JP3290034B2 - Chip component recognition method and apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for recognizing a chip component for obtaining the position and inclination of the chip component from an image of the chip component adsorbed on a component mounting head of a mounting machine. .

[0002]

2. Description of the Related Art Conventionally, in a mounting machine for mounting a chip component such as a transistor on a printed circuit board, a chip component adsorbed on a component mounting head is imaged in order to correct a chip component mounting position and the like. 2. Description of the Related Art A chip component recognizing apparatus that obtains the position and inclination of a chip component is known.

[0003] That is, the above mounting machine automatically picks up a chip component from a component supply unit by a component mounting head having a suction nozzle, transfers the chip component to a positioned printed board, and mounts the chip component on the printed board. It is supposed to be done. The recognition device mounted on the mounting machine has an image pickup means such as a CCD camera for picking up an image of the chip component adsorbed on the component mounting head, and captures an image of the chip component picked up by the image pickup means. The position and inclination of the chip component are obtained by this image processing means, and the correction of the mounting position of the chip component with respect to the printed circuit board is performed based on this.

As a method of recognizing a chip component by the image processing means, a processing window representing an image of the chip component is moved in an X-axis direction (horizontal direction) or a Y-axis direction (vertical direction).
Are scanned along the scanning line, and the edge and the like of the chip component are examined by this scanning, and the position and the inclination of the chip component are obtained based on this.

Various methods for obtaining the position and inclination of the chip component based on the above scanning have been considered. For example, in the case of recognizing a chip component whose image has a rectangular shape or a similar shape, a processing window is used. The change in density value is examined along a number of parallel scanning lines in the X-axis direction or the Y-axis direction, and a side portion of the chip component is detected by obtaining a straight line portion from the positional relationship of the density value change points. Then, there is a method of calculating the position and inclination of the center of gravity of the chip component from data such as the position and angle of the side portion.

For example, as disclosed in Japanese Patent Application Laid-Open No. 4-332199, when a chip component having a large number of leads on four sides is recognized, an X-axis direction or a Y-axis direction in a processing window is required. Of the scan lines crossing the lead row on the same side, the center of the lead row is obtained based on the density value data along this scan line, and thus each side is A method is also known in which the position of the center of gravity and the inclination of the chip component are obtained from the center of the lead row that has been examined.

[0007]

As described above, according to the conventional method for obtaining the position and inclination of the center of gravity of a chip component based on scanning by a scanning line in the X-axis direction or the Y-axis direction, when a very small chip component is recognized. In the case where the image of the chip component is extremely small as described above, an error is likely to occur, and it is difficult to increase the accuracy.

More specifically, according to the former of the above two methods, when the image of the chip component is small, the accuracy of the linear component obtained based on the scanning in the X-axis direction or the Y-axis direction is small. Therefore, it is difficult to obtain the chip component sufficiently, so that an error easily occurs in the detection of the side portion of the chip component and the calculation of the position of the center of gravity and the inclination based on the detection. In addition, Japanese Patent Laid-Open No.
According to the method disclosed in JP-A-332199, it is not easy to extract a scanning line that crosses a lead row when the lead is small on the screen, and the position of the extracted scanning line is apt to vary. As a result, measured values vary, and errors tend to occur.

The present invention has been made in view of the above circumstances, and provides a chip component recognition method and apparatus which can easily and accurately determine the position of a chip component even when the image of the chip component is small. The purpose is to:

[0010]

The invention according to claim 1 is
After imaging the chip component adsorbed on the component mounting head of the mounting machine, the image is displayed in the processing window, and based on the scanning in the processing window, the chip component recognition method for determining the position and inclination of the chip component is performed. A scanning line for scanning the inside of the processing window is set in a direction inclined with respect to the X-axis direction and the Y-axis direction of the processing window, and image data on the scanning line is determined while moving the scanning line in parallel. When the scanning line crosses two sides sandwiching the apex of the corner of the chip component, each position on the two sides is detected by comparing the density value with a predetermined threshold value, thereby detecting the corner of the chip component. The position and the inclination of the chip component are determined based on the detected data of the corner portion. Preferably, the scanning line is set at 45 ° with respect to the X-axis direction and the Y-axis direction.

According to a second aspect of the present invention, the corner part of the chip component is detected by detecting each position on the two sides when the scanning line crosses two sides sandwiching the vertex of the corner part of the chip part. In addition to this, image data at two positions in the direction perpendicular to the direction of the scanning line on the center side of the chip component from each position on the two sides is determined. .

The invention according to claim 3 is the invention according to claim 1 or 2
In the method of (1), a chip component whose image is a rectangle or a shape similar thereto is to be recognized, and the detection of the corner portion is performed by scanning the chip component with respect to the upper left corner portion and the lower right corner portion of the four corners on the processing window. The direction of the line is set obliquely upward to the right, and the scanning line direction for the lower left corner and the upper right corner is set obliquely to the lower right, so that the scanning is performed sequentially. The left side profile measurement line based on the left side detection data of one side part on the upper side of the chip component and the opposite side part of the detection data at the two positions of the part, A right side profile measurement line is obtained based on the right side detection data, and the obtained left and right side profile measurement lines are obtained. The center of gravity and the inclination of the chip component are determined from the center line connecting the center points of the chip components and the center point of the center line, or the detection data of the two positions at each corner is located on the left side of the chip component. An upper side profile measurement line based on each upper side detection data at a certain one side and an opposite side side, and a lower side profile measurement line based on each lower side detection data. The center of gravity and inclination of the chip component are obtained from the obtained center line connecting the center points of the upper and lower side profile measurement lines and the center point of this center line.

