JPS6221011A - Measuring apparatus by light cutting method - Google Patents

Abstract

PURPOSE:To enable the recognition of 3-D shape information and measurement at a high accuracy, by taking an image of an object varying the intensity of a slit light projected thereto gradually to process a plurality of image data obtained. CONSTITUTION:A density image data outputted from a camera 4 is written into an image memory 6 through a fetching section 5. At this point, a CPU 7 varies the intensity of a slit light 2 at three steps for the same part of the same object 3 without changing the position of a light cutting line 2a to memories 6a-6c three kinds of image data obtained from the camera 4 at each intensity of light. The density values of the image data are compared at the image address (x, y) to detect the density value for the brightest level but darker than a specified threshold as representative value at the image address (x, y) and one image data is synthesized by a light cutting line image 2b partially extracted. The synthesized image data is stored into a memory 6b and processed thereby enabling the recognition of 3-D shape information and measurement at a high accuracy.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention 1] The present invention relates to a measuring device that obtains three-dimensional shape information of an object using a light cutting method.

[Technical background of the invention and problems 1 FIG. 5(A) shows an outline of the optical system for the optical cutting method.

A slit light 2 from a light source 1 is projected onto an object 3, and the reflected light is imaged by a television camera 4.

From the television camera 4, a slit light 2 on the surface of the object 3 is emitted.
Bright line 2a (hereinafter referred to as the light cutting line)
m-manufactured image data with gradations including images of .

An example of image data is shown in FIG. 5(B). 2b in this figure is a light section lfl'1 line image corresponding to the light section line 2a. This image data is processed appropriately and
The three-dimensional shape information is obtained.

In order to perform highly accurate measurement using the light sectioning method, it is necessary to obtain a light section line image 2b with a density that can be clearly distinguished from the background and as narrow as possible. However, since the reflectance of the surface of the object 3 differs partially and the distance between the light section line 2a and the camera 4 also differs partially, the density of the light section line image 2b in the image data is not uniform. , its dynamic range is generally very wide.

If the density of the light section line image 2b is too dark, it will be difficult to distinguish it from the density of the background (this will cause erroneous measurements. Saturation occurs in the part, causing blooming, smearing, etc. Then, not only the density of the corresponding part of the light-cut line image 2b becomes brighter, but also the line width becomes thicker, and the object 3
It becomes difficult to accurately detect three-dimensional shape information. camera 4
First, a line thinning process is performed on the image data obtained from the process to reduce the line width of the light-cut line image 2b contained therein.
After that, three-dimensional shape recognition processing is usually performed. As this thinning process, for example, JP-A-56-70
The technique disclosed in No. 407 is known.

No matter what kind of line thinning processing is performed, if the dynamic range of the density of the light sectioned line image 2b is too large, the expected processing result cannot be obtained, and the light sectioned line image 2b may partially disappear or be deformed. The lines may be thinned.

[Object of the Invention] The present invention was made in view of the above-mentioned conventional problems, and its purpose is to obtain image data in which the dynamic range of the density of a light-cut line image is small and uniform at an appropriate level; It is an object of the present invention to provide a measuring device using a light cutting method that can recognize highly accurate three-dimensional shape information and perform 31 measurements.

[Summary of the Invention] A measuring device using a light cutting method according to the present invention includes a decoy image means for obtaining gradation image data by layering a light cutting line generated by a light cutting method using a slit light; A means for temporarily storing a plurality of image data obtained by the image carrying means by changing the slit light intensity for the same part of the same object, and a means for comparing and specifying the density value of the same pixel address of the plurality of image data. The pixel address has means for extracting the brightest density value which is darker than the threshold value of the pixel address as a representative value of this pixel address. FIG. 1 is a block diagram illustrating each of these constituent elements.

[Embodiments of the invention] The slit light source 1 shown in FIG. Slit light 2° light cutting line 2a, object 3. The configuration of the optical system of the television camera 4 is the same in the present invention.

As shown in FIG. 2, the grayscale image data output from the camera 4 is written into the image memory 6 via the capture unit 5, and the computer 7 processes the image data stored in the memory 6 as detailed below. Add processing. The slit light source 1 uses a laser, and the intensity of the slit light 2 is variably controlled by a computer 7 via a control section 8.

The computer 7 calculates the same part of the same object 3 (
Without changing the position of the light cutting line 2a), the intensity of the slit light 2 was changed to three levels: 1-strong, 1-medium, 1-weak, and three types of images were obtained from the camera 4 at each light intensity state. The data is stored in the image memory 6.

As shown in FIG. 2, the image memory 6 has four image memories (a) (b) (
c) (d) matches. The image data with the light intensity "strong" is stored in the image memory (a), and the image data with the light intensity "medium" is stored in the image memory (a).
The image data at "low" light intensity is stored in the image memory (b), and the image data at "low" light intensity is stored in the image memory (C). Examples of these three types of image data are shown in FIG.

