JP2000024880A - Outline recognizing method for tool image, tool diameter measuring method, scatter preventing method for tool chip, and devices therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately recognize the outline position of a tool and accurately measure the diameter of the tool by calculating the picture element lines largely changed with the picture element luminance cummulative sums of the picture element lines toward a tool image from a background image, and using them as the outline position of the tool image. SOLUTION: The system of an image processing device is provided with an image processing computer 1, a photographing device 2 photographing a tool T1, and an image input board 3. The computer 1 is provided with an arithmetic process section arithmetically processing the luminance data of received picture elements and a memory storing the program of processing procedures and luminance data. The photographing device 2 is constituted of a lens 4 and a CCD camera main body 5. The lens 4 photographs the side face of the tool T1 from the direction perpendicular to the tool axis together with a background image. The picture element luminance data of the optical image of the tool T1 formed on the CCD camera main body 5 are read by the computer 1 through the image input board 3 for each picture element. The outline of the tool image is recognized from the picture element luminance data, thereby the diameter of the tool T1 is measured.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an NC machine tool such as a machining center for automatically recognizing a tool outer shape from a photographed image of a tool to be replaced and measuring a tool diameter. The present invention relates to a method of recognizing a tool outer shape of a tool image, a method of measuring a tool diameter, a method of preventing a tool chip from scattering, and a device for preventing the scattering of a tool chip by obtaining an allowable rotation speed of the tool.

[0002]

2. Description of the Related Art Binary processing is generally used as a conventional method of automatically recognizing a tool outer shape by such image processing. The binarization process determines a threshold value that is intermediate between the pixel luminance of the background image and the pixel luminance of the tool image. For example, when the background image is brighter than the tool image, and when the pixel luminance is larger than the threshold value, 1 is set to 0 when the size is small, and the tool image is distinguished from the background image.

[0003]

However, the conventional 2
In the binarization process, when the contrast between the background image and the tool image is small, the tool image and the background image cannot be clearly distinguished, and the tool outline cannot be extracted properly.

[0004] When a tool having a throw-away tip such as a milling cutter is used in a machining center or the like, an accident in which the tip is scattered during high-speed rotation has occurred.
There is a need for a technique of measuring the tool diameter in advance and using the tool at a rotation speed at which the chips are not scattered.

However, when a tool is photographed at a tool standby position in a magazine or the like, the contrast between the tool image and the background image is small due to dirt such as oil or dust, and in the conventional binarization processing, the tool maximum diameter is determined. It was difficult to do.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art. It is an object of the present invention to make it possible to more accurately recognize a tool outer shape position and accurately measure a tool diameter. Provided are a method for recognizing a tool outer shape of a tool image, a method for measuring a tool diameter, a method for preventing a tool chip from scattering, and a device for preventing the scattering of a tool chip by calculating a permissible number of rotations of a tool based on a tool diameter that has been accurately measured. It is in.

[0007]

In order to achieve the above object, in a method for recognizing a contour of a tool image according to the present invention, pixel brightness data of a tool image taken from a direction orthogonal to a tool axis is read, and For each pixel column arranged in at least one of the direction parallel to the axis and the direction orthogonal to the tool axis, a pixel luminance cumulative sum of a large number of pixels constituting each pixel column is calculated, and the pixel luminance cumulative sum of each pixel column is set as a background. It is characterized in that a pixel row greatly changed from an image to a tool image is calculated and used as an outer position of the tool image.

[0008] By calculating the pixel luminance cumulative sum of each pixel column, even in the case of an image in which the contrast between the tool image and the background image is small, the contrast is further emphasized and the outer shape can be accurately recognized.

By calculating the pixel luminance cumulative sum of a pixel row parallel to the tool axis, the outer position of the tool image in the direction orthogonal to the tool axis can be determined, so that the tool diameter can be calculated.

[0010] If the pixel luminance cumulative sum of a pixel row orthogonal to the tool axis is calculated, the outer shape position of the tool in the direction parallel to the tool axis can be detected. Are different, the boundary position is known, and the axial length of the tool can be calculated. In addition, the shape pattern of the tool such as which of the cutter portion and the holder portion of the tool image has a larger diameter can be found.

The data of the pixel-sum cumulative sum may be processed as it is, or may be processed by an integer conversion process in which the data is allocated to a predetermined number of integers from 0 to M. In this way, the data is smoothed and the amount of processed data is reduced, thereby enabling high-speed processing.

The change position of the pixel luminance cumulative sum may be differentiated, for example. Since the processing data is a histogram, the differentiation processing is substantially a difference between the pixel luminance cumulative sums of adjacent pixel rows.

The apparatus for recognizing a tool outer shape includes an image input means for inputting pixel luminance data of a tool image photographed from a direction orthogonal to a tool axis, and a tool axis for image luminance data input by the image input means. The pixel luminance cumulative sum of a large number of pixels constituting each pixel row aligned in the parallel direction or the direction orthogonal to the tool axis is calculated, and the pixel luminance cumulative sum of each pixel row changes greatly from the background image toward the tool image. Image processing means for executing a processing procedure for calculating the outer shape position of the tool image by calculating a column.

As the image input means, a tool image may be captured and input by a CCD camera or the like, or may be input from a flexible disk or the like in which pixel brightness data of the tool image has been read, or a captured image may be input by a scanner. May be read.

