Spie Press Book • on saleImage Performance in CRT Displays
|Format||Member Price||Non-Member Price|
- INTRODUCTION / 1
- Chapter 1 CRT GLASS / 5
- 1.1 Color Glass / 6
- 1.2 Monochrome Glass / 8
- 1.3 Glass as an Insulator / 8
- 1.4 Usable Area / 9
- 1.5 Contrast Enhancement / 10
- 1.6 X-Ray Compliance / 11
- References / 11
- Chapter 2 CRT ARCHITECTURES / 13
- 2.1 Color CRT Mask / 14
- 2.2 Monochrome CRT / 16
- 2.3 CRT Mechanicals and Safety System / 17
- References / 19
- Chapter 3 ELECTRON OPTICS / 21
- 3.1 Basic Structures / 21
- 3.2 Lower Gun Structure (Triode) / 24
- 3.3 Upper Gun Structure (Lens) / 28
- 3.4 Cathode Selection / 29
- 3.4.1 Cathode life / 30
- 3.4.2 Cathode drift / 32
- 3.4.3 Cathode drive gamma / 32
- 3.5 Focus Considerations / 34
- References / 35
- Chapter 4 PHOSPHORS / 37
- 4.1 Phosphor Efficacy / 38
- 4.1.1 Thermal quenching / 39
- 4.1.2 Decay time / 40
- 4.2 Phosphor Aging / 41
- 4.2.1 Short-term aging / 42
- 4.2.2 Long-term aging / 42
- 4.2.3 Color shift / 43
- 4.3 Phosphor Spatial Noise / 44
- 4.4 Phosphor Screen Weight / 45
- References / 46
- Chapter 5 GENERATING THE PIXEL / 47
- 5.1 The Gaussian Pixel / 47
- 5.2 Raster Addressability Ratio (RAR) / 49
- 5.3 Control of the Pixel / 50
- 5.3.1 Vertical pixel modulation / 51
- 5.3.2 Horizontal pixel modulation / 52
- 5.4 Dynamic Response as a System / 52
- 5.4.1 Video amplifier response / 53
- 5.4.2 Depth of modulation / 56
- 5.4.3 Calibration errors / 60
- References / 62
- Chapter 6 LUMINANCE UNIFORMITY / 63
- 6.1 Sweep Circuit Contributions / 64
- 6.2 Beam Landing Angle / 65
- 6.3 Focus and Current Density / 65
- 6.4 High-Voltage Stability / 66
- 6.5 Compensating for Nonuniformity / 67
- 6.5.1 CRT manufacturing process controls / 67
- 6.5.2 Variable gain amplifier / 67
- 6.5.3 Raster modulation / 68
- 6.6 Considerations on Performance / 68
- References / 69
- Chapter 7 TEST PATTERNS AND HOW TO READ THEM / 71
- 7.1 The SMPTE Test Pattern / 71
- 7.1.1 Overview at the macro level / 72
- 7.1.2 Details reveal bandwidth / 74
- 7.2 Briggs Test Pattern 4 / 75
- 7.2.1 Basic panel construct / 76
- 7.2.2 Quadrant construct / 76
- 7.2.3 Briggs test pattern scoring / 77
- 7.3 AAPM Quality Control Test Patterns / 80
- 7.4 Additional Test Patterns / 81
- References / 82
- Chapter 8 VIDEO CARDS AND IMAGE QUALITY / 83
- 8.1 Video Card Basics / 83
- 8.2 Eight-Bit versus Ten-Bit Internal / 85
- 8.3 Pixel Formats Supported / 86
- 8.4 Connections / 87
- 8.5 Calibration of Display and Card / 88
- 8.5.1 Preprocessing in software / 89
- 8.5.2 Processing in hardware / 90
- 8.5.3 Monochrome alternative with a color card / 91
- References / 93
- Chapter 9 WINDOW AND LEVEL / 95
- 9.1 Brightness and Contrast Review / 95
- 9.2 Window Width and Window Level / 96
- References / 98
- Chapter 10 OVERVIEW OF CRT RASTER DISPLAY / 99
- 10.1 Sync Signals / 100
- 10.2 Video Signal Distortions / 101
- 10.3 High-Voltage Source / 102
- 10.4 Interlaced and Sequential Scan / 102
- 10.5 Geometric Distortions / 103
- 10.6 Digital versus Analog Control / 108
The purpose of this book is to bring a broad spectrum of information related to cathode ray tube (CRT)-based displays into a single easy-to-understand narrative. It requires no working knowledge of a television or how one programs a video cassette recorder (VCR). The starting point of each chapter will be basic information that is followed by detailed explanations and insight into the design trade offs that influence the image observed. The sequence of topics follows that in a workshop prepared by the author, and the chapters may be read in any order. However, the information in each chapter does build upon the material in the preceding chapters. It should be noted that all references to a cathode ray tube in this book are to only the glass part giving off light in a display. They do not include the entire display or monitor with the associated electronics, a point of confusion at times even within the industry.
All CRTs use glass as a starting point; formulas involved (glass melt) provide the variations in performance. In the 1950s when big names in television such as RCA, Sylvania, and Philco were working on color CRT technology, they all had the same problem: monochrome glass could not withstand the high voltages required to make color workable. A breakthrough in glass additives solved this shortcoming and made it possible to achieve 4 and 5 megapixel medical displays. The performance properties and safety limits of the various glass melts are discussed in this book as they relate to monochrome applications in medical imaging.
An overview of the architectural differences between color and monochrome CRTs discusses how they are manufactured and the compromises required by their respective design limits. This leads into the subject of electron optics. Here the main focus is on monochrome optics because it provides performance beyond 1k-line displays.
In addition to the performance of electron optics, the cathode (electron beam source) is examined as a failure mode for all CRTs. The types of cathodes available and their life expectancy are discussed in terms of cost of ownership, with an example calculation. For medical applications, the inability to render full image fidelity is the true failure mode, not the failure of the CRT to emit luminance. The way in which the electron beam is formed and controlled through the optics determines the shape of the pixel and thus the image quality. The influence of electron optics on the CRT gamma and related performance compromises are discussed in conjunction with phosphor selection.
To say the CRT is a mature product is stating the obvious. Sir William Crookes developed the progenitor of the modern electron gun in 1878 as he experimented with variations on the Geisler discharge tube. Then in 1897, the German physicist Karl Ferdinand Braun demonstrated a tube intended to display electrical waveforms. It was not until 1920 that Vladimir Zworykin of Westinghouse Electric developed the other components needed for the first camera and picture tube, respectively called the iconoscope and kinescope.
In the 1930s the first broadcast architecture was tested using a format that became the standard for North America. The National Television Standards Commission (NTSC) established a format of 520 lines interlaced at 60 Hz refresh. Given the level of performance available with vacuum tubes, interlacing the video with odd and even lines was a necessity. In this way, the video amplifier wrote only half the lines with each vertical scan. This in turn kept the horizontal scan rate down to 15 kHz. An NTSC television, to this day, displays broadcast signals the same way it did in the late 1940s when commercial television became a reality. Because of other limitations at the time, only about 480 lines of the 520 in the signal can be seen. This is called overscan, meaning that the active video is larger than the viewing space provided. Many control problems could be hidden in the area just outside of what is visible.
Today's color monitors and monochrome medical-grade displays run at frequencies well above those of television and put all the information within the available viewing area. A SVGA boot format of 800 X 600 pixels starts at 30 kHz and climbs to 105 kHz to support 1600 X 1200 at 72 Hz refresh. In a medical portrait orientation, 118 kHz is required to support the same pixel format. A five megapixel (2560 lines) display tops out at 180 kKHz. How does this differ from TV technology in terms of design application? A TV set today can be reduced to a handfull of integrated circuits (ICs) that include the power output for a number of circuits. The performance required for medical-grade displays is not to be found in a handful of ICs.
Phosphors are more than just a color preference to be based on historical film usage. There are efficacy and long-term aging considerations that determine calibration cycles and the ability to color match old and new displays in multihead workstations. The image quality as defined by the individual pixel requires careful consideration of both the electron optics and phosphor performance. In addition, spatial noise is a factor to be considered with all blended phosphors against the image complexity of the source modality.
It is relatively easy to generate pixels. Being able to resolve them is the key to superior image quality. The metrics involved in defining a pixel and how distortions can influence the net results are illustrated using Microvision scans of both individual pixels and a series of pixels at the Nyquist frequency with two types of optics and video amplifier. Pixel fidelity is also separated into vertical and horizontal aspects of performance, which are controlled by the optics and video amplifier, respectively, in a raster-scanned device. This leads to a net performance as illustrated by a depth of modulation (DMOD) scan for the same optic/video combinations.
Luminance uniformity on a CRT display is generally better than that of an average light box. What contributes to this phenomenon is a multitude of events working against the intended result. Compensation can be utilized, but there is a price to be paid, and how CRT uniformity is defined is still subject to question. Two potential approaches to defining uniformity are reviewed and weighed against uncompensated results.
Compliance with the DICOM grayscale standard display function (GSDF) is reviewed to illustrate how important it is to specify a display's performance with hard numbers, particularly video bandwidth. The background information from the preceding chapters is needed to fully appreciate this discussion.
Test patterns and how to read them for information about a display's performance can prevent second-guessing in the absence of test equipment. Use of the test pattern of the Society of Motion Pictures and Television Engineers and Briggs test pattern 4 as quality assurance tools illustrates the benefits of proper utilization and indicates what is not acceptable. The American Association of Physicists in Medicine (AAPM), Task Group 18 has developed quality control patterns specifically for medical imaging.
The video card, whether it is a commercial graphics or a custom medical card, is part of the video path and should always be tested in conjunction with the intended display. Video card performance varies with manufacturer, and not all digital-to-analog converters are created equal, all of which contribute to the shape of the pixel. Medical grade cards are discussed with alternative paths using software preprocessing or software compensation based on commercial color cards for monochrome displays.