1、EIA TEPLOS-17 90 3234600 0557463 74T Systems I In O W t- r n TEPAC PU B L I CAT 10 N MTF Test Method for Monochrome CRT Display TEP1 05-1 7 JULY 199 ELECTRONIC INDUSTRIES ASSOCIATION ENGINEERING DEPARTMENT EIA TEPLO5-L7 90 a 3234600 0559462 686 m NOTICE EIA Engineering Standards and Publications are
2、 designed to serve the public interest through eliminating misunderstandings between manufacturers and purchasers, facilitating interchangeability and improvement of products, and assisting the purchaser in selecting and obtaining with minimum delay the proper product for his particular need. Existe
3、nce of such Standards and Publications shall not in any respect preclude any member or nonmember of EIA from manufacturing or selling products not conforming to such Standards and Publications, nor shall the existence of such Standards and Publications preclude their voluntary use by those other tha
4、n EIA members, whether the standard is to be used either domestically or internationally. Recommended Standards and Publications are adopted by EIA without regard to whether their adoption may involve patents on articles, materials, or processes. By such action, EIA does not assume any liability to
5、any patent owner, nor does it assume any obligation whatever to parties adopting the Recommended Standard or Publication. Technical Publications are distinguished from EIA Recommended Standards or Interim Standards, in that they contain a compilation of engineering data or information useful to the
6、technical community, and represent approaches to good engineering practices that are suggested by the formulating committee. This Publication is not intended to preclude or discourage other approaches that similarly represent good engineering practice, or that may be acceptable to, or have been acce
7、pted by, appropriate bodies. Parties who wish to bring other approaches to the attention of the formulating committee to be considered for inclusion in future revisions of this Publication are encouraged to do so. It is the intention of the formulating committee to revise and update this Publication
8、 from time to time as may be occasioned by changes in technology, industry practice, or government regulations, or for other appropriate reasons. Published by ELECTRON IC IN DUSTRIES ASSOCIATION Engineering Department 2001 Pennsylvania Ave., N.W. Washington, D.C. 20006 Copyright I990 Ail Rights Rese
9、rved Printed in U.S.A. EIA TEPLOS-L7 90 3234600 0559463 512 TEP 105 - 17 MTF TEST METHOD FOR MONOCHROME CRT DISPLAY SYSTEMS EIA JT-20 COMMIlTEE ON CATHODE RAY DISPLAY DEVICES This test method details the calculation of the Modulation Transfer Function (MTF) from resolution data such as those obtaine
10、d using EiA TEPAC 105-7, It A test method for line profile measurements“. A FORTRAN-language software program is provided to allow the implementation of this method with a minimum of personnel training. Since the MTF is derived from CRT line profile data, it is recommended that line profile measurem
11、ents be the preferred specification for CRTs in order to reduce the possibility of computational errors and misunderstandings between suppliers and customers on contractual requirements. The Fourier transform algorithm included in the software package was excerpted from the computer program, “Fast F
12、ourier Transform Algorithm“ by Bergland and Dolan appearing in the IEEE Press Book - Programs for Digital Signal Processing. This algorithm is used with the permission of the Institute for Electrical and Electronics Engineers. i EIA TEP105-17 90 m 3234600 05.59464 459 m TEP 105- 17 MODULATION TRANSF
13、ER FUNCTION TEST PROCEDURE Section that is, to assess how faithfully the CRT reproduces the input signais on the display screen. Resolution Memme nt Techniaues). AU Statement of Purpose The purpose of this test method is to provide a procedure to compute the MTF of monochrome CRTs in order to quanti
14、fy the discemability and quality of displayed information. While the mathematical concepts of the MTF are complex, this test method provides a software package which aiiows users to easily convert spot profile data into MTF data., There is no need for mathematical calculations by the user. The softw
15、are package is designed for use on any computer system with a standard FORTRAN-77 compiler, and a second version of the software for an IBM-PC (or l%-compatibles) persnal computer provides graphical presentations of the spot profile and associated MTF data EIA TEPLOS-L7 90 E 3234600 0559467 T30 W TE
16、P105-17 Page 5 2. DERIVATION: COMPUTATION OF THE MTFs BY FOURIER TRANSFORMS The discrete Fourier transform (in one dimension) of a function f (x) is given as in which and N-1 x=o F (o) = f (x) exp -j2mmtNl j =fi o=- , fork = 0,1,2 ,., (N-1)/2, NAd where Ad denotes the discrete sampling rate, and N d
17、enotes the total number of discrete samples. If x has units of distance, then o is spatial frequency in cycles/distance units. In general, the transform is a complex-number function. It wli be a real-number function only if f(x) has the mathematical property of even symmetry. A most useful property
18、of the MTF for display evaluations is its simplification of the convolution operation. For example, if I (.), m (.), and n (,) are functions with Fourier transforms L (.), M (.), and N (.), respectively, then the convolution expression may be stated as L(o) = M(o) N(o). (4) In words, convolution bec
19、omes simple multiplication in the Fourier transform domain. It is well-understood that the output signal spectrum from a linear display device represents the multiplication of the input signal spectrum with the system (modulation) transfer function. Thus, in Equation 4, if M (.) and N (.) denote the
20、 MTF and the input signai spemm, respectively, then L (.) denotes the output signal spectrum. Note that the algebraic structure of Equation 4 defines the MTF as a ratio of output signai to input signai. Equation 4 provides the basis for an efficient technique to assess the MTF of display system. Tha