In the method for measuring a tool diameter of a tool image according to the present invention, pixel brightness data of a tool image photographed from a direction perpendicular to a tool axis is read, and
In order to detect the leading edge of the tool image, a pixel luminance cumulative sum is calculated for each pixel row aligned in a direction orthogonal to the tool axis, and the pixel luminance cumulative sum is calculated from the background image side toward the tool image side in a direction parallel to the tool axis. The tool column extraction including a pixel row in a direction orthogonal to a predetermined number of tool axes arranged on the tool background image side and the tool image side with respect to the tool image tip side based on the tip position of the tool image with the pixel row greatly changed in the tool image. An image processing area is set for the image processing area for tool diameter extraction, and a pixel luminance cumulative sum of each pixel row arranged in a direction parallel to the tool axis is calculated for the pixel luminance data of the image processing area for tool diameter extraction. It is characterized in that a pixel row that changes greatly from the image side to the tool image side is determined as both end positions of the tool image, and the tool diameter is calculated.

The tool has a cutter part and a holder part,
In the present invention, the tool diameter is the diameter of the cutter portion.

According to the present invention, even in the case of an image in which the contrast between the tool image and the background image is small, the contrast is further emphasized and the tool diameter can be accurately measured.

Further, since the image processing area for extracting the tool diameter by detecting the tip position of the tool image is narrowed, image processing can be performed at high speed.

Further, a first image processing area for removing a background noise image around an image frame is set for the read pixel luminance data, and a tip of a tool image is detected in the first image processing area. Then, image processing can be performed at higher speed.

The tool image diameter measuring apparatus according to the present invention further comprises an image input means for inputting pixel luminance data of a tool image photographed from a direction perpendicular to the tool axis, and a pixel luminance input by the image input means. For the data, in order to detect the leading edge of the tool image, a pixel brightness cumulative sum is calculated for each pixel row aligned in a direction orthogonal to the tool axis, and the pixel brightness cumulative sum is calculated from the background image side toward the tool image side. The pixel row that has changed greatly in the parallel direction is defined as the tip position of the tool image, and includes a predetermined number of tool axes arranged in the tool background image side and the tool image side on the basis of the tip position of the tool image and pixel rows in a direction orthogonal to the tool image side. An image processing area for tool diameter extraction is set, and the pixel brightness data of the image processing area for tool diameter extraction is calculated, and the pixel brightness cumulative sum of each pixel row aligned in the direction parallel to the tool axis is calculated. Japanese background painting Characterized by comprising an image processing means for performing a procedure for calculating a tool radius rows of pixels which changes increase from the side of the tool image side determines that the end positions of the tool image.

The image processing means sets a first image processing area for removing the background noise image around the image frame with respect to the read pixel luminance data, and sets a tip of the tool image in the first image processing area. Preferably, a procedure for performing the detection is performed.

In another tool diameter measuring method according to the present invention, pixel brightness data of a tool image photographed from a direction perpendicular to a tool axis is read, and the pixel brightness data is stored in each pixel column aligned in a direction parallel to the tool axis. Calculate the pixel luminance cumulative sum of
A pixel row in which each pixel luminance cumulative sum changes greatly from the background image side toward the tool image side is determined as both end positions of the maximum diameter portion of the tool image, and the maximum diameter is calculated, while being arranged in a direction orthogonal to the tool axis. Calculate the sum of pixel brightness for each pixel column, compare the sum of pixel brightness of the cutter part of the tool image with the sum of pixel brightness of the holder part, determine whether the maximum diameter part is the cutter part or the holder part, and determine the maximum diameter part as the cutter part. In this case, the maximum diameter is determined as the tool diameter.

According to the present invention, even in the case of an image in which the contrast between the tool image and the background image is small, the contrast can be further enhanced and the tool diameter can be accurately measured.

Further, by distinguishing the cutter part and the holder part of the tool image from each other and setting the maximum diameter part to the tool diameter only when the cutter part is larger than the holder part, the subsequent processing can be simplified.

Further, the tool diameter measuring apparatus of the present invention comprises: an image input means for inputting pixel luminance data of a tool image photographed from a direction perpendicular to the tool axis; and a pixel luminance data input by the image input means. Calculate the pixel luminance cumulative sum of each pixel row aligned in the direction parallel to the tool axis, and determine the pixel row where each pixel luminance cumulative sum changes greatly from the background image side to the tool image side at both ends of the maximum diameter portion of the tool image. And calculate the maximum diameter.On the other hand, calculate the pixel luminance cumulative sum for each pixel row aligned in the direction orthogonal to the tool axis, and compare the pixel luminance cumulative sum of the cutter part and the holder part of the tool image to determine the maximum. Image processing means for determining whether the diameter part is a cutter part or a holder part, and executing a processing procedure for determining the maximum diameter as a tool diameter when the maximum diameter part is a cutter part.

In the method for preventing scattering of tool tips according to the present invention, a tool diameter is measured at the tool standby position of the automatic tool changer by each of the above-described tool diameter measuring methods, and the measured tool diameter is set in advance. The tool allowable rotation speed is set from the allowable rotation speed data, and the tool command rotation speed commanded by the NC device is compared with the tool allowable rotation speed. If the tool command rotation speed is larger than the tool allowable rotation speed, a warning is issued. It is characterized by emitting.

