JPS63192179A - Mark reader - Google Patents

Abstract

PURPOSE:To eliminate a mechanical movable part, to shorten an image pickup time and to improve reading reliability by applying an enhancement processing to a picture signal from a television camera and binarizing the enhancement processed signal by a set threshold. CONSTITUTION:A picture processing part is constituted of an enhancement processing part 20, for emphasizing a character part to the image up by a CCD camera (photoelectric element type) 19, a picture storing memory for storing the processed picture, a binarization circuit 22 for binarizing video signal, a binarized picture memory 23 for storing the binarized picture and a microcomputer 25 for making access to the memories 21, 23. Accordingly, the image of two dimensional image is picked up under a stationary state, a signal level difference from a background part a threshold level for the binarization is easily set, the unevenness in the picture of a mark contour part is reduced and an erroneous decision ratio is reduced. Thereby, the stability in a binarization image processing is improved and a reading with high accuracy is executed.

Description

[Detailed Description of the Invention] [Industrial Field of Application] This invention is applicable to mirror-like wafers or exposure devices in semiconductor manufacturing processes. The present invention relates to a device that is often used when capturing an image of a target object such as a soap and reading characters, codes, or other marks displayed thereon.

[Conventional technology]

Fig. 4 shows this type of device conventionally used, and Fig.
A rectangular optical fiber illumination device 2 that illuminates a target object 3 from a certain angle as shown in the figure, and a target object 3
1':) Line sensor 1 for in-imaging, X-Y table 4 for vertically moving the target object 3, mechanism control section 5 for operating this X-Y table 4, information sent from line sensor 1 It is comprised of a character recognition section 6 for recognizing characters from text, and an R8232C type communication line 7 for communicating data between the mechanism control section 5 and the character recognition section 6.

Next, the operation will be explained. x-Y by mechanism control unit 5
Operate table 4 and arrow target object 3! After being conveyed to the reading position as shown in , a reading start signal is sent to the character recognition section 6 through the signal line 7. After receiving this signal, the character recognition unit 6 scans the line with the light/sensor 1 and reads the character 8. 9 indicates the field of view. Thereafter, a table movement signal is sent to the mechanism control section 5, and the next lie is scanned in the same manner. This operation is repeated until the image of the read character is imported into the memory. Next, a method of processing scan signals will be explained using FIG. 6.

Assume now that a scan signal as shown in FIG. 6(a) is obtained. By differentiating this signal, a differential signal corresponding to the signal level displacement amount ΔLK with respect to the minute time displacement amount ΔtK is obtained. This is shown in the same figure (b).

This differential signal is further applied to a constant level threshold value Ltht.
- Binarize as a reference to obtain a binarized signal as shown in FIG. 4(c). The above processing is executed by dividing one line time by t-256, and the results are input into the binarization memory built into the character recognition section 6.

After the character t is recognized by processing the data taken in by the above processing, the result is sent to the mechanism control section 5 via the communication line 7.

[Problem that the invention seeks to solve]

Since the conventional character reading device is configured as described above, it requires an X-Y table for moving the target object, which complicates the device configuration and takes time to take an image.

Further, there were other problems such as noise from the mechanical section affecting the communication section, resulting in communication errors, and disturbance noise easily affecting the binary signal as shown by A in FIG.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide an apparatus that has no mechanically movable parts, can shorten the image capturing time, and has high reading reliability.

[Means for solving problems]

The mark reading device according to the present invention utilizes a television camera, a means for performing enhancement processing (contour enhancement processing) on an image signal from the television camera, and a means for performing binary value processing based on a histogram based on the brightness of the enhanced image signal. The present invention is provided with means for setting a threshold value for encoding and means for binarizing the enhanced image signal using this threshold value.