[0013] The invention according to claim 4 is the invention according to claim 1 or 2.
In the method of (1), a chip component whose image is a rectangle or a shape similar thereto is to be recognized, and the detection of the corner portion is performed by scanning the chip component with respect to the upper left corner portion and the lower right corner portion of the four corners on the processing window. The direction of the line is set obliquely upward to the right, and the scanning line direction for the lower left corner and the upper right corner is set obliquely to the lower right, so that the scanning is performed sequentially. The left side profile measurement line based on the left side detection data of one side part on the upper side of the chip component and the opposite side part of the detection data at the two positions of the part, The right side profile measurement line and the right side profile measurement line are obtained based on the detection data on the right side. A process of obtaining two specific points offset by a predetermined amount on the upper and lower sides from the respective center points, and a straight line that intersects the left and right side profile measurement lines and passes through each upper specific point, A straight line passing through each lower specific point is set, the midpoint of both side edge points on these straight lines is obtained, and the center of gravity and inclination of the chip component are obtained based on each midpoint, or Among the detection data at the two positions of the corner portion, an upper side profile measurement line based on each upper side detection data at one side portion on the left side of the chip component and the opposite side portion, A lower side profile measurement line is obtained based on the detection data of each lower side, and two points offset by a predetermined amount to the left and right sides from the respective center points on the upper and lower side profile measurement lines are obtained. A process for obtaining a fixed point, a straight line passing through the upper and lower side profile measurement lines and passing through each left specific point, and a straight line passing through each right specific point are set, and on both these straight lines Of the two side edge points are obtained, and the center of gravity and the inclination of the chip component are obtained based on the respective midpoints.

The invention according to claim 5 is the invention according to claim 1 or 2
In the method described above, when a chip component having a plurality of pins on both upper and lower side portions is to be recognized, the detection of the corner portion is performed by detecting the outer corner portion and the lower side portion of the left end pin of the pin row in the upper side portion of the chip component. The direction of the scanning line with respect to the outer corner of the right end pin of the pin row in the part is set obliquely upward to the right, and the outer edge of the right end pin of the pin row on the upper side of the chip component and the left end of the pin row on the lower side The direction of the scanning line with respect to the outer corner portion of the pin is set to be inclined downward and to the right, and the processing for obtaining the position and inclination of the chip component includes the above two positions of the outer corner portion of the right end pin in the upper side portion. Upper side based on the detection data on the right side of the detection data and the detection data on the left side of the above two positions of the detection data at the outer corner of the left end pin. Find the profile measurement line in the pin row direction, find the position of the center point of each pin on the left and right ends on the profile measurement line in the upper pin row direction, and find the outer corner of the right end pin in the lower side. The lower pin row direction profile measurement is performed based on the detection data on the right side of the detection data at the two positions and the detection data on the left side of the detection data at the outer corner of the left end pin at the two positions. The process of obtaining the line, obtaining the position of the center point of each pin on the left and right ends on the obtained lower pin row direction profile measurement line, and the process of obtaining each pin corresponding to the center point of each pin at the upper left end and upper right end Of the pin row in the upper side based on the edge point of each pin,
A process of obtaining the position of the edge point of each pin corresponding to the center point of each pin at the lower left end and the lower right end, and obtaining the center of the pin row in the lower side based on the edge point of each pin, A process of calculating the center of the center of the pin row on the upper side and the center of the pin row on the lower side and the inclination of a straight line connecting the centers of the two pin rows as the center of gravity and the inclination of the chip component is performed. On the other hand, when a chip component having a plurality of pins on both left and right side portions is to be recognized, the detection of the corner portion is performed at the outer corner portion and the right side portion of the upper end pin of the pin row on the left side portion of the chip component. The direction of the scanning line with respect to the outer corners of the lower end pins of the pin row is set obliquely upward and to the right, and the outer corners of the lower end pins of the pin row on the left side and the upper ends of the pin rows on the right side of the chip component The direction of the scanning line with respect to the outer corner portion of the pin is set obliquely to the lower right, and the processing for obtaining the position and the inclination of the chip component includes the two positions of the outer corner portion of the lower end pin on the left side portion. A left side pin row direction profile measurement line is determined based on the lower detection data of the detection data and the upper detection data of the two positions at the outer corners of the upper end pin. The position of the center point of each of the upper and lower pins on the profile measurement line in the direction of the left pin row is determined, and the lower side of the detection data of the two positions of the outer corners of the lower pin on the right side In the pin row direction on the right side based on the detection data at A process for determining the center point of each pin at the upper and lower ends on the profile measurement line in the direction of the pin row on the right side, and a process corresponding to the center point of each pin at the upper left and lower left ends. The position of the edge point of the pin is determined, the center of the pin row on the left side is determined based on the edge point of each pin, and the edge point of each pin corresponding to the center point of each pin at the upper right end and the lower right end And the process of calculating the center of the pin row on the right side based on the edge point of each pin, and the midpoint between the center of the pin row on the left side and the center of the pin row on the right side And the inclination of a straight line connecting the centers of these two pin rows are obtained as the center of gravity and the inclination of the chip component.