In this way, even if the tool in the tool magazine is wrong, the actual tool diameter is always measured, the allowable tool rotational speed is calculated and compared with the tool command rotational speed. Can be prevented beforehand.

Further, the apparatus for preventing scattering of a tool tip according to the present invention comprises the above-mentioned tool diameter measuring devices, and the tool allowable rotational speed based on the tool diameter obtained by the tool diameter measuring device and preset allowable rotational speed data. Setting means for comparing the tool commanded rotation speed commanded from the NC device with the tool permissible rotation speed and issuing a warning signal when the tool commanded rotation speed is larger than the tool permissible rotation speed. It is characterized by.

[0029]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below based on the illustrated embodiment.

[First Embodiment] FIG. 1A shows the configuration of an image processing system in which the present invention is used for measuring tool diameters of various tools used in a machining center.

The machining center 100 is provided with a machining tool T0.
Processing apparatus 102 having a main shaft 101 to which a workpiece is attached
And a table 103 to which the workpiece is fixed, a tool holding magazine 104 for holding the tool T1 to be used, and a tool T0 on the spindle 101 for the tool T1 to be used on the tool magazine 104.
And an automatic tool changer 105 for automatically changing the tool image. The photographing of the tool image is performed at the standby position of the tool T1 to be used.

The system configuration of the image processing apparatus is as follows: an image processing computer 1 as image processing means, a photographing apparatus 2 as an input apparatus for photographing the tool T1, and an image signal from the photographing apparatus 2 to the image processing computer 1. And an image input board 3 to be taken in. The image processing computer 1 includes an arithmetic processing unit such as a CPU for performing arithmetic processing on luminance data of pixels to be taken in a predetermined procedure, and a RAM, ROM, and flexible storage means for storing a processing procedure program and luminance data. A memory such as a disk or a hard disk is provided.

The photographing device 2 includes a lens 4 and a CCD.
And a camera body 5. As shown in FIGS. 1B and 1C, the lens 4 of the photographing device 2 is attached to a stay 104a provided so as to protrude from the tool storage device 104 above the to-be-used tool T1 at the tool standby position. Then, the side surface of the tool T1 is photographed together with the background image from a direction orthogonal to the tool axis. A background plate may be arranged on the tool background to remove image noise.

The optical image of the tool imaged from the lens 4 on the CCD camera body 5 is composed of a large number of lines in a grid by a large number of lines drawn in parallel with the tool axis by the CCD camera body 5 and a large number of lines perpendicular to the tool axis. , And the pixel luminance data is read into the image processing computer 1 through the image input board 3 for each pixel.

Next, the procedure for recognizing the outer shape of the tool image from the read pixel luminance data and measuring the tool diameter based on the outer shape is shown in FIGS.
This will be described with reference to the schematic diagram shown in FIG.

The tool image TI to be detected is shown in FIG.
As shown in (1), it has a holder HI and a cutter CI for attaching a chip, and the tool diameter is the diameter of the cutter CI.

First, the tool is photographed, and pixel luminance data of the photographed image is read (see step 1 in FIG. 2).

In this embodiment, as a condition of the tool image to be read, the luminance value of the background image is larger than that of the tool image, and the tool tip is positioned above the image frame F as shown in FIG. Is set. The image frame F has a rectangular shape, and the coordinates of each pixel are set as coordinate axes with the upper left corner of the image frame F as the origin, the upper side of the image frame F as the X axis, and the left side as the Y axis. The Y axis is parallel to the tool axis, and the X axis is orthogonal to the tool axis.

The pixel luminance data of the input photographed image is data of the entire image area projected in the image frame F of the photographing device 2. As shown in FIG. 4A, the largest area of the data of the entire image area which does not include the image noise of the tool background is set as the first image processing area R1 (FIG. 2, step 2). For example, when a background plate is provided, it is set within the range of the background plate.

Hereinafter, as shown in FIG. 4B, the upper side of the first image processing region R1 is the X axis, the left side is the Y axis, the pixel at the upper left corner is p (x1, y1), and the pixel at the lower right is To p (xn, y
The description will be made assuming that the coordinates have been converted to n).

In the following description, as shown in FIG. 4C, a pixel row parallel to the tool axis is defined as a vertical pixel row lx (a row of pixels having the same X coordinate in parallel with the Y axis). A pixel row in a direction orthogonal to the tool axis is referred to as a horizontal pixel row ly (a row of pixels whose Y coordinates are arranged in parallel with the X axis).

Next, as shown in FIGS. 4B and 4C, the noise N in the first image processing area R1 is removed. The removal of the noise N is performed by, for example, a median filter or the like.

Next, the position of the tip A of the tool image TI is calculated from the pixel luminance data of the first image processing region R1, and the second image processing region as a tool diameter detection range based on the position of the tool tip A is used as a reference. R2 is reset (FIG. 2, steps 3 and 4).

The point of the detection of the tool tip position is to utilize the fact that the background pixel luminance above the tool image TI in which nothing is shown is the maximum value, as shown in FIG. 4C.

