JPS58736A - Evaluating method for image - Google Patents

Abstract

PURPOSE:To determine the response of square wave gratings easily without requiring test charts for square waves, etc. by irradiating a diffusion surface light source to an edge, inputting the image thereof to an optical system to be measured, measuring the inclination of the response thereof and operating the same. CONSTITUTION:A diffusion surface light source 15 is irradiated to a knife edge 16 to make an edge image. This image is inputted to an optical system 17 to be measured, and the output image thereof is scanned at a right angle to the edge direction by a one-dimensioal scanning system 18, whereby an edge response curve is determined. The inclination of the rising straight line thereof is determined and an analog output is obtained. After this output is digitized with an A/D converter 19, it is operated with a computer 20 to determine the optical transmission function of the system 17. In accordance with this, the response of a square wave grating is determined and the figure is outputted to an X-Y plotter 21. Therefore, the image is easily evaluated simply by measuring the inclination of the edge response without using test charts for square waves and sine waves.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image evaluation method that can easily obtain a rectangular wave grating response of an arbitrary period of a measured system such as an m-image transmission line.

FIGS. 1 and 2 show examples of conventional IIII image shields, respectively. In both conventional examples, the OT F of the optical system to be measured
(Optical Transfer Function
n; optical transfer function).

In the conventional example shown in FIG. The image is photographed on film 5 through the camera. At this time, the optical system 3 to be measured is moved randomly, for example, by rotational movement.

By scanning the image of the film 5 with the densitometer 6,
A drum shape 7 representing the record end 6cKOTF is recorded. Note that 6a is a light receiving element, and 6b is an amplifier.

In the conventional example of the second core, the light source 8. Lens system 9 and diffuser plate l
The image of the slit plate 11 is formed on the optical system 3 to be measured by O, and this slit image is formed on the sine wave drum 13 via the lens system 12. The sine wave drum 13 is made by winding a sine wave test chart 13a whose spatial frequency changes sequentially around a drum 13b. While rotating this sine wave drum 13, the light receiving device 14 receives the superposition of the slit image and the sine wave, and the 91 superposition becomes a 7-layer variation of the slit image, so this output itself is OT
Dress F.

However, the rectangular wave test chart 1 in Fig. 1 and the sine wave test chart 13a in Fig. 2 require gradual changes in their spatial frequency and require high precision, so they are very difficult to manufacture. It is difficult and extremely costly.

In view of the above-mentioned prior art, the present invention makes it possible to easily evaluate optical systems such as m-image transmission lines without using rectangular wave or sine wave test charts that are difficult to manufacture and expensive.
The overall purpose is to provide an image evaluation method. Therefore, in the present invention, an expanded WI surface light source is irradiated onto the naive edge to create edge #, this edge image is input to the optical system to be measured, and this output image is scanned perpendicular to the edge direction to create edge response curve I.
i5! The slope of the rising straight line of this edge response curve is calculated, and the rectangular wave grating response is calculated from this slope as an inverse value or a graph.

FIG. 3 shows a block configuration of a rectangular wave grating response measuring device constructed by applying the present invention. In FIG. 3, reference numeral 15 indicates a diffused surface light source 1. For example, as shown in the second factor, light source 8. Realized by a combination of parallel light lens system 9 and diffuser plate 10,
Generates uniform light. Reference numeral 16 denotes a knife with Tor. By illuminating the back of the knife 16 with a diffuse surface light source 15, an edge image represented by a step function U (X) shown in FIG. 4 is created. However, xVi is a coordinate in a direction perpendicular to the edge direction within the edge plane, that is, in the X direction, and the edge image is normalized as follows.

Is 17 the optical system to be measured? +, Outputs an edge response image expressed by the response function e (x) shown in FIG. 4 with respect to the input edge image. In general optical systems such as lens systems, photographic recording materials, or original transmission lines, the edge response function e(x) can be estimated experimentally.

It can be modeled as Since the edge response image is the input edge image changed by the OTF of the optical system 17 to be measured, if the above edge response model is used, the OTF can be determined by knowing the rise coefficient α of Equation 1 (2). This coefficient α is the tangent line 10#4 of the edge response function e(x) at X=Q
I want more from you.

α 2　　　　　　　　　　　　°Formula (3) There is a relationship between

Reference numeral 18 denotes a photoelectric conversion system that scans the output image from the optical system to be measured 17 at right angles to the edge direction.

For example, a one-dimensional scanning system is used in which a photoelectric conversion element with a minute pinhole or slit at its tip is moved in a straight line, and analog No. 16 with the same change as the function e (x) in Figure 4 is output as it is scanned. do. Output from photoelectric conversion system 18 J! If we differentiate r with respect to x-〇, we can find the steepness m of tangent IIMt in Fig.
After being digitized by a digital converter 19, it is processed by a computer (CPU) 20 for digital calculation.

By the way, if the impulse response of the thread consisting of kh (X) is expressed, then the OTF of the system, H(ω), is:

H(ω) = / h(x)e" dx -, expressed by equation (4). The edge response model of creation (2) shows that the nine-pulse response is b h(x) ", 1. e
(x) and α in equation (3) can be found by α = 2 m.

becomes. By substituting this into equation (4), the OTF tr! of this optical system to be measured 17 is obtained. .

FIG. 6 shows the OTF of equation (6), and if necessary, the computer 20 causes the X-Y plotter 21 to output the figures.

When the OTF of the optical system 17 to be measured is determined according to equation (6)', the viewer 20 calculates the square wave grating response from this OTF, and outputs the result to the 1x-YfH vine 21 as a figure, or The printer 22 is made to output numerical values of contrast and resolution.