The invention according to claim 6 is the invention according to claim 1 or 2
In the method, when a chip component having a plurality of pins on at least one of the upper and lower one side portions is to be recognized, the detection of the corner portion is performed by detecting the left end pin of the pin row in the one side portion of the chip component. The direction of the scanning line with respect to the outer corner portion of the right side pin is set to be diagonally upward or diagonally right, and the direction of the scanning line with respect to the outer corner portion of the right end pin is set to be diagonally right or diagonally right. The process for determining the position and inclination of
The above 2 of the outer corner portion of the right end pin in the side portion
The profile measurement line in the pin row direction is determined based on the detection data on the right side of the position detection data and the detection data on the left side of the two position detection data at the outer corner of the left end pin. Processing of determining the position of the center point of each pin on the left and right ends on the profile measurement line in the direction of the pin row, and determining the position of the edge point of each pin corresponding to the center point of each pin; A process of obtaining the inclination of the chip component based on the arrangement of the pins, and a process of obtaining the center of the pin array at the edge on one side based on the edge point of each pin, and obtaining the center of the pin array at the edge of the one side. From the profile of the density value on the center line obtained based on the inclination of the chip component and the inclination of the chip component, an edge point on the other side on the center line is detected, and the pin row at the edge of the one side is detected. From an edge point of the heart and the other side portion side of the center line performs a process of obtaining the centroid of the chip component,
On the other hand, when a chip component having a plurality of pins on at least one of the left and right sides is to be recognized, the detection of the corner portion is performed outside the upper end pin of the pin row in the one side portion of the chip component. The direction of the scanning line with respect to the corner portion is set to be diagonally upward or diagonally right, and the direction of the scanning line with respect to the outer corner portion of the lower end pin is set to be diagonally diagonally right or diagonally right. And the process of obtaining the inclination include the detection data at the lower side of the detection data at the outer corners of the lower end pin and the detection data at the two positions at the outer corners of the upper end pin in the side edges. The profile measurement line in the pin row direction is obtained based on the detection data on the upper side of the A process of obtaining the position of the center point of each pin, and obtaining a position of an edge point of each pin corresponding to the center point of each pin, and a process of obtaining the inclination of the chip component based on the arrangement of the edge points of each pin And the center of the pin array at the edge on one side based on the edge point of each pin, and the center of the pin array at the edge of the one side and the inclination of the chip component. From the density value profile on the center line, the edge point on the other side on the center line is detected, and the center of the pin row at the edge on the one side and the edge point on the other side on the center line are detected. Thus, the processing for obtaining the center of gravity of the chip component is performed.

According to a seventh aspect of the present invention, there is provided an apparatus for recognizing a chip component, comprising: an image pickup device for picking up an image of a chip component adsorbed on a component mounting head of a mounting machine; and an image of the chip component picked up by the image pickup device. Image reading means, and an image reading means for reading out the image from the image storing means, scanning the processing window representing the image, and performing an arithmetic processing based on the image. The processing means scans the processing window. Scan lines are processed in the X-axis direction and Y direction of the window.
By setting the direction inclined with respect to the axial direction, determining the density value of the image on the scanning line while moving the scanning line in parallel, the scanning line crosses two sides sandwiching the vertex of the corner of the chip component. A corner detecting means for detecting each position on the two sides where the density value has changed by comparing the density value with a predetermined threshold value; and a center of gravity and inclination of the chip component based on the detected data of the corner. And calculating means for calculating.

[0017]

According to the method of the present invention, the corner of the chip component in the image in the processing window can be easily formed by scanning the processing window in a direction inclined with respect to the X-axis direction and the Y-axis direction. , And is reliably detected.
Then, based on the detection of the corner portion, the position and inclination of the center of gravity of the chip component can be accurately obtained.
Particularly, the scanning line is set at 45 ° with respect to the X-axis direction and the Y-axis direction.
, Setting of the scanning line and scanning along the scanning line are also easy. Further, in the detection of the corner part of the chip part, when the scanning line crosses two sides sandwiching the vertex of the corner part of the chip part, each position on the two sides is detected. In addition, image data at two positions in the direction perpendicular to the direction of the scanning line on the center side of the chip component from the positions on the two sides may be determined. In this case, erroneous detection due to noise can be prevented.

According to the method of the third aspect of the present invention,
When a chip component whose image has a rectangular shape or a similar shape is to be recognized, four corners of the chip component are detected by oblique scanning as described above,
Based on the detection data, the profile measurement line in the direction along the side at the left, right, upper, or lower side is properly determined. Then, the center points at both sides are obtained based on the profile along the profile measurement line, and the center of gravity of the chip component and the center line obtained from the center points obtained from the center points at both sides are obtained. A slope is required. In this case, with the profile measurement line being properly defined,
The accuracy of the center of gravity and the inclination of the chip component obtained as described above is also enhanced.