As shown in FIG. 5 (a), the detection of the tool tip position is performed in a horizontal pixel row lyi arranged in a direction orthogonal to the tool axis.
[(X1, yi), (x2, yi), ... (xn, y
i)], the cumulative sum Glyi of the respective luminances of the pixels p (x1 to xn, yi) forming each horizontal pixel row lyi is calculated, and a histogram is created as shown in FIG. 5A (FIG. 3). (A), Step 2-1).

As shown in the histogram of FIG. 5A, the pixel luminance cumulative sum Glyi of each horizontal pixel row lyi is obtained.
Is that the background area at the top of the tool tip is the largest, falls sharply at the tool tip position, and the area corresponding to the cutter IC is the smallest,
It gradually increases at the boundary position between the cutter CI and the holder HI, and is larger than the cutter CI in the holder HI and smaller than the upper background area.

The determination of the tip position of the tool image TI is performed by sequentially comparing the pixel luminance cumulative sums downward from the pixel row located at the upper end of the first image processing area R1. It is determined that the horizontal pixel row lya at which the accumulated luminance Gly first largely changes is the pixel row located at the tool tip.

The change in the pixel luminance cumulative sum Gly is calculated by, for example, differentiating the pixel luminance cumulative sum Gly. Since the processing data is a histogram, the differentiation processing substantially takes a difference.

By differentiating the accumulated luminance sum, FIG.
As shown in the figure, there is a zero peak in the tool upper background area, a negative peak at the tool tip (cutter tip), and a cutter CI
It is zero in the area, positive in the boundary area between the cutter part CI and the holder part HI, and zero in the holder part HI area. By calculating the horizontal pixel row lyi at which this negative peak is obtained, the tool tip position is recognized.

Next, based on this tool tip position, an image area including a predetermined number of horizontal pixel rows on the upper tool background image side and the lower tool image side is subjected to the second image processing of the tool diameter extraction amount. The image processing area is reset as the area R2 (FIG. 2, step 4).

The second image processing area R2 is set to increase the processing speed by limiting the processing data to a range necessary for extracting the tool diameter, and is shown in FIG. 5 (c). As described above, the area may be limited to only a part of the cutter part CI, or may be a region including the entire cutter part CI and a part of the holder part HI as shown in FIG.

Here, the tools are shown in FIGS. 4 and 13 (a).
The milling type shown in FIG. 13B has a larger diameter of the cutter portion CI than the holder portion HI, and the drill type shown in FIG. 13B has a smaller diameter of the cutter portion CI.
There are types. Fig. 13
As shown in (a), a tip t is attached to the tool tip. Since the mounting diameter of the tip t is substantially equal to the maximum diameter of the tool tip, the tool diameter is the maximum diameter of the tool tip. In the case of a drill-type tool, as shown in FIG. 13B, since the tip of the cutter portion CI is sharp, the sharp portion C
In order not to mistake I0 as the diameter of the cutter part CI, the diameter of the cylindrical part CI1 formed into a cylinder at the tip of the tool is defined as the tool diameter. Since the tip shape of the drill type can be recognized from the pattern of the histogram of the pixel density accumulation sum Gly of the horizontal pixel row lyi, the range from the tip to the range including the cylindrical portion CI1 area is the second.
May be set as the image processing region R2.

Next, the tool diameter is calculated by calculating the pixel luminance data of the second image processing area R2 (FIG. 2, step 5).

FIG. 3B shows a processing procedure for extracting the tool diameter.

First, with respect to the pixel luminance data of the second image processing area ROI2, the vertical (Y-line) pixel rows lxi of the vertical pixel rows lxi arranged in the direction parallel to the tool axis.
Is calculated, and a histogram is created as shown in FIGS. 4D and 4G (step 5-1).

The pixel luminance cumulative sum Glx has a pattern that is maximum in the background image areas on both the left and right sides of the tool image TI and sharply decreases at both end positions of the tool image TI. From this histogram, the pixel luminance cumulative sum G of each vertical pixel row lxi is obtained.
lxi, lxia, lxi, in which a vertical pixel row lxi greatly changes in a direction orthogonal to the tool axis from the background image side to the tool image TI side
xb is determined to be both end positions of the tool image TI. That is,
The vertical pixel row lxa which first looks at the pixel luminance cumulative sum Glx in the increasing direction from the first column number lx1 of the vertical (Y line) pixel row lxi is used as one end of the tool diameter. The pixel luminance cumulative sum Glx is sequentially looked at in the direction of decreasing from the last column number lxn of the pixel column, and the position at which the first large change occurs is the vertical pixel column lx at the other end of the tool diameter.
b (FIG. 3B, steps 5-2 and 5-3).

The vertical pixel rows lxa and lxb are shown in FIG.
As shown in (e) and (h), it can be calculated by differentiating the pixel luminance cumulative sum Glx. That is, the differential value takes a peak value at the left and right end positions of the tool image.

The tool diameter can be calculated by multiplying the number of pixel rows included between the vertical pixel rows lxa and lxb by the actual pixel row width (preset) (step 5-4).

By calculating the pixel luminance cumulative sum Glx in this way, even in the case of a captured image in which the contrast between the tool image and the background image is small, the contrast is further emphasized and the tool diameter can be accurately measured.

