assembling Appearance: Color Constancy, Rod Vision and HDR 279

30.5 Summary 280

30.6 References 280

Section F HDR IMAGE PROCESSING 283

31 HDR Pixel and Spatial Algorithms 285

31.1 Topics 285

31.2 Introduction – HDR Image Processing Algorithms 285

31.3 One Pixel – Tone Scale Curves 286

31.3.1 One Pixel – Histogram 286

31.3.2 One Pixel – LookUp Tables 287

31.3.3 Using 1-D LookUp Tables for Rendering HDR Scene Captures 287

31.4 Some of the Pixels – Local Processing 288

31.5 All of the Pixels 289

31.6 All Pixels and Scene Dependent – The Retinex Extended Family 289

31.7 Retinex Algorithms 290

31.8 ACE Algorithms 290

31.9 Analytical, Computational and Variational Algorithms 290

31.10 Techniques for Analyzing HDR Algorithms 290

31.11 The HDR Story 291

31.12 References 291

32 Retinex Algorithms 293

32.1 Topics 293

32.2 Introduction 293

32.2.1 Three Independent Spatial Retinexes Calculate Three Lightnesses – 1964 293

32.2.2 Ratio-Product 294

32.3 How to Calculate Lightness Using Ratio-Products 297

32.3.1 Normalization 297

32.3.2 Models that Include the Limitations of Vision 298

32.3.3 Gradient Threshold 298

32.3.4 Reset 299

32.3.5 Average Old and New Products 300

32.3.6 Scene Dependent Output 301

32.3.7 Models Using Imposed Limits in Computation 301

32.4 A Variety of Processing Networks 301

32.5 Image Content 302

32.5.1 Normalization in Vision – Neither Local, Nor Global 303

32.5.2 Tools to Control a Response to Scene Content 305

32.5.3 Edges and Gradients 305

32.6 Real Images – 1975 307

32.6.1 Real Images in Real Time 308

32.6.2 Goal: Lightness Field Replaces Reﬂ ectance 311

32.6.3 Dynamic Range Compression from Threshold or Reset 312

Preface

Vision and visual reproduction are a giant puzzle. By comparison, it is much more complicated than a

jigsaw puzzle that lies on a ﬂ at table. We know a lot about the ﬂ at puzzle piece before we start. We

know the surface that lays on the table, and just need to ﬁ gure out its location and orientation.

Vision is more like a wooden three - dimensional puzzle in which the craftsman passes the jigsaw

through a volume of wood at many angles and adds many curves. All the pieces of such solid block

puzzles have complex irregular shapes. To solve these puzzles we need to look at each piece in all

possible ways. We need to inspect each surface to recognize its shape, and then search for its matching

surface. We rotate each piece in our hands to see it from all possible directions.

Vision has far more than three dimensions. However, the principle is the same. We need to look at

the properties of vision from all possible directions to see if we can ﬁ nd pieces that ﬁ t together.

All HDR imaging is the result of having variable illuminations falling on objects. This variable

illumination is the result of the 3 - D world that has multiple illuminants, shadows, specular and diffuse

reﬂ ections from other objects. The scene components in a 3 - D world play a big part in controlling the

light coming from them. This is quite different from ﬂ at 2 - D surfaces in uniform illumination. Just as

with jigsaw puzzles, 3 - D scenes are more interesting and complex than spots of light, and 2 - D ﬂ at

scenes.

This text integrates three topics that are thought of as independent disciplines: artists ’ renditions of

scenes and their illumination, color and lightness constancy, and digital HDR imaging. Each is a rich

ﬁ eld of study; each is a piece of the puzzle. The shapes of these topics ﬁ t together. Their contours match

because measurements of appearances in each give the same results. This book is about those measure-

ments of appearances, and models of the results.

This text differs from others in the Wiley - IS & T Series in Imaging Science & Technology in that it

concentrates on measuring color appearances in complex scenes. It integrates and extends the work of

two photographers who were close friends: Edwin Land and Ansel Adams. One began as an optical

physicist and inventor, and then built a great company around instant photography. The other began as

a pianist, and turned to photography to express his incredible artistic talent. Both Adams and Land were

IS & T Fellows; both were awarded the United States Presidential Medal of Freedom.

Adams taught his Zone System to many thousands of photographers. His books have many scientiﬁ c

plots of ﬁ lm - response functions measured with spots of light. The Zone System taught photographers

HDR image capture. That is the record of the scene; it is the analog of a musical score. Making the

print, the analog of performing the score, departed from the physics of spots of light and turned to

implement the skills of painters in controlling the spatial information content. By controlling the light

from the negative in each spatial region of the print, Adams taught how to perform HDR compression.

He not only ﬁ t the very large dynamic range of scenes in the U.S. southwest into a lower dynamic

range print, he synthesized the print ’ s appearance using spatial image processing. Adams did not accu-

rately reproduce the HDR world; he captured it, transformed it, and rendered his artistic visualization

of it in the print.

At the same time that Adams started to write about the Zone System, his friend Edwin Land started

to create instant photography. Ten years later, after receiving hundreds of photographic patents, Land

剩余391页未读，继续阅读

haz2yy

粉丝: 5
资源: 24

探索HDR成像的艺术与科技：融合传统与未来

The Art and Science of CCD Astronomy / edited by David Ratledge.

英文原版-The ESC Textbook of Cardiovascular Imaging 1st Edition

the early second paper about Optical computational imaging

Describe the background information of Significance of analyzing metal-transfer images for quality control and process optimization in detail

Contrast recovery

The Mathematics of Computerized Tomography

帮我检索癫痫核磁共振近五年发表的文献，并简单说明文献内容

cannot import name _imaging from PIL

Voxel-based morphometric magnetic resonance imaging (MRI) postprocessing

ImportError: DLL load failed while importing _imaging

detector layer

write me a ultrasound channel data generator

缺陷检测的期刊，结合深度图像的

Compositional Human Pose Regression

自旋回波序列扫描需要几分钟

virtual imaging technology

Image Deconvolution

外文文献 基于OpenCV的图像处理技术以及谷歌开源的图像识别工具Tesseract

最新资源

外文文献基于OpenCV的图像处理技术以及谷歌开源的图像识别工具Tesseract