JPS5913903A - Position detecting method by image pickup element - Google Patents

Abstract

PURPOSE:To detect an image formation position with precision more than the resolution of an element, by moving a body to be measured until the logical level of a binary-coded signal is inverted, and measuring the moved distance with resolution more than the detection resolution of the element. CONSTITUTION:The output signal (video signal) of a photodiode linear array 1 is coded by a binary coding circuit 27 into a binary signal and the binary- coded signal of the display of a specific element is read by a signal reading circuit 28 and stored in a storage circuit 29. A driving circuit 30 indicates the moved direction and extent of a driving table 26, which is moved through a pulse motor driver 31. The signal reading circuit 28 reads the binary-coded signal of the specific element display again after specific-extent driving (e.g. 0.5mum). A binary-coded signal read by a CPU32 previously and the binary-coded signal read after the driving are compared with each other to decide on a position where the logical level of the binary-coded signal is inverted. Simultaneously, the movement extent required for the inversion is counted to calculate an error value to find a real position. This method improves the detection resolution which is limited by element pitch.

Description

DETAILED DESCRIPTION OF THE INVENTION FIELD OF INDUSTRIAL APPLICATION The present invention relates to a method for improving the resolution of T-piece position detection using an imaging probe in the field of photoelectric measurement.

Conventional configuration and its problems Photoelectric measurement has been widely used in various fields in recent years. A typical example is one that uses a photographic element to measure the position (dimensions) of an object based on the difference in reflectance on its surface.

The image sensor is a sensor that uses the photoelectric effect to apply a load according to the brightness of incident light and detect it as an amount of electricity.

An example of an application of the image pickup device is a photodiode linear array in which a plurality of image pickup devices are arranged in a straight line. As shown in Fig. 1(a), a photodiode linear array (1) with 512 elements (photodiodes 5 (2) spaced at intervals of 28 μm and 512 elements) is currently in practical use. If the bright and dark images (bright area (3), dark area (4)) caused by the difference in ratio are focused on the photodiode linear array (1) using an optical lens, the individual elements (2) will be as shown in Figure 1. One output signal (δ) (video signal) is generated according to the amount of incident light as shown in (b).This output signal (5) is binarized using a predetermined threshold value (6), and the A binary signal @-(7) as shown in Figure 1(c) is obtained.In this binary signal (7), for example, the dark part (4) of the object to be measured is detected.
) position and width can be measured. The details can be known from the element (T, for example, the element (9) indicating the b LOW level) indicating the logical inversion position (8) of the binary signal (7). The width can be measured by the number of elements that exhibit the same logic.

FIG. 2 shows one conventional embodiment using a photodiode linear array. Photodiode linear array (1)
An optical lens α is positioned between the object to be measured OQ and the image of the object to be measured OQ to the photodiode linear array (1).
It forms an image. Photodiode linear array (1
) is subjected to predetermined measurements using the procedure described above. 03 is a binarization conversion circuit, @ is a signal readout circuit, α→ is a CPU% (to) is a display section.

The object to be measured Q1 in this case is a thin film having a predetermined reflectance formed on a substantially optically transparent substrate.

Specifically, it measures the pattern edge position and pattern width of a photomask, color filter, etc. As mentioned earlier, the detection resolution of the currently put into practical use is 28μ per element spacing.
The result of combining a photodiode linear array with 512 m5 elements and an optical lens is about 5 μm.

This detection resolution is the element l! determined by the l interval.

This will be explained using FIG. Bright part of the record of objects to be measured (3
) and the dark area (4) are placed in one element,
The output signal a obtained by the element is generated at an intermediate level between the white level indicating the bright area (3) and the black level indicating the dark area (4). No matter which level the threshold value α is set between the white level and the black level, the resulting binary signal (total) can only be detected in units of element intervals. If a boundary between bright and dark edges exists within one element, an error within the element interval will inevitably occur.