Also, in this embodiment, the pixel luminance cumulative sums Glx, Gly are shown in FIGS. 5 (a), (d),
(G) is schematically illustrated in a simplified manner. However, in practice, since there is a small variation, the process may be performed by an integer conversion process of assigning a predetermined number of integers from 0 to M. In this way, the data is smoothed and the amount of processed data is reduced, thereby enabling high-speed processing.

[Second Embodiment] Next, a second embodiment of the present invention will be described.

As described above, there are two types of tools, a milling type in which the diameter of the cutter portion is larger than the holder portion, and a drill type in which the diameter of the cutter portion is smaller than the holder portion. Is provided only for milling type tools.

In the first embodiment, the second image processing area R2 for extracting the tool diameter is set by judging the position of the tool tip, and the cutter CI
In the second embodiment, first, the maximum diameter of the tool is determined without distinguishing between the cutter part and the holder part. This maximum diameter is the tool diameter of the cutter part CI in the case of the milling type, and the outer diameter of the holder part HI in the case of the drill type. Therefore, the shape pattern of the tool is determined, and in the case of the milling type, the maximum diameter is determined as the tool diameter. Things.

The flowcharts of FIGS. 6 and 7 and FIG.
The image processing procedure will be described in detail based on the image explanatory diagrams shown in FIGS.

First, as shown in FIG. 8A, a tool is photographed and pixel brightness data of an image is read (see FIG. 6, step 11).

Next, as shown in FIG. 8B, noise is removed (see FIG. 6, step 12). The removal of noise is performed by, for example, a median filter or the like. For the drill type, the image with noise removed is shown in FIG.
It is described in (a).

The pixel luminance data is image data including the cutter part CI and the holder part HI, and the image processing area may be data of all images projected in the image frame F of the photographing apparatus. Similarly, the image processing area may be narrowed to remove a noise portion of the tool background.

Next, the tool maximum diameter is calculated by calculating the pixel luminance data of the image processing area (see step 13 in FIG. 6).

FIG. 7A shows a processing procedure for extracting the tool diameter.

First, with respect to the read pixel luminance data, a pixel luminance cumulative sum Glxi is calculated for a vertical pixel row lxi arranged in a direction parallel to the tool diameter, and FIG. 8 (c) (milling type) and FIG. 10 (b) As shown in (drill type), a histogram is created (step 13-1).

From this histogram, each vertical pixel row l
The pixel rows lxa and lxb in which the pixel luminance cumulative sum Glxi of xi greatly changes in the direction perpendicular to the tool axis from the background image side to the tool image TI side are determined as both end positions of the maximum diameter portion of the tool image.

That is, the pixel luminance cumulative sum Glx is sequentially looked at in the increasing direction from the first column number lx1 of the vertical (Y line) pixel column, and the position where the first large change is taken as one end of the maximum diameter portion. The pixel luminance cumulative sum Glx is sequentially viewed in the direction decreasing from the last column number lxn of the direction pixel column, and the position where the pixel luminance first changes greatly is defined as the vertical pixel column lxb at the other end of the largest diameter portion (FIG. b), Steps 13-2 and 13
-3). A vertical pixel row lx at both left and right ends of the maximum diameter portion
a and lxb are the same as those in the first embodiment, as shown in FIGS.
It can be obtained by the differential processing shown in FIG.

By multiplying the number of pixel columns included between the pixel columns lxa and lxb by the actual pixel column width (preset),
As shown in FIG. 8E, the maximum diameter maxD of the tool can be calculated (step 13-4).

However, it is unknown whether the maximum diameter is the cutter CI or the holder HI.

Then, it is determined whether the maximum diameter portion of the tool is the cutter portion CI or the holder portion HI (step 14 in FIG. 6). If the maximum diameter portion is the cutter portion CI, the maximum diameter portion is recognized as the tool diameter. (FIG. 6, step 15), if the maximum diameter portion is the holder portion HI, the tool diameter is ignored without recognition (FIG. 6, step 16). Of course, it can also be used as data on the outer diameter of the holder HI.

FIG. 7B shows this determination procedure.

To determine whether the maximum diameter is the cutter portion CI or the holder portion HI, a horizontal pixel row ly in the direction orthogonal to the tool axis is used.
, A histogram is created by calculating the pixel luminance cumulative sum Gly (step 14-1). In the case of the milling type (see FIG. 9A) and the case of the drill type (FIG. 10)
(d)), a clear difference appears in the histogram,
Whichever of the cutter part CI and the holder part HI can determine the maximum diameter.

That is, in the case of the milling type, as shown in FIG.
Since ly is smaller than the holder HI, for example, if the subtracted value is negative, it can be determined that the milling type is used. Since the background is made brighter, the cutter portion CI having the smaller sum of pixel luminance Gly is the maximum diameter portion. As shown in FIG. 9B, the horizontal pixel rows lya and lxb corresponding to the cutter section CI and the holder section HI can be calculated by differentiating.

On the other hand, in the case of the drill type, FIG.
As shown in (d), since the cumulative sum of pixel brightness is smaller in the holder portion HI than in the cutter portion CI, the holder portion HI having the smaller sum of pixel brightness is the maximum diameter portion. The horizontal pixel rows lya and lyb corresponding to the cutter section CI and the holder section HI are shown in FIG.
It can be calculated by differentiating as shown in FIG.