[Effect]

By using a television camera, two-dimensional images are captured while remaining stationary. In addition, it is possible to emphasize the image signal of the mark part, since the signal level difference between the mark part and the background part is large.9, it becomes easy to set the threshold level for binarization, and it also makes it easier to set the threshold level for binarization. 〇 [Example] This invention will be explained in detail below. In Figure 1, C
The CD camera (photoelectric device type camera) 19 is for taking an image of the characters on the target object 3, and is for taking an image two-dimensionally, whereas the line sensor 1 is taking an image one-dimensionally. Next, an enhancement processing unit 20 that emphasizes text in the image captured by the CCD camera 19, an image storage memory 21 that stores the processed image, and a binarization circuit that binarizes the video signal. 22, a binarized image memory 23 that stores binarized images, and a microcomputer 25 that accesses the memory 21 and 23 through the multipath 24.

Next, the operation in the enhancement processing section 20 will be explained using FIG. 2. A video signal as shown in FIG. 4(a) obtained from the CCD camera 19 is differentiated with respect to the time axis to obtain a signal as shown in FIG. 2(d). Next, this differential signal is inverted and amplified to obtain a signal as shown in FIG. , the original video signal (
This is the edge emphasis signal in FIG. 2(a). Therefore, adding this second-order differential signal to the video signal K
As a result, an enhanced signal as shown in FIG. 4(e) is obtained.

The effect of performing such enhancement processing will be explained using FIG. 3. 2(a) shows a video signal corresponding to FIG. 2(a)K, and FIG. 2(b) shows an enhanced signal corresponding to FIG. 2(e). In contrast, the same figure (e)
, (d) shows histograms according to the brightness of each dot of the stored image in each case.

As shown in FIG. 4C, when no enhancement processing is performed, the distribution of character parts is not uniform compared to the brightness distribution of the background. Therefore, it is not easy to set the threshold Lth for binarization. On the other hand, in the same figure (d)
As is clear from the above, after the enhancement processing, the distribution is divided into the distribution in the background and the distribution in the character part, and it is easy to set the threshold value. That is, an appropriate level Ltht'' can be selected within the range from Ltbl to Lth2.

In this way, a threshold value is set from the brightness distribution after the enhancement processing, and the binary conversion circuit 2
Binarization is performed in step 2, and character recognition is performed using the binary image. Even if there is noise as indicated by A in FIG. 2(&), it does not affect the binary signal as is clear from Ge) in the same figure.

In the above embodiment, the characters were enhanced to improve their reading reliability, but by performing the same process with a wafer pattern defect and dust adhesion recognition device, defects and dust adhesion areas were removed. It is emphasized and produces the same effect as the above embodiment.

〔Effect of the invention〕

As described above, according to the present invention, mechanically movable parts are eliminated, and effects such as miniaturization of the optical part and reduction of noise can be obtained compared to using a television camera. Further, by performing the enhancement process, the stability of the binarized image processing is increased, and highly accurate reading can be performed.

[Brief explanation of the drawing]

1 to 3 are diagrams showing an embodiment of the present invention, in which FIG. 1 is a configuration diagram of a reading device, FIG. 2 is a waveform diagram for explaining the enhancement process, and FIG. 3 is a waveform diagram for explaining the enhancement process. A waveform diagram and a histogram to explain the effect, Fig. 4 is a configuration diagram showing a conventional example, Fig. 5 is a plan view showing an example of the target object, and Fig. 6 is a waveform to explain the operation of Fig. 4. It is a diagram. 3...Target object, 19.Fist...CCD camera, 20
... Enhancement processing unit, 21 ... Game image storage memory, 22 ... Binarization circuit, 23 ... Binarization image memory, 25 ... Microcomputer -0

Claims (1)

[Claims] a television camera that captures an image of a mark attached to a target object;
means for performing enhancement processing on the image signal from the television camera; means for creating a histogram of the brightness of the enhanced image signal; and means for setting a threshold for binarization from this histogram; A mark reading device comprising: means for binarizing the enhanced image signal; and means for recognizing marks from the binary image.