In the image data in memory (a), many parts of the light section line image 2b are saturated and thick, but there are no parts that are too dark to disappear. In the image data in the memory (b), a part of the light section line image 2b is saturated and a part has disappeared. In the image data in the memory (C), disappearance is seen in many parts of the light section line image 2b, but there is no saturated part.

In this invention, the best part of the light section line image 2b is extracted from the above three types of image data, and the partially extracted light section line! &2b to synthesize one image data. The composite image data is stored in a memory <d2), and this data is used for recognition and measurement processing of three-dimensional shape information.

In the composite image illustrated in Figure 4 (d), the bottom line is extracted from memory (a), the middle line is extracted from memory (b), and the top line is extracted from memory (C). It is something.

When creating the above composite image data, data is extracted based on the following processing criteria. Compare the density values of the same pixel address of the above three types of image data, and set a predetermined threshold value T.
One density value that is darker than ll (set to a value slightly lower than the saturation level) and is the brightest among them is extracted as a representative value of this pixel address. The processing procedure when the above processing is performed by the computer 7 is shown in the flowchart of FIG.

In steps 100, 101, and 102, the intensity of the slit light 2 is changed in three steps, and three types of image data are stored in the memory (
Store them in a), (b), and (c), respectively.

To create the above-mentioned composite image data, x, y of the image data
The following processing is performed for each pixel on the plane. First, in steps 103 and 104, the pixel address (x, y) to be processed is initialized.

In the next step 105, the density value Va(x, y) of the pixel address (x, y) is read from the image memory (a) and compared in magnitude with the aforementioned threshold Th.

If the density fiiVa(x, y) is smaller than the threshold Th, then the data in the memory <a> is data at a light intensity of "strong", so Va(x, y) is darker than the threshold Th, and other memory (b) (c) Brighter than i9 [
In step 106, the density value Va(
Write the same density value as (x, y) to the pixel address (x, y) of the image memory (d).

Then, in step 110, the pixel address y is updated, and in step 111, it is confirmed that y is not the final address M, and the process returns to step 105 described above.

If it is determined in step 105 that Va (X, V) is larger (brighter) than the threshold Th, this is the memory (a
) means that the S degree value of this pixel address is almost at the saturation level. In this case, the density value Vb (x, y) of the above pixel address (x, y) is read from the image data in the memory (b) obtained with the light intensity "medium",
The magnitude is compared with the threshold value Th. If vb (x, y) is smaller than the threshold Th (118), step 108
The density value and the pixel address (X
, V). If Vb (x, y) is larger than the threshold Th (bright), the light intensity is set to "weak" in step 109.
The density value Vc (x, y) of the pixel address (x, y) of the memory (C) where the image data at is stored is written to the pixel address (X, V) of the image memory (d). This processing is performed up to the final y address M, and further up to the final X address N, processing the entire x and y planes of the image data. As a result, composite image data as shown in FIG. 4 is generated in the memory (d).

[Effects of the Invention] As explained in detail above, in the measurement device using the light sectioning method of the present invention, the intensity of the slit light projected onto the object is changed stepwise to form a silver image, and the obtained plurality of Since the best density part of the light section line image is extracted from the image data of The dynamic range of 12 degrees of the image becomes very small and is almost at an appropriate level.Therefore, three-dimensional shape information can be recognized.

Measurements can be made with high precision.

[Brief explanation of drawings]

FIG. 1 is a block diagram illustrating the constituent elements described in the claims, FIG. 2 is a block diagram schematically showing the hardware configuration of an apparatus according to an embodiment of the present invention, and FIG. 3 is the computer shown in FIG. 2. FIG. 4 is a flow chart showing the operation of the apparatus of the present invention, and FIG. 5 shows the basic configuration (A) of the optical system of the light sectioning method and the image data (B ). 1...Slit light source 2...Slit light 2a...Light cutting line 2b...Light cutting line image 3...Object 4...Delephoto camera 6...Image memory 7...Computer 8 ...Light intensity Ruri i71S Fig. 1 Fig. 2 Prisoner Fig. 4 (a) a (b) medium (C1 weak (d)
Synthesis! 5i11 table

Claims (1)

[Claims] (1) An imaging means that obtains gradation image data by imaging a light cutting line generated by a light cutting method using slit light, and a method that changes the intensity of the slit light for the same part of the same object. means for temporarily storing a plurality of image data obtained by the imaging means, and a density value at the same pixel address of the plurality of image data is compared, and a density value which is darker than a predetermined threshold value and which is brightest among them is determined. A measuring device using a light sectioning method, characterized in that it is equipped with means for extracting a representative value of this pixel address.