In the second embodiment, the shape pattern is recognized after measuring the maximum diameter. However, the order is reversed, the shape pattern is recognized first, and the milling type tool or the drill type tool is recognized. The maximum diameter (tool diameter) may be measured in the case of a milling type tool.

[Embodiment 3] Next, a method for preventing the scattering of a tool tip by calculating the allowable rotation speed of the tool from the measured tool diameter will be described.

FIG. 11 shows the system configuration of the tool tip scattering prevention device. The same components as those in FIG. 1 are denoted by the same reference numerals, and description thereof will be omitted. In this system configuration, the image processing computer 1 functions not only as an image processing unit but also as a tool chip scattering determination unit.
Connected to the tool, sets a tool allowable rotation speed based on the measured tool diameter and preset allowable rotation speed data, compares a tool command rotation speed commanded from the NC device with the tool allowable rotation speed, and sets a tool command rotation speed. When the number is larger than the tool allowable rotation speed, a warning signal is output.

FIG. 12 is a flowchart of the procedure for determining the scattering of tool tips.

First, a tool image is read, and image processing as described in the first and second embodiments is performed to measure a tool diameter (steps 1 and 2).

Thereafter, the permissible rotational speed data is read, and a permissible rotational speed NM corresponding to the tool diameter is calculated (step 3).

Next, the tool command rotation speed N is read from the NC device (step 4) and compared with the tool allowable rotation speed NM calculated from the tool diameter (step 5).

If the tool command rotation speed N is larger than the tool allowable rotation speed NM (N> NM), a warning signal is output, and if the tool designated rotation speed N is less than the tool allowable rotation speed (N ≦ NM). In), the tool tip scattering determination processing is terminated as it is.

In this way, the scattering of the tool tip can be prevented.

[Other Embodiments] In each of the above embodiments, a tool image is photographed and read by a photographing device equipped with a CCD camera. In addition, a flexible disk in which pixel luminance data of a tool image is read in advance is used. Or the like, or an output image of a captured image may be read by a scanner.

The description has been made on the assumption that the diameter of the cutter portion is measured as the tool diameter. However, since the outer shape of the tool can be recognized, the diameter of the holder portion can be measured. It is also possible to measure the axial length.

[0091]

As described above, according to the first and second aspects of the present invention, each pixel row is formed for each pixel row aligned in at least one of a direction parallel to the tool axis and a direction orthogonal to the tool axis. Calculates the pixel luminance cumulative sum of a large number of pixels, and computes a pixel row in which the pixel luminance cumulative sum of each pixel row greatly changes from the background image toward the tool image, thereby recognizing the outer position of the tool image. Therefore, even in the case of an image in which the contrast between the tool image and the background image is small, the external shape can be accurately recognized by further enhancing the contrast.

By calculating the pixel luminance cumulative sum of the pixel row parallel to the tool axis, the outer position of the tool image in the direction orthogonal to the tool axis can be determined, so that the tool diameter can be calculated.

By calculating the pixel luminance cumulative sum of the pixel row orthogonal to the tool axis, the tool outer shape position in the direction parallel to the tool axis can be detected. For example, the tool tip position, the diameter of the tool image holder and the diameter of the cutter part can be detected. Are different, the boundary position is known, and the axial length of the tool can be calculated. In addition, the shape pattern of the tool such as which of the cutter portion and the holder portion of the tool image has a larger diameter can be found.

According to the third and fifth aspects of the present invention, the tip of the tool image is detected, and the image processing area for extracting the tool diameter is set based on the tip position of the tool image. Since the tool diameter is calculated from the pixel luminance cumulative sum of each pixel row aligned in the direction parallel to the tool axis in the image processing area for extraction, even if the contrast between the tool image and the background image is small, the contrast can be reduced. Is more emphasized and the tool diameter can be measured accurately.

Further, since the image processing area for extracting the tool diameter by detecting the tip position of the tool image is narrowed, image processing can be performed at high speed.

According to the fourth and sixth aspects of the present invention, the first image processing area for removing the background noise image around the image frame is set for the read pixel luminance data. Since the detection of the tip of the tool image is performed in, a higher-speed processing can be performed.

According to the seventh and eighth aspects of the present invention, the maximum diameter of the tool image is calculated from the cumulative sum of pixel pixel luminances of the respective pixel columns arranged in the direction parallel to the tool axis. Judge whether the largest diameter part is a cutter part or a holder part from the pixel luminance accumulation sum for each pixel row.If the largest diameter part is a cutter part, use the largest diameter as the tool diameter.If the largest diameter part is the holder part. Since the tool diameter is not determined in the case of, even when the contrast between the tool image and the background image is small, the contrast is further emphasized and the tool diameter can be accurately measured.

Further, by distinguishing the cutter portion of the tool image from the holder portion and setting the maximum diameter portion to the tool diameter only when the cutter portion is larger than the holder portion, the subsequent processing can be simplified.

According to the ninth and tenth aspects of the present invention, at the tool standby position of the automatic tool changer, the actual tool diameter is measured to set the allowable rotational speed of the tool. The tool command rotation speed is compared with the tool allowable rotation speed, and a warning is issued when the tool command rotation speed is larger than the tool allowable rotation speed. Chip scattering can be prevented beforehand.