21、t is, if a unity input modulation specmm is transmitted through a display device, the modulation spectrum of the resulting output signal represents the MTF directly. Mathematically, a signal which possess a unity modulation spectrum is the EIA TEPLOS-L7 90 0 3234600 0559470 752 TEP105-17 Page 6 delt
22、a-function or ”spike” signal. Presenting a delta-function to a CRT device activates one pixel on the display screen. Conceptually, therefore, the MTF may be computed by taking the Fourier transform of a single raster-line (pixel) profile. Mathematically, the MTF is defied as the ratio of normalized
23、Fourier spectra of the output to the input, if the input modulation spectra is unity across the display bandpass, then the Eq. 5 simplifies to For Gaussian-shaped spot profiles, where s denotes the 50% spot width. The corresponding MTF is Notice with x that the transform has the same functional depe
24、ndence as the original function replaced by o and s replaced by Us. For this reason, Gaussian behavior is frequently assumed. Resolution of a CRT may be taken directly from the MTF curve. The MTF analysis, however, inherently is a tool for the analysis of linear systems. Consequently, the MTF analys
25、is expressions do not hold rigorously for power law expressions, such as the relation between the video signal and the beam current in a CRT. This limitation of MTF analysis must be considered carefully in the evaluation of CRT spot size data. For example: 1) The light output from the phosphor scree
26、n is not a linear function of incident electron density. 2) The beam current is a nonlinear function of drive voltage. Typically, i = cv2J (9) in which i is the beam current and Vis the drive voltage. TEP105-17 Page 7 3) The CRT spot size is a function of the beam current. The Fourier Transform of a
27、 CRT spot or line profile can be calculated in several ways. The simplest is the explicit analytical transformation of an assumed profile expression, as shown in Eq. 8 for a Gaussian spot. If the MTF is needed at only a few explicit frequencies, this solution procedure is very useful. An alternative
28、 method is to use a Fast Fourier Transform (FFT) algorithm to numerically transform a sampled or tabulated spot profile. If the FFT is chosen, it is recommended that the EEE algorithm be used (see bibliography for the reference to a book having a FORTRAN listing of a suitable routine). Regardless of
29、 the method of calculation chosen, the accuracy of the numerical results must be compared to known analytic results. Useful functions for this purpose include the Gaussian discussed above and the square-impulse function 1.0 if 1x1 1/2 0.0 otherwise Both of these functions can be transformed easily b
30、y an analytical method and are weil behaved for numerical computation. The article by Johnson listed in the bibliography is a useful source of MTF results for testing purposes. EIA TEP105-17 90 3234600 0559472 525 TEP1 O 5- 17 Page 8 3. EQUIPMENT REQUIRED: COMPUTER SOFWARE FOR MTF METHOD 3.1 Compute
31、r, IBM Compatible PC with minimum of 512K memory, 8087 math co- processor. Two disk drive preferred. EGA capability is required if plots of data are necessary. 3.2 3.3 Printer or plotter compatible with 3.1. Equipment to measure spot profile (see EL4 TEPAC 105-7A for monochrome CRTs or 105-9 for col
32、or CRTs). 3.4 MTF Calculation Program disc. EIA TEPLOS-17 90 3234600 0559473 461 TEP1 O 5 - 17 Page 9 4. SOFTWARE: INPUT DATA REQUIREMENTS FOR MTF SOFTWARE There are two methods available to enter spot profile data into MTF analysis program: (1) directly from a keyboard or (2) from a disk file. The
33、former method is useful for smail amounts of data generated during program testing or cursory analyses; while, the latter method is preferred for actual display performance analyses since large amounts of data can be be analyzed efficiently. 4.1 Direct Keyboard Data Entry At the MTF program prompt,
34、select the option to enter the spot profile data from the keyboard. The next prompt will ask for the spatial sampling rate. The spatial sampling rate is defined as the center-to-center distance between adjacent spot profile samples. It is required that the distance between samples be equal across th
35、e entire spot profile. Next, the program will prompt for the luminance estimates defining the spot profile. The program will increment the spatial position index automatically as successive luminance values are entered. To terminate the keyboard data entry, enter two successive carriage- rems (Le.,
36、the ENTER key on an IBM-PC keyboard). 4.2 Data File Data Entry At the MTF program prompt, select the option to read the spot profile data from a disk- resident data file. The disk file must be formatted properly in order for the MTF program to accept the spot profile data. The data Ne format is simp
37、le, but must be followed strictly. Each line in the data file must contain two values; the first value is the spatial position of the spot profile sample, while the second value is the luminance reading. The two values on each data file line must be separated by a space character, and the end of eac
38、h data file line must be marked with a carriage-rem character. No other formatting requirements are needed. EIA TEPLOS-17 90 m 3234600 0559474 3T8 m TEP1 O 5-1 7 Page 10 5. PROCEDURE: OPERATION OF MTF SOFTWARE 5.1 For IBM compatible PC * * (Standard Version) 5.1.1 Load System DOS. 5.1.2 Insert disk
39、in drive A. 5.1.3 Press “A:“ and “RETURN“ 5.1.4 Press “CD A:STD“ and “RETURN“. Note: The DOS disk subdirectory entitled “STD“ contains the executable MTF program file as weii as FORTRAN source code files required to compile the program for non-IBM/DOS computers. J 5.1.5 Press “MTF“ and “RETURN“ 5.1.
40、6 Follow prompts to spec* title, input/output options and sample spacing. Defaults are indicated in brackets. 5.1.7 Enter values of spot profile amplitude at each increment of sample spacing entered in paragraph 5.1.6. 5.1.8 Enter “Q“ to end data entry. 5.1.9 Enter “H for HELP or “M“ to compute MTF.
41、 5.1.10 Press “RETURN“ for each page of h4TF data (10 pages total). 5.1.11 Enter “Q“ to begin next analysis 5.1.12 To end session enter “E“ for EXIT or “RETURN“ will default to EXIT. 5.2 5.2.1 Load system DOS. For IBM compatible PC *GA QJ an MTF value of unity indicates total transmission of the inp
42、ut signal, wheFeas an MTF less than unity indicates a degree of attenuation in the output signai relative to the input signai. Only an idealized system possesses an MTF of unity across the entire spatial frequency passband. Aii real systems possess MTFs that approach zero at some high spatial freque
43、ncy lunit. In most cases, the larger the area under the MTF, the higher the image quaiity, and, hence, the performance of the display system. EIA TEPLOS-17 90 3234600 0559480 bTL TEP105-17 Page 16 9. ANNOTATED BIBLIOGRAPHY: MTF TEST METHODS Abbott, F., “Practical Aspects of Transfer Function Measure
44、ment“, Proceedings SPIE Seminar on Modulation Transfer Function, March 1968. Review of concept and theory; emphasis on measurement equipment, procedure and technique. Anstey, G. and Dore, M. J., “Automatic Measurement of Cathode Ray Tube MTFs“, Royal Signals and Radar Establishment, Memorandum No. 3
45、301 (dist. by US Defense Technical information Center), August 1980. Techniques, procedures, equipment/instrumentation, operational considerations. System described is semi- automatic; addresses monochrome CRT displays, image intensifiers. Banbury, J. R. and Whitfield, F. B., “Measurement of Modulat
46、ion Transfer Function for Cathode Ray Tubes“, Displays, January 1981, pp. 189-198. Methods, equipment, effects of CRT characteristics, equipment considerations; proposed method employs fixed slit, electronic signal. Banbury, J. R., “Evaluation of MTF and Veiling Giare Characteristics of CRT Displays
47、“, Displays, January 1982, pp. 23-29. Methods, CRT operating characteristics, veiling glare, contrast index. Bartfeld, D. P. and Bela J. P., “Semi-Automatic Measurement of the Two-Dimensional Intensity Distribution of a Color CRT Electron-Beam Spot“, Society for Information Display, 1987 Digest, May
48、 1987, pp. 225-227. Measurement technique, equipment, noise reduction by contrast enhancement. Acquires accurate contour profiles of electron beam in shadow mask color CRT. Bergland, G. D. and Dolan, M. T., “Fast Fourier Transform Algorithms“, In IEEE Press Book, “Programs for Digital Signal Process
49、ing“, Institute of Electrical and Electronics Engineers, New York, 1979, pp 1.2-1 to 1.2-18. FORTRAN computer language listing for Fast Fourier Transform algorithms. Bloomfield, P., “Fourier Analysis of Time Series: An Introduction“, Wiiey, 1976. Text, Chapter 5, pp. 79-101. Spectrai analysis of agriculture yields; analogous to two dimensional variability of CRT spot deflected across faceplate. Spatial auto- regression; regression and spatial autocorrelation. Clodfelter, R., “Modulation Transfer Function for the Display Engineer“, SPIE A
