Calibration method for video image reproduction with electronic printer and video monitor.

0001The present invention relates to the calibration of a cathode ray tube (CRT) multiformat camera (MFC) and of a video monitor to provide a consistent reproduction of a video image both on film and on the monitor, using a gray scale which matches the visual perception characteristics and provides equal visualization for equal image value changes over the entire film density range.

0002It is well known that the sensitivity of the human eye to small variations in shades of gray is intrinsically nonlinear with respect to the optical density of film. According to the literature (see: C.J. Bartleson and E.J. Breneman, "Brightness perception in Complex Fields", J. Opt. Soc. Am., vol. 57, pp. 953-957 (1967). J.L. Mannos and D.J. Sakrison, "The Effects of a Visual Fidelity Criterion on the Encoding of Image", IEEE Trans. Inf. Theory, vol. 20, pp. 525-536 (1974).), the relationship between perceived lightness (S_f) and film density (D_f) was measured to be<maths id="math0001"><img file="EP0542012A1_D0001.tif" /></maths> where c₁ and c₂ are normalization constants, L_f is the luminance from film, K is the view-box luminance, and p is close to 1/3.

0003Conventionally, a Cathode Ray Tube Multi-Format Camera (CRT MFC) is calibrated to produce a gray scale on film which matches the gray scale of the monitor (see: R.P. Schwenker, "Film Selection Considerations for Computed Tomography and Ultrasound Video Photography", SPIE, vol. 173, pp. 75-80 (1979); E.R. Ritenour, S.N. Sahu, R.P. Rossi, and T.R. Nelson, "Quantitative Methods for Hard-Copy Device Adjustment", SPIE, vol. 767, pp. 529-535 (1987); H. Shalit, "Method and System in Video Image Hard Copy Reproduction", US-A-4,939,581; L.G. Glasser, A.H. McKinney, C.D. Reilly, and P.D. Schnelle, "Cube-Root Color Coordinate System", J. Opt. Soc. Amer., vol. 48, pp. 736-740, 1958.) However, there is no universal or standard gray scale for the monitor. Usually, the gray scale on the monitor is subjectively set up by the individual radiologist or technician. Therefore, the "look" of the images on film will vary and depend on the setup of the monitor.

0004Alternatively, a CRT MFC can be calibrated to give a gray scale on film with uniform quantization in density, that is, the output film density varies linearly with the input code value. From Eq. (1), the perceived lightness will be a nonlinear function of digital image code value. As a result, it would be difficult to see the differences in gray shades in film at high-densities under standard view-box conditions.

0005The following patents also disclose various video printers, none of which provide a solution to the problem loss of low gray scale and high gray scale detail.    US-A-4,742,397; US-A-4,794,460; US-A-4,586,085; US-A-4,962,433; US-A-4,709,271; and US-A-5,018,085.

0006According to the present invention, the calibration of a CRT MFC and a video monitor involve the correction of a video image, on a pixel-by-pixel basis, and the production of a gray scale on a photosensitive medium, such as film, and on a monitor with uniform quantization in perceived lightness (that is, perceived lightness (or gray shade) varies linearly with digital image code value). Consequently, there will be no loss of small-signal and strong-signal structures, and all the image information will be recorded and clearly displayed.

0007FIG. 1 is a block diagram of one embodiment of the system of the present invention.

0008FIGS. 2-4 are graphical views useful in describing the calibration technique of the present invention in determining a LUT for a video printer.

0009FIG. 5 is a graphical view of a representative LUT for a video printer.

0010FIGS. 6-8 are graphical views useful in describing the calibration technique of the present invention in determining a LUT for a video monitor.

0011FIG. 9 is a graphical view of a representative LUT for a video monitor.

0012FIG. 10 is a block diagram of the embodiment of FIG. 1 including LUTs determined by the present invention.

0013Referring now to the Figures, there will be described a preferred embodiment of the present invention. Although the invention will be described with respect to a cathode ray tube multiformat camera (CRT MFC), the invention is also applicable to other types of electronic printers, such as laser-film printers, xerographic laser printers, ink-jet printers and thermal dye transfer printers, which use various types of recording media including paper and film. The digital images to be reproduced may come from any digital image source, such as medical imaging modalities (ultrasound (US), nuclear medicine (NM), computerized tomography (CT), digital subtraction angiography (DSA), magnetic resonance imaging (MRI), computed radiography (CR), digital fluorography (DF)), x-ray film digitizers and archived digital images.

0014FIG. 1 is a block diagram of an embodiment of the present invention. As shown, CRT MFC 10 and video monitor 22 are calibrated to provide equal visualization for equal image value changes over the visual range of each device. A video signal generator 12 produces a test image such as a step tablet of graduated gray scale values. The test image is applied to CRT MFC 10 and video monitor 22. The CRT MFC exposes film 14 which is developed by film processor 16 to produce hard copy (developed film) 14'. Densitometer 18 measures density values of copy 14' which are sent to computer 20 to produce a first look-up table LUT₁. Photometer 24 measures gray scale values on the test image displayed on monitor 22, which are sent to computer 20 to produce a second look-up table LUT₂.