[Brief description of the drawings]

FIG. 1A is a diagram showing an image processing system configuration of a machining center to which a tool diameter measuring device according to a first embodiment of the present invention is applied, and FIGS. 1B and 1C are diagrams showing the same; It is the side view and front view which show the lens attachment part of the imaging device of a).

FIG. 2 is a flowchart illustrating an image processing procedure for extracting a tool diameter according to the first embodiment;

3 (a) is a flowchart showing a procedure for detecting a tool tip position in FIG. 2, and FIG. 3 (b) is a flowchart showing a procedure for extracting a tool outer shape in FIG. 2;

4A is a schematic diagram showing a first image processing region in a captured image, FIG. 4B is a schematic diagram showing noise in the first image processing region, and FIG. FIG. 8 is a schematic diagram showing an image of a first image processing area after removing the image.

FIG. 5A is a histogram showing a pixel luminance cumulative sum for each horizontal pixel column in the first image processing area in FIG. 4;
(b) is a graph showing the amount of change in the accumulated pixel brightness between adjacent horizontal pixel columns in (a) of the same figure, and (c) is a second image processing area set for measuring the tool diameter Schematic diagram showing the same figure
(d) is a histogram showing the pixel luminance cumulative sum for each vertical pixel column in the second image processing area of FIG. (c), and (e) of FIG.
(d) is a graph showing the change amount of the pixel luminance cumulative sum between adjacent vertical pixel columns, (f) is a schematic diagram showing a second image processing region having a different setting region, and (g) is the same. A histogram showing the pixel luminance cumulative sum for each vertical pixel column in the second image processing area in FIG. (F), and FIG. (H) shows a pixel luminance cumulative sum between adjacent vertical pixel columns in FIG. 6 is a graph showing the amount of change.

FIG. 6 is a flowchart illustrating an image processing procedure of tool diameter extraction according to the second embodiment of the present invention.

7A is a flowchart showing a procedure for extracting a tool maximum diameter shown in FIG. 2, and FIG. 7B is a flowchart showing a procedure for determining whether or not the tool maximum diameter is a cutter portion.

8 (a) is a view showing a photographed image of a milling type tool, FIG. 8 (b) is a view obtained by removing the noise of FIG. 8 (a), and FIG. 8 (c) is a vertical view in the photographed image; (D) is a graph showing the amount of change in the pixel luminance cumulative sum between adjacent horizontal pixel columns in FIG. (C), and FIG. It is a figure which shows the maximum diameter measured.

9 (a) is a histogram showing a pixel luminance cumulative sum for each horizontal pixel column in the captured image of FIG. 8 (b), and FIG. 9 (b) is an adjacent horizontal direction of FIG. 8 (a). 9 is a graph showing a change amount of a pixel luminance cumulative sum between pixel columns.

10A is a diagram showing a photographed image of a drill type tool, and FIG. 10B is a histogram showing a pixel luminance cumulative sum for each vertical pixel column in the photographed image of FIG. 10A; , FIG.
FIG. 4B is a graph showing the change amount of the pixel luminance cumulative sum between adjacent horizontal pixel columns in FIG. 4B, and FIG. 5D is a pixel luminance for each horizontal pixel column in the captured image in FIG. FIG. 14E is a histogram showing the cumulative sum, and FIG. 14E is a graph showing the change amount of the pixel luminance cumulative sum between adjacent horizontal pixel rows in FIG. 14D.

FIG. 11 is a diagram showing a system configuration of a tool tip scattering prevention system using a tool diameter measured by the tool diameter measuring method of the present invention.

FIG. 12 is a control flowchart of the system in FIG. 11;

13 (a) is a schematic side view of a milling type tool in a tip mounted state, FIG. 13 (b) is a schematic side view of a drill type tool, and FIG. 13 (c) is FIG. 4) is a histogram showing the pixel luminance cumulative sum of the horizontal pixel row of FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Automatic tool changer 4 Gear box (apparatus main body) 5 Center axis 6 Exchange arm 7 1st servomotor (1st rotation drive source) 8 1st power transmission mechanism 9 2nd servomotor (2nd rotation drive source) 10th 2 Power transmission mechanism 13 Tool gripping part 14 Gripping arm 16 Key holder part 17 Rotational motion transmission mechanism 18 Nut shaft 19 Nut holding part 30 Ball screw mechanism (feed screw mechanism) 31 Ball nut 32 Ball screw shaft 33 Ball 37 Cap 60 Main shaft T1 Processing Used tools T2 Tools to be used 70 Sequencer (control means)

Claims (10)