In the measurement of ordinary photomasks, color filters, etc., problems arise when the detection resolution is 5 μm. This is because a detection resolution of about 0.6 .mu.nl or 1 .mu.m is currently required. The first possible measure to improve this resolution is to increase the optical magnification. There is a natural limit to increasing optical magnification, which also poses problems in terms of workability.

In other words, the working distance (W, D) and depth of focus become shallow, which causes many practical problems. Therefore, it is desired to improve the detection resolution by reducing the depth of focus due to the long working distance and deep focus, which are the advantages of the photodiode linear array.

Therefore, if the conditions limited by the optical lens magnification are realistic and practical, and r = , then one possible way to improve the detection resolution is to reduce the element spacing of the photodiode linear array. First, a feature of the photodiode linear array is that the element spacing is small. However, the current element spacing is 28 voices.
It is difficult to stably manufacture anything smaller than this with current semiconductor technology.

OBJECTS OF THE INVENTION The object of the present invention is to provide a detection method that can obtain a resolution higher than the detection resolution limited by the element spacing in 1 (dimension) detection using an image sensor.

Structure of the Invention In order to achieve the above-mentioned objects, the present invention forms an image of an object to be measured on an image sensor assembly made up of a plurality of light receiving elements that produce differences in output signals due to the photoelectric effect. In a method of converting an output signal of each element constituting a binary signal into a 2 (Ill) signal and detecting a boundary position between a High level and a Low level, either of the two elements constituting the boundary of the binary signal On one side, the object to be measured is moved in the positive or negative direction with respect to the scanning direction of the image sensor assembly until the logic of the binary signal is reversed, and the moving distance is set at a resolution higher than the detection resolution of one element. It measures the length and detects the image formation position on the element of the object to be measured from the moving distance until reversal.This has made it possible to detect the image formation position with an accuracy higher than the resolution of the element. It's a small thing.

Description of examples Embodiments of the present invention will be described below based on the drawings.

First, the principle will be explained with reference to FIG. As shown in Figure 4G), the boundary between the bright part (3) and dark part (4) of the object to be measured is -
In this case, as shown in Figure 4tb), the output signal V obtained from that element is determined by m of the white level and black level.
occurs in l. This output signal is binarized at a threshold (E) at a level close to the black level as shown in Fig. 4 (C), and
Obtain the valued signal ■υ. Here, binarize on the black level side and T
This is because the black level is more stable, and the white level must be adjusted to avoid fluctuations due to differences in illumination and surface conditions of the object to be measured.

Two elements (a) that constitute the boundary of the binarized signal Qυ (
Focus on one of (a), such as the element (a).

The first position detected by the cylindrical iron is the true boundary (c)
is displayed with an error of X. In this state, the binary signal of ) is stored in memory in the element. In other words, perform the same measurement as before, and select 2 (memorize whether the III conversion signal is High level or OW level.Here, it is the element).The reason for this is that the black level side is binarized. Then, there is an error and one element always shows a high level. As a result, when the black level side is binarized, attention is paid to the element showing the High level.

Furthermore, the binary signal to be stored is always at the Hlgh level. It is binarized on the white level side, and the opposite is true.

Next, the object to be measured is moved in the A direction by a predetermined pitch (for example, 0.5 μm) as shown in FIG. 4(a). The direction of movement of the object to be measured is determined by setting the threshold value to either the black level side or the white level side, or by setting the binarized signal to High.
It is possible to give either a positive or negative direction to the scanning direction of the photodiode linear array (1) depending on whether it is converted from Low to High. A known pulse motor drive is used as the moving means. With the current pulse motor drive, 05- feed can be easily achieved. As the sub-length means, it is sufficient to count the number of generated pulses. Alternatively, a linear scale may be used in combination for continuous movement.

In that state, observe the binary signal level of the element (a),
First, it is stored in memory and compared with the binary signal level.