0015Calibration of CRT MFC 10 and video monitor 22 is effected as follows.

0016At the customer site, the CRT MFC 10 is initially set up by a technical representative from the manufacturer to obtain an approximation of the desired "look" or gray scale (D_min, D_max, and density gradation) on film. This is done by printing a gray scale test pattern supplied from a video signal generator 12 and by adjusting brightness (B), contrast (C), and density (D) of the camera 10 settings. The resulting B, C, and D values (B₀, C₀, and D₀) are saved for later use by the calibration algorithm.

0017Now the camera 10 is calibrated as follows. <ul id="ul0001" list-style="none"><li>(1) Expose gray scale test pattern (for example, code value (CV) = 0 to 255 with increment of 17) using the above C₀ and D₀ settings with various values of B (for example, from 20 to 80 with increment of 10). The test pattern images can be exposed on a single film 14 since camera 10 is capable of exposing plural images on film 14 (such as 4, 6 or 9 images).</li><li>(2) Process film and measure film density for the darkest level (corresponding to CV = 0 for a positive image).</li><li>(3) Use FIG. 2 to find the new B setting (B') to give the desired D_max.</li><li>(4) Expose the same test pattern with camera 10 on new film 14.</li><li>(5) Process new film 14 in processor 16 and measure film density of new processed film 14' with densitometer 18 for each gray level. Calculate film density by computer 20 for other code values using cubic spline or other interpolation techniques.</li><li>(6) Convert density values into perceived lightness values by using the visual perception characteristic curve of FIG.. 3, which is calculated from Eq. (1).</li></ul>

0018Finally, (7) construct a look-up table (LUT) from the measured and desired tone curves (FIG. 4). The resulting LUT₁ is shown in FIG. 5. All calculations are carried out by the computer 20.

0019Similarly, the video monitor 22 is calibrated as follows. <ul id="ul0002" list-style="none"><li>(1) Vary the brightness and contrast controls of the monitor 22 to obtain the desired minimum luminance (L_min) and maximum luminance (L_max) which correspond to code values CV=0 and CV=255, respectively. Luminance from the monitor 22 is measured by a spot photometer 24.</li><li>(2) Measure monitor 12 luminance for each level of the same gray scale test pattern (CV=0 to 255 with increment of 17). Calculate monitor 22 luminance for other code values using cubic spline or other interpolation techniques (FIG. 6).</li><li>(3) Convert monitor 22 luminance (L_m) into perceived lightness (S_m) by using the visual perception characteristics (FIG. 7), <maths id="math0002"><math display="inline"><mrow><msub><mrow><mtext>S</mtext></mrow><mrow><mtext>m</mtext></mrow></msub><msub><mrow><mtext> = c₁(L</mtext></mrow><mrow><mtext>m</mtext></mrow></msub><msup><mrow><mtext>)</mtext></mrow><mrow><mtext>p</mtext></mrow></msup><mtext> + c₂   (2)</mtext></mrow></math><img file="EP0542012A1_D0002.tif" /></maths> where c₁ and c₂ are normalization constants and p is close to 1/3.</li></ul>

0020Finally, (4) construct a LUT from the measured and desired tone curves (FIG. 8). The resulting LUT₂ is shown in FIG. 9.

0021Referring now to FIG. 10, there will be described the use of LUT₁ and LUT₂ in the reproduction of a digital image on film that has the same gray scales as displayed on a video monitor, both of which are matched to human visual perception characteristics. When a digital image from a digital image data source 26 is printed out, the image data are first corrected by LUT₁ 28 on a pixel-by-pixel basis, then converted into analog signals by a digital-to-analog converter (DAC) 30, and finally fed to the CRT MFC 10 to produce an exposed film 14. Film 14 is processed in processor 16 to produce hard copy film 14'. The resulting image on film 14' will have a tone scale with uniform quantization in perceived lightness (that is, perceived lightness varies linearly with code value).

0022Similarly, when the same digital image is displayed on monitor 22, the image data from source 26 are first corrected by LUT₂ 32 on a pixel-by-pixel basis, then converted into analog signals by a DAC 34, and finally fed to the video monitor 22 to produce an image on the monitor screen. Like the film image, the resulting screen image will have a tone scale with uniform quantization in perceived lightness.

0023Although the film 14' has a higher dynamic range than the video monitor 22, the step size of the calibrated tone scale on film, ΔS_f, can be made to closely match the corresponding monitor's step size, ΔS_m, by adjusting the luminance of the view-box on which the film is viewed.

0024With the proposed calibration technique, the resulting gray scales on film and on monitor, which are well matched to the human perception characteristics, provide equal visualization for equal image value changes, and avoid the loss of information both at the low end and high end (in image value) of the image. Also, the invention is applicable to the reproduction of color images, in which a color CRT monitor is used to display the image and a color printer (for example, an ink-jet printer or a thermal dye transfer printer) is used to print the image. The invention will provide an optimum luminance scale in reproducing the image on the color monitor and on the recording media.

0025The present invention has application in digital image processing, such as in the reproduction on film of a medical diagnostic digital image viewed on a video monitor.