According to the method of the fourth aspect of the present invention, the profile measurement line in the direction along the side at the left or right or upper or lower side is properly determined. Then, two straight lines passing through a specific point on this profile measurement line are set, and the center of gravity and the inclination of the chip component are obtained from the midpoint on each straight line. Even for a component having a dent, the center of gravity and the like are required to be obtained with high accuracy.

According to the method of the fifth aspect of the present invention,
When a chip component having a plurality of pins on both side portions is to be recognized, the outer corner portions at both ends of the pin row on both side portions of the chip component are detected by the oblique scanning as described above, and the detection data is obtained. Based on the above, profile measurement lines in the pin row direction are appropriately determined at the both side portions. Then, the position of the center point of each pin on this line is determined from the profile along this profile measurement line, and the position of the edge point of each pin corresponding to each center point, the center of the pin row on both sides The position of the point, the position of the center of gravity of the chip component, and the inclination are sequentially obtained. In this case, as the profile measurement line is properly determined, the accuracy of the center of gravity and the inclination of the chip component obtained as described above is also increased.

According to the method of the invention according to claim 6,
In the case where a chip component having a plurality of pins on at least one side is to be recognized, the outer corners at both ends of the pin row in the one side are detected by the oblique scanning as described above. Based on the data, the profile measurement line in the pin row direction at this side is properly determined. Then, the position of the center point of each pin on this line is determined from the profile along this profile measurement line, and the position of the edge point of each pin corresponding to each center point, the inclination of the chip component, one side The position of the center point of the pin row at the edge on the side, the position of the corresponding edge point on the other side, and the position of the center of gravity are sequentially obtained. In this case, as the profile measurement line is properly determined, the accuracy of the center of gravity and the inclination of the chip component obtained as described above is also increased.

According to the device of the invention according to claim 7,
An image of the chip part is read from the image storage means by the processing means including the corner part detection means and the arithmetic means, and the corner part is detected from the image by oblique scanning and the chip part is determined based on the detection data. The detection of the position of the center of gravity and the inclination is automatically performed.

[0023]

An embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows an embodiment of a chip component recognition apparatus. In this figure, reference numeral 1 denotes a component mounting head mounted on a mounting machine, which is relatively movable with respect to a component supply unit and a printed board (not shown). The nozzle 1a is provided so as to be able to move up and down and rotate. Then, after sucking the chip component 2 from the component supply unit by the suction nozzle 1a, the component mounting head 1 moves onto the printed board, and mounts the chip component 2 at a predetermined mounting position on the printed board. I have.

Reference numeral 3 denotes a CCD camera as an image pickup means, which is installed at a predetermined position of the mounting machine, and the component mounting head 1 moves onto the camera 3 after the chip components are sucked, and the camera 3 After that, the mounting of the chip component 2 on the printed circuit board is performed. The camera 3 images the chip component 2 sucked by the suction nozzle 1a from below.
In the embodiment shown in FIG. 1, lamps 4 are arranged on both sides of the camera 3, and an image is taken by reflected illumination. Depending on the chip component, an illuminating means may be arranged above and imaging may be performed by transmitted illumination, or reflected illumination and transmitted illumination may be selectively used.

The mounting machine is provided with a mounting machine controller 5 for controlling the operation of the component mounting head 1 and the like, and an image processing unit 10 for processing an image of the chip component 2 captured by the camera 3. Is provided. The mounter controller 5 is provided with a component data storage unit 6 for storing various data on the chip components 2 sucked by the component mounting head 1.

The image processing unit 10 includes an A / D converter 11 for converting an image from the camera 3 into an analog-to-digital signal, and an image memory (an image memory) for storing an image having passed through the A / D converter 11. Storage means) 12 and processing means 1 for processing an image read from the image memory 12
3, the processing means 13 and the mounting machine controller 10
Memory 1 that exchanges information with the
And 4.

The processing means 13 comprises the image memory 12
An image is read out of the processing window, and a processing window representing the image is scanned and an arithmetic processing based on the image is performed. The image processing apparatus includes, in particular, a corner detection unit 15 and an arithmetic unit 16 as described below.

The corner detecting means 15 sets a scanning line for scanning the inside of the processing window in a direction inclined with respect to the X-axis direction and the Y-axis direction of the processing window, and moves the scanning line in parallel with the scanning line. Determine the density value of the image of
The position at which the density value changes when the scanning line crosses the corner of the chip component is detected. The inclination angle of the scanning line is set to be larger than the limit of the range of the angle deviation (the inclination of the chip component) of the chip component when it is sucked by the suction nozzle 1a. 45 ° with respect to the axial direction and the Y-axis direction. The calculating means 16 calculates the position and inclination of the center of gravity of the chip component based on the detected data of the corner by the corner detecting means 15.

Then, the data of the position of the center of gravity and the inclination of the chip component obtained by the processing means are sent to the mounting machine controller 5 via the dual port memory 14, and
The mounting machine controller 5 corrects the mounting position and angle of the chip component with respect to the printed circuit board in accordance with the deviation of the center of gravity from the nozzle center and the inclination.