To explain the calculation, first consider the response to a single rectangular wave. The input single rectangular wave is a function r1<x) with an amplitude of - and a transition point of X-±Σa as shown in Figure 7-1.
It is thought that.

It is expressed as The 7-Rie transformed form of this single rectangular wave rl(X) is (*)=F(rl(x)).

, ω mn Σ R1(-) □ ...Since formula (8),
The response ft(x) of the optical system to be measured 17 to a single rectangular wave is determined by the inverse Fourier transform using the OTF' of equation (1) and equation (6).

f 1(X) -F-(Rt (4') H('') ]
That's what I find. fs (X) in Fig. 7 represents the approximate t- of this equation (9). The response f, (x) to a general rectangular wave r~(X) as shown in Fig. 8 is I This is because it can be considered as a superposition of ±np (where n is an integer and p is a pitch) of the response fs (X) to the shaped wave fl (X).

It is expressed as This 2〇(l is the square wave grating response of the optical system to be measured 17, and 1. Computer 20 Vc and p formula (
9) and the equation (1(l) are calculated, and graphically displayed as f.o(X) in FIG. 8 using the X-Y plotter 21 if necessary.

From the square wave grating response $f,(X), the contrast and resolution are determined as follows.

The contrast of image 1←) C(i (x) ) is expressed by a linear equation 〇υ using a maximum value of 1 m□ and a minimum value of 1-t-.

i wax −1= C(1(x))−1 i5]−1 ・・・2〇υ Therefore, the contrast C(f−(
x)) is given by. However, Type 1 (9) and 2〇〇.

fm=Σft(No.-np) n-10-xx-asut(ma) as scotth(mp)
Set I, for simplicity, the duty is 0.5
If we consider the rectangular wave of
p) ... Formula C1612 is obtained.

By substituting the equations and the equations α◆ or the equations αS and the equations αer into the equations α, the contrast for the input rectangular wave of any pitch can be found by knowing “mmt”.

A numerical value is output to the printer 22. Note that the calculation for calculating the contrast is based on the contrast curve "C (:
P ) is calculated in advance and prepared as a table, and it is preferable to read out a control 2 strike of an arbitrary pitch corresponding to 1 m from the table.

Next, resolution will be explained. Let C* be the allowable contrast value, and let P-2mp that satisfies C* be P*.

Once the electric current is determined, P is determined from the contrast curve shown in FIG. 9, and by calculating the following formula αn from the previously determined m.

is output numerically.

Alternatively, the contrast curve C(
f, (x)) can be expressed as shown in FIG. Using this car f't table, determine the resolution of the optical system to be measured 17 from the determined m and contrast tolerance C*. − Can be read.

Of course, the contrast ct- for a rectangular wave of an arbitrary pitch can be calculated from the previously determined m using the contrast curve in diagram h#110.

The slope of the edge response of the optical system to be measured 17, showing a numerical example.

Assuming that m=50, the contrast CFi for a square wave of p=0.05m, (a) In Fig. 9, P=2mp =2X50X0.
From 05=5.

C=0.69 Than.

C=0.69. On the other hand, since the allowable contrast C* is C*=0.33 (1) P*=2 in FIG. 9, the formula an
yo(b) In Figure 10, m;50 curve and C*=0
．． From 33.

becomes.

As explained above in detail, one aspect of the present invention is as follows.

The measurement can be carried out by simply irradiating the edge with a diffuse surface light source, inputting this edge 1# into the optical system to be measured, and measuring the slope of the response, so there is no need for a square wave or sine wave test chart as in the past. It looks strange. Also, by simply measuring the slope of the edge response.

Since many of the necessary characteristics can be determined through calculations, responses to rectangular wave gratings of any period (pitch) can be obtained freely. As described above, the present invention is extremely effective for image evaluation in general, including general optical systems such as lens systems, and resolution evaluation devices for various types of image transmission paths.

[Brief explanation of drawings]

Figures 1 and 2 are block diagrams showing examples of conventionally used OTF measuring devices, Figure 3 is a block diagram of an example of a device implementing the present invention, Figure 4 is a graph of edge response, and Figure 5 is a diagram showing the configuration of an example of a conventional OTF measuring device. Graph of in/malus response, Figure 6 is the graph of OTF, Figure 7
Figure 8 is a graph of the nest-square wave response, Figure 8 is a graph of the square wave grating response, and Figures 9 and 10 are graphs showing contrast curves. In the drawing. 15 is a diffused surface light source. 16 is Naifetsuji. 17 is an optical system to be measured. 18 is a one-dimensional running system. 19 is an analog/digital converter. 20 is a computer. 21 is the X-Y7" rotor. 22 is the printer, u(x) is the input edge image. e(x) is the modeled edge response - WA. f, , , (x) is the square wave grating response curve. m is the Slope of the edge response. p is the pitch cH contrast of the square wave. Patent applicant Sumitomo Electric Industries Co., Ltd. Agent Patent attorney Tsutomu Mitsuishi (and 1 other person) Figure 1 Figure 2 Figure 4/ Figure 5 Figure 7

Claims (2)

[Claims] (1) An edge image forming system that inputs an edge image illuminated from behind by a diffuse surface light source into the optical system to be measured, a photoelectric conversion system that scans the output image of the optical system to be measured perpendicular to the edge direction, and a single source The slope of the rising straight line of the output image is calculated based on the output signal from the conversion system, and the optical transfer function of the optical system to be measured is calculated from this slope and a predetermined edge response function model.
Single square wave response characteristic, arbitrary period square wave grating response characteristic,
An image evaluation method that consists of a calculation processing system that calculates optical characteristics such as contrast and resolution. (2) Claim 1 comprising an output display system that displays the optical characteristics in numerical values, graphs, etc.
Image evaluation method described in section.