[Claims] 1. A method for reading pixel brightness data of a tool image taken from a direction orthogonal to a tool axis, and forming each pixel row for each pixel row aligned in at least one of a direction parallel to the tool axis and a direction orthogonal to the tool axis. A pixel row in which the pixel luminance cumulative sum of each pixel row greatly changes from the background image toward the tool image is set as the outer position of the tool image. Image outline recognition method. 2. An image input means for inputting pixel luminance data of a tool image photographed from a direction perpendicular to the tool axis, and the image luminance data input by the image input means is set in a direction parallel to the tool axis or in a direction parallel to the tool axis. The tool calculates the pixel luminance cumulative sum of a large number of pixels constituting each pixel row arranged in the orthogonal direction and calculates the pixel row in which the pixel luminance cumulative sum of each pixel row greatly changes from the background image toward the tool image. Image processing means for executing a processing procedure for obtaining an outer shape position of an image,
An outer shape recognition device for tool images, comprising: 3. Reading pixel luminance data of a tool image photographed from a direction perpendicular to the tool axis, and for each pixel row aligned in a direction perpendicular to the tool axis with respect to the pixel luminance data in order to detect a tip of the tool image. Calculate the pixel luminance cumulative sum, and set the pixel row in which the pixel luminance cumulative sum greatly changes in the direction parallel to the tool axis from the background image side toward the tool image side as the tip position of the tool image, and determine the tip position of the tool image. An image processing area for tool diameter extraction including a pixel row in a direction orthogonal to a predetermined number of tool axes aligned on the tool background image side and the tool image side with respect to the tool background image side is set. Calculates the pixel luminance cumulative sum of each pixel row aligned in the direction parallel to the direction, and determines the pixel row where each pixel luminance cumulative sum changes greatly from the background image side toward the tool image side as the end position of the tool image, and determines the tool diameter. Calculate Tool diameter measuring method of the tool image characterized by. 4. The tool image according to claim 3, wherein an image processing area for removing a background noise image around the image frame is set for the read pixel luminance data, and a tip of the tool image is detected in the image processing area. Tool diameter measurement method. 5. An image input means for inputting pixel luminance data of a tool image photographed from a direction orthogonal to a tool axis, and for detecting the tip of the tool image with respect to the pixel luminance data input by the image input means. A pixel brightness cumulative sum is calculated for each pixel row aligned in a direction orthogonal to the tool axis, and the pixel row in which the pixel brightness cumulative sum greatly changes in the direction parallel to the tool axis from the background image side toward the tool image side is calculated as the tool image. The tip position of the tool image, a predetermined number of tool axes arranged on the tool background image side and the tool image side with reference to the tip position of the tool image and a tool diameter extraction image processing area including a pixel row in the orthogonal direction are set, Regarding the pixel luminance data of the image processing area for tool diameter extraction,
Calculate the pixel luminance cumulative sum of each pixel row aligned in a direction parallel to the tool axis, and determine the pixel row in which each pixel luminance cumulative sum changes greatly from the background image side toward the tool image side as both end positions of the tool image. An apparatus for measuring a tool diameter of a tool image, comprising: an image processing means for executing a procedure for calculating a tool diameter. 6. The image processing means sets a first image processing area for removing a background noise image around an image frame from the read pixel luminance data, and sets a tip of a tool image in the first image processing area. 6. The method according to claim 5, wherein the detecting step is performed.
A tool diameter measuring device for a tool image according to item 1. 7. A pixel luminance data of a tool image photographed from a direction orthogonal to a tool axis is read, and a pixel luminance cumulative sum of each pixel row arranged in a direction parallel to the tool axis is calculated for the pixel luminance data. A pixel row in which the cumulative sum of luminance changes greatly from the background image side to the tool image side is determined as both end positions of the maximum diameter portion of the tool image, and the maximum diameter is calculated, while each pixel arranged in the direction orthogonal to the tool axis is calculated. Calculate the pixel luminance cumulative sum for each column, compare the pixel luminance cumulative sum of the tool image cutter part and the holder part, determine whether the maximum diameter part is the cutter part or the holder part, and if the maximum diameter part is the cutter part A tool diameter measuring method for a tool image, wherein a tool diameter is not obtained when a maximum diameter portion is a tool diameter and a holder portion is a maximum diameter portion. 8. An image input means for inputting pixel luminance data of a tool image photographed from a direction orthogonal to a tool axis, and for each pixel arranged in a direction parallel to the tool axis, the pixel luminance data input by the image input means is provided. Calculate the pixel luminance cumulative sum of the column, judge the pixel row where each pixel luminance cumulative sum changes greatly from the background image side to the tool image side as both end positions of the maximum diameter part of the tool image, and calculate the maximum diameter. On the other hand, the pixel luminance cumulative sum is calculated for each pixel row aligned in the direction orthogonal to the tool axis, and the cutter pixel of the tool image and the pixel luminance cumulative sum of the holder are compared to determine whether the maximum diameter part is the cutter part or the holder part. A tool diameter measuring device for a tool image, comprising: image processing means for executing a processing procedure for determining the maximum diameter as a tool diameter when the maximum diameter portion is a cutter portion. 9. A tool diameter is measured by the tool diameter measuring method according to claim 3, at a tool standby position of the automatic tool changer, and the measured tool diameter and preset allowable rotation speed data are set. The tool allowable rotation speed is set from, and the tool command rotation speed commanded from the NC device is compared with the tool allowable rotation speed. If the tool command rotation speed is larger than the tool allowable rotation speed, a warning signal is output. A method for preventing tool chips from scattering. 10. A tool diameter measuring device according to claim 5, 6 or 8, wherein a tool allowable rotational speed is set from a tool diameter obtained by the tool diameter measuring device and preset allowable rotational speed data. N
A scatter determining unit that compares a tool command rotation speed commanded from the C device with the tool allowable rotation speed and outputs a warning signal when the tool command rotation speed is greater than the tool allowable rotation speed. Tool chip scattering prevention device.