This movement and comparison is repeated to obtain a binary signal that differs in logic from the initial binary signal. In FIG. 4, the high level is inverted to the low level. The logic of the binary signal is reversed, the movement of the object to be measured is stopped at the T position, and the sum of the movement amounts is counted. (T), the boundary between brightness and darkness of the image of the object to be measured moves to the neutral position of the element (C) and (C), as shown here.

As a result, it is necessary to invert the logic of the binary signal, and the T movement My
If the one-element detection resolution 2 is clear, the boundary between brightness and darkness of the object to be measured can be calculated. In this case, the one-shot error value X whose movement direction is negative is obtained by X-roz-y. If there are 5 conditions for movement in the positive direction, the error value X will be x=y. Therefore, by subtracting the error value from the f value measured by the element indicating the boundary of the binary signal (T) described above, the true boundary position (c) can be found.

1, an error occurs due to the level difference between the previously set threshold value (e) and the black level in this case. In this first step, the threshold value (A) must be kept as close to the black level as possible.

As a result of the above, it is now possible to measure positions and dimensions with high precision without being affected by the element spacing.

An embodiment according to the present invention is shown in FIG. The object to be measured θQ is held on a drive table (c). Measured object OQ
An optical lens (+) is located between the photodiode linear array (1) and the first diode linear array (1), and forms an image of the object to be measured 00 on the photodiode linear array (1).

This binary signal obtained from the photodiode linear array (1) is read and stored in a storage circuit. The drive circuit (to) instructs the drive table to move in the direction and amount of movement, and causes the drive table to move via the pulse motor driver 0υ.

After driving by a certain amount (for example, 0.5 μm), read the signal 9
The circuit (c) reads the binary signal of the predetermined element display again. Then, the CPU compares the previously read binary signal with the binary signal read after pulsation, and the logic of the binary signal is inverted to determine the position. At the same time, the amount of movement required for reversal is counted, the error value is calculated, and the true position is determined. (to) indicates the display section.

Effects of the Invention According to the present invention, it is possible to easily improve the detection resolution which is limited by the strand pitch. Therefore, the conditions limited by the optical lens magnification (working distance exclusion, sunspot holding pressure) can be made realistic, and the great effect of making the detection power and resolution as desired can be obtained.

[Brief explanation of the drawing]

Figure 1 shows conventional photodiode linear array measurement.
FIG. 2 is a configuration diagram showing a conventional example; FIG. 8 is a diagram illustrating the focal point during measurement using a conventional photodiode linear array; FIG. 4 is a diagram showing the present invention FIG. 5, which is a diagram showing the principle of measurement, is a configuration diagram showing an example of application to a photodiode linear array according to the present invention. (])...Photodiode linear array, (3)...
...Bright part, (4)...11d part, 01...Boundary part between bright part and dark part is located within one element lit L f: Output signal obtained by the element when, (e)...・Threshold value, Q
])...Binarized signal, @(E)(C)...Light receiving probe, @...Boundary part, ■...Direction agent for moving the object to be measured at a predetermined pitch Yoshihiro Morimoto Figure 1 - Figure 2 Figure 3 Figure 4''・Figure 5

Claims (1)

[Claims] 1. An image of an object to be measured is formed on an image sensor assembly consisting of a plurality of light receiving elements whose outputs differ depending on the number due to the photoelectric effect, and the image of each element constituting the image sensor assembly is In a method of converting an output signal into a binary signal and detecting a boundary position between a high level and a low level, either one of two elements constituting the boundary of the binary signal converts the output signal into a binary signal. The object to be measured is moved in the positive or negative direction with respect to the scanning direction of the image sensor assembly until the logic is reversed, and the distance of movement is measured with a resolution higher than the detection resolution of one element, and the time required until the logic is reversed. A position detection method using an image sensor having a similar feature, which detects the imaging position on the element of the object to be measured from the moving distance with an accuracy higher than the resolution of the element.