The processing means 13 including the corner detecting means 15 and the calculating means 16 has programs for various recognition methods (first to fourth examples of recognition methods) which will be described later. Depending on the type of the chip component 2, a program of one of the methods is selected and the method is executed.

Here, several examples of chip components to be recognized will be described with reference to FIGS.

FIGS. 2A and 2B show a chip resistor 2A and an image thereof. The chip resistor 2A has electrodes 2 on both sides.
1 is a square block shape. When imaging the chip resistor 2A, either reflection illumination or transmission illumination may be used. In the case of reflection illumination, the electrode portion shown by a solid line in FIG. 2B becomes a bright image. The whole main body (shown by a broken line) becomes a dark image. FIG. 3 shows an image of the tantalum capacitor 2B. The tantalum capacitor 2B also has electrodes 22 on both sides and has a shape similar to that of the chip resistor 2A. Imaging may be performed by either reflection illumination or transmission illumination. . FIG. 4 shows an image of the aluminum capacitor 2C.
C has electrodes 23 on both sides at the center of the bottom. The image is taken by reflection illumination, and the image is shown by a solid line.

These chip parts 2A, 2B, 2C are:
The image becomes a rectangle or a shape similar to the rectangle, and becomes a recognition target in a first example of the recognition method or a second example of the recognition method described later. In addition to this, for example, the first example of the recognition method or the second example of the recognition method as long as the captured image is a rectangle or a shape similar thereto even in a cylindrical shape such as a color resistor or the like.
Become an example recognition target.

FIGS. 5A and 5B show a mini-transistor 2D and an image thereof.
Is a plurality (two in the figure) of pins 24 on both sides.
And is a recognition target of a third example of a recognition method described later. The imaging of the mini-transistor 2D is based on reflected illumination, and the pin 24 indicated by a solid line becomes a bright image.

FIG. 6 shows an image of the power transistor 2E. The power transistor 2E has three pins 25 on one side and one pin 26 on the other side. This is the recognition target of the fourth example. The imaging of the power transistor 2E is based on reflected illumination, and the pins 25, 26 and the like indicated by solid lines become bright images.

Next, various specific examples of the method for recognizing chip components by the above-described recognition device will be described. In the following description, it is assumed that the reflection value is used for the density value, but when the transmission illumination is used, the magnitude of the density value (bright and dark) may be considered in reverse.

(1) First Example of Recognition Method FIG. 7 shows the chip resistor 2A and the tantalum capacitor 2 described above.
The flowchart shows a first example of a recognition method suitable for a case where a chip component such as B, an aluminum capacitor 2C, and a color resistor, whose image has a rectangular shape or a similar shape, is to be recognized. This method will be described with reference to FIGS.

First, in step S1, a threshold value for detecting a corner portion is set based on the density values at a plurality of locations in the processing window 30 representing the image of the chip component. Specifically, as shown in FIG. 8, the processing window 30 including a large number of pixels 31 is scanned every few pixels to check the density value, and, for example, one pixel for each section of 5 pixels vertically and horizontally ( Check the density value for the hatched area). A maximum value Dmax and a minimum value Dmin are selected from the density values of these pixels, and a minimum value Dmi smaller than the maximum value Dmax is selected.
a suitable value greater than n, for example Dmin + (Dmax-Dmi
n) · 1/2 is set as a threshold value.

Next, in step S2, each corner of the chip component 2 is detected by scanning with the scanning line Ls inclined at 45 ° with respect to the X axis and the Y axis. That is, as shown in FIGS. 9 and 10, when the upper left corner of the chip component 2 is first detected, a scanning line is set at an oblique angle of 45 ° to the lower left, and is directed from the upper left corner to the center of the processing window. By comparing the density value of the pixel on the scanning line with the threshold value while sequentially moving the scanning line in parallel, the position of the scanning line when the scanning line crosses a corner of the chip component is detected.

More specifically, a point NW1 where it is first found that the density value of the target pixel is equal to or higher than the threshold value and the density value of the pixel perpendicular to the scanning direction and located on the center side of the chip component is equal to or higher than the threshold value; A point NW2 immediately before the density value of the target pixel becomes the threshold value or less again is obtained. In this case, the condition that the density value of the pixel perpendicular to the scanning direction and on the center side of the chip component is also equal to or greater than the threshold value is to prevent erroneous detection due to noise. The two points NW1 and NW2 are points at which the scanning line crosses two sides sandwiching the vertex of the corner, and these are referred to as corner candidate points.

Similarly, for each of the other corner portions, two candidate corner points are found. However, when the lower left corner is detected, the scanning line L that is slanted 45 ° to the lower right
s is translated from the lower left corner of the processing window 30 toward the center, and when the upper right corner is detected, the scanning line Ls inclined downward at an angle of 45 ° is translated from the upper right corner of the processing window 30 toward the center. When the lower right corner is detected, the scanning line Ls inclined downward at an angle of 45 ° is moved in parallel from the lower right corner of the processing window 30 toward the center.

In this way, as shown in FIG. 11A, eight convenient corner candidate points NW1, NW2, SW1,
SW2, NE1, NE2, SE1, SE2 are obtained, and the coordinates (NW1x, NW1y) (NW1
2x, NW2y) (SW1x, SW1y) (SW2x,
SW2y) (NE1x, NE1y) (NE2x, NE2
y) (SE1x, SE1y) (SE2x, SE2y) are stored.

When the detection of the corner portion is completed, a process of obtaining the position of the center of gravity and the inclination of the chip component 2 is performed based on the detected data. In this embodiment, the mounting machine controller 5
It is determined whether the electrodes of the chip component 2 are arranged left and right or up and down based on the data read from the component data storage unit 6 (step S3). The processes in S4 to S15 are performed, and when the electrodes are arranged vertically, the processes in steps S16 to S27 are performed.

[Processing for Electrode Left / Right Arrangement] In step S4, the profile of the left electrode portion (left side portion) is checked. Specifically, as shown in FIG. 11A, two points (upper side and lower side) NW of each corner candidate point of the upper and lower corners in the left electrode portion.
1, a profile measurement line LA1 passing through SW2 is set. For example, the coordinates (Xs, Ys) of the start point ST and the coordinates (Xs) of the end point EN of the profile measurement line LA1 are set as follows.
e, Ye).

[0046]

Xs = NW1x- (SW2x-NW1x) (1) Ys = NW1y- (SW2y-NW1y) (2) Xe = SW2x + (SW2x-NW1x) (3) Ye = SW2y + (SW2y-NW1y) ... (4) Then, the line LA1 from the start point ST to the end point EN
The density value at each position above is examined, and a profile of the density value as shown in FIG. 12 is obtained.

Subsequently, in step S5, the position of the vertical center point b1 of the left electrode portion as shown in FIG. 11B is calculated from the above profile. Specifically, a maximum value Dmax and a minimum value Dmin of the density value are obtained in the profile, and a value smaller than the maximum value Dmax and larger than the minimum value Dmin, for example, Dmin + (Dmax-Dmin)
・ Set 1/2 as the threshold value. Then, looking at the density values of the profile in order from the starting point side, first find a point that exceeds the threshold, and calculate the sub-pixel by linear interpolation from the density value of this point, the density value of the immediately preceding point, and the threshold value. The Y coordinate of the converted upper edge point a1 is calculated.

The sub-pixel conversion by the linear interpolation is as follows.
This is for obtaining the position in a unit smaller than the pixel unit (sub-pixel unit). When the density value exceeds the threshold value at the i-th point counted from the starting point of the profile for the first time, the density value of that point is calculated. D _i, obtained in the (i-1) -th D _i-1 the density value of a point, when Dth the threshold (see FIG. 12), the sub-pixels Y coordinate of the top edge point a1 (Ya) is the following formula Can be

[0049]

[Number 2] _{Ya = Ys + i- (D i} -Dth) / (D i -D i-1) ... (5) In addition, go look in order the density value of the profile from the end point side, the upper edge points a1 The Y coordinate of the lower edge point a2 is obtained in the same manner as the calculation of the Y coordinate of. And
The Y coordinate of the upper edge point a1 and the Y coordinate of the lower edge point a2
The middle value of the coordinates is defined as the Y coordinate of the vertical center point b1 of the left electrode portion, and the center point b1 corresponding to the Y coordinate is obtained from the equation of the straight line (profile measurement line LA1) passing through the profile start point ST and end point EN. Find the X coordinate.

Next, in steps S6 and S7, step S
4 and S5 are performed on the right electrode portion (right side portion). That is, the coordinates (Xs, Ys) of the start point and the coordinates (Xe, Ye) of the end point of the profile measurement line LA2 passing through the two points NE2 and SE1 of the corner candidate points at the upper and lower corners are determined as follows.

[0051]

Xs = NE2x- (SE1x-NE2x) (6) Ys = NE2y- (SE1y-NE2y) (7) Xe = SE1x + (SE1x-NE2x) (8) Ye = SE1y + (SE1y-NE2y) … (9) Then, take a profile from the start point to the end point,
On the basis of this profile, the Y coordinate of the upper edge point a3 and the Y coordinate of the lower edge point a4 are obtained in subpixel units by linear interpolation, and the vertical center point b of the right electrode portion is obtained.
2 and the corresponding X coordinate of the center point b2 (see FIG. 11B).

As described above, each center point b of the left and right electrode portions is
By calculating 1, b2, the following steps S13 to S13
The position of the center of gravity and the inclination of the chip component can be obtained in the process of S15. However, in this embodiment, in order to improve the accuracy, in steps S8 to S12, the profile measurement line is corrected for each of the left and right electrode portions, and the vertical Processing for calculating the direction center point is performed (see FIGS. 11B and 11C).

That is, if the center points b1 and b2 of the both-sided electrode portions obtained in steps S5 and S7 are referred to as provisional center points, in step S8 the two provisional center points b
The inclination of the straight line LB passing through the lines 1 and b2 is calculated. Then, in step S9, an equation of a straight line LC1 which is orthogonal to the straight line LB and passes through the provisional center point b1 of the left electrode portion is obtained, and the Y coordinate on this straight line LC1 is calculated based on the electrode width from the Y coordinate of the provisional center point b1. A profile is taken with a point smaller by the start point and a point larger by the electrode width as the end point, and step S
In step 10, a vertical center point c1 of the left electrode is calculated in the same manner as in step S5. Further, in step S11, an equation of a straight line LC2 orthogonal to the straight line LB and passing through the provisional center point b2 of the right electrode portion is obtained.
A profile is set on C2 with a point whose electrode is smaller than the Y coordinate of the provisional center point b2 by the electrode width as a start point and a point whose end point is larger than the Y coordinate by the electrode width in step S12. Vertical center point c2
Is calculated.

Next, in step S13, FIG.
An equation of a straight line (center line) LD passing through the center points c1 and c2 of the above both-side electrode portions as shown in FIG.
Then, on the straight line LD, the length of the line segment is [(inter-electrode distance) + (electrode width) × 2], and the distance from the start point ST to the center point c1 of one electrode portion and the end point EN The start point ST and the end point EN of the profile are set so that the distance to the center point c2 of the other electrode part is equal,
The profile on this straight line LD is taken.

Subsequently, in step S15, the position and inclination of the center of gravity of the chip component are calculated. Specifically, the straight line L
Dmax and minimum D of the density values in the profile on D
min, and the minimum value Dm smaller than the maximum value Dmax
A value larger than in, for example, Dmin + (Dmax-Dmin) · 1
/ 2 is set as a threshold value, and the X coordinate of the left edge point d1 and the X coordinate of the right edge point d2 on the straight line LD are calculated in subpixel units by the same method as the calculation of the Y coordinate of the edge performed in step S5. calculate. Then, the X coordinate of the middle point is calculated from the X coordinates of the edge points d1 and d2, and the Y coordinate of the middle point is calculated from the X coordinate and the equation of the straight line LD. Then, this midpoint is referred to as chip component 2
And the gradient is calculated from the equation of the straight line LD, and this gradient is defined as the inclination θ of the chip component.

[Processing for Electrode Vertical Arrangement] In step S16, a profile of the upper electrode portion (upper side) is obtained. In step S17, the center of the upper electrode portion in the left-right direction is calculated from this profile.
At 18, a profile of the lower electrode portion (lower side portion) is obtained, and at step S19, the center of the lower electrode portion in the left-right direction is calculated from this profile. These processes are the same as steps S4 to S7 described above. However, as shown in FIG. 13, the profile measurement line L of the upper electrode portion
A1 passes through two points NW2 and NE1 at the upper left and right corners, and the profile measurement line LA of the lower electrode portion
2 passes through the two points SW1 and SE2 at the lower left and right corners. That is, the upper profile measurement line LA
The coordinates (Xs, Ys) of the start point and the coordinates (Xe, Ye) of the end point of 1 are determined as follows.

[0057]

Xs = NW2x- (NE1x-NW2x) (10) Ys = NW2y- (NE1y-NW2y) (11) Xe = NE1x + (NE1x-NW2x) ... (12) Ye = NE1y + (NE1y-NW2) (13) Further, the coordinates (Xs, Ys) of the start point and the coordinates (Xe, Xe) of the end point of the lower profile measurement line LA2 are obtained.
Ye) is determined as follows.

[0058]

Xs = SW1x− (SE2x−SW1x) (14) Ys = SW1y− (SE2y−SW1y) (15) Xe = SE2x + (SE2x−SW1x) (16) Ye = SE2y + (SE2y−SW1y) (17) Then, the center point in the left-right direction on each profile measurement line LA1, LA2 is obtained.

Subsequently, in step S20, the inclination of a straight line passing through the center point (temporary center point) of the upper and lower electrode portions is calculated. In step S21, the profile of the upper electrode portion is obtained. The center point is calculated. In step S23, a profile of the lower electrode portion is obtained. In step S24, the center point of the lower electrode in the left-right direction is calculated. Further, in step S25, an equation of a straight line passing through the center point of the upper and lower electrodes is calculated. In step S26, a profile between the upper and lower electrodes is obtained.
In step S27, the position and inclination of the center of gravity of the chip component are calculated. These processes are the same as the processes of steps S8 to S15 in the case of the above-described electrode left-right arrangement (however, the left side is replaced with the upper side and the right side is replaced with the lower side).

According to the above-described first example of the recognition method, the four corners of the chip component are easily and reliably detected by scanning in the oblique direction. The profile measurement lines LA1 and LA2 in the direction along the side are properly determined, and the profile measurement lines LA1 and LA2 do not vary and the accuracy of profile measurement is increased. Then, the center point in the side portion is obtained based on the profiles along the profile measurement lines LA1 and LA2, and the center of gravity and the inclination of the chip component are calculated based on the center points of the side portions on both sides. In this way, the center of gravity and the inclination of the chip component whose image is a rectangle or a shape similar thereto can be obtained with high accuracy.

(2) Second Example of Recognition Method FIG. 14 shows a second example of a recognition method which is another specific example of a recognition method suitable for a case where a chip component whose image has a rectangular shape or a similar shape is to be recognized. It will be described with reference to FIG. The second example of the recognition method is a modification of the first example of the recognition method described above, and performs the same processing as that of the first example of the recognition method except for the following changes. The flowchart is omitted.

In the second example of the recognition method as well, after setting a threshold value for detecting a corner portion, each corner of the chip component 2 is detected by scanning at an angle of 45 °, and both side portions are detected based on the detected data. The process of setting the profile lines of each and taking the profile of each density value is the first recognition method.
This is the same as the example (steps S1 to S3 and steps S4 to S9, S11 or steps 16 to S21,
S23). Then, based on the profiles along the side profile lines (preferably the lines corrected as described above) LC1 and LC2, the center of gravity w and the inclination of the chip component are obtained as shown in FIG.

That is, from the profile along each of the lines LC1 and LC2, two specific points which are offset by a predetermined amount on both sides from the center of the component side portion on each of the lines LC1 and LC2 are determined. For example, on one (left side in FIG. 14) line LC1, each of the edge points a11 and a12 on both sides (upper and lower sides in FIG. 14) is 1/1 of the part length (the length between both edge points a11 and a12). A first specific point c11 and a second specific point c12 are defined at the position of No. 4 and the other (right side in FIG. 14) line LC2 is similarly set to a part length of 1 from the both side edge points a21 and a22.
A first specific point c21 and a second specific point c22 are determined at the position of / 4.

Next, a straight line LD1 crossing each of the lines LC1 and LC2 and passing through the first specific points c11 and c21 and a straight line LD2 passing through each of the second specific points c12 and c22 are set. Profiles are respectively taken along LD1 and LD2, and from each profile, a first midpoint d3 which is a midpoint between the two edge points d11 and d21 on one straight line LD1, and both edge points d12 on the other straight line LD2. , D22 and a second midpoint d4. Then, a midpoint between the first and second midpoints d3 and d4 is calculated and used as the center of gravity w, and a straight line passing through the first and second midpoints d3 and d4 (or a straight line orthogonal to this).
Is calculated, and this is defined as the inclination of the chip component.

According to such a second example of the recognition method, according to the chip component 2 whose image is a rectangle or a shape similar thereto,
The center of gravity and inclination of are determined with high accuracy. In particular, among components of this type, a component having a depression 28 or the like in the middle of one side electrode, such as a tantalum capacitor 2B shown in FIG. It is effective when That is,
In the case of the tantalum capacitor 2B, the center line LD is obtained as in the first example of the recognition method described above, and the center of gravity w is calculated from the center line LD.
Is calculated, an error occurs when the center line LD passes through the recess 28, but according to the second example of the recognition method, the error
Since the straight lines LD1 and LD2 pass through the position avoiding the recess 28, and the center of gravity w is obtained from the respective middle points d3 and d4 on the two straight lines LD1 and LD2, an error may occur due to the presence of the recess 28 or the like. There is no.

(3) Third Example of Recognition Method FIG. 15 shows a chip component in which a plurality of lead pins are disposed on both sides (left and right sides or upper and lower sides) of the mini-transistor 2D and the like. Is shown in the form of a flowchart in a third example of a recognition method suitable for a case where is a recognition target. This method will be described with reference to FIG.

First, in step S101, a threshold value for detecting a corner portion is set in substantially the same manner as in step S1 in the first example. Specifically, the inside of the processing window representing the image of the chip component is scanned every few pixels, and for example, the density value of each pixel is checked for each of the two vertical and horizontal pixels, and the density values of these pixels are checked. Medium to maximum value Dma
x and the minimum value Dmin are selected, for example, Dmin + (Dmax
−Dmin) · 1/2 is set as the threshold value.

Next, in step S102, four corners are detected in the same manner as in step S2 in the first example. That is, as shown in FIG.
By scanning with the scanning line Ls inclined at 45 ° with respect to the axis and the Y axis, the outer corners of the pins at both ends of each of the pin rows on both sides of the chip component 2D are detected. Eight corner candidate points NW
1, NW2, SW1, SW2, NE1, NE2, SE
1, SE2 is obtained, and the coordinates of each corner candidate point are stored.

When the detection of the corner portion is completed, a process for obtaining the position and inclination of the center of gravity of the chip component is performed based on the detected data. In this case, based on the data read from the component data storage unit 6 of the mounter controller 5, it is determined whether the electrodes of the chip component 2D are arranged left and right or up and down (step S103). If it is arranged in step S104 to S11
2 are performed, and when the electrodes are arranged vertically, the processes of steps S113 to S121 are performed.

[Processing for Left-Right Electrode Arrangement] In step S104, a profile on the left side is obtained. In other words, the starting point S is set so that the profile measurement line LA1 passes through the upper and lower corner candidate points NW1 and SW2.
The coordinates of T and the end point EN are set by the above equations (1) to (4) (see FIG. 16A), and a profile of density values along this line LA1 is obtained.

Next, in step S105, the center of each pin on the left side is calculated based on the profile. Specifically, the maximum value Dma of the density value in the above profile
A value smaller than x and larger than the minimum value Dmin, for example, Dmin + (Dmax-Dmin)） / 3 is set as a threshold. Is calculated in subpixel units by shape interpolation, and the coordinates of the midpoint are calculated as the center points e1 and e of the respective pins.
2 (see FIG. 16B).

In step S106, left edge points f1 and f2 of each pin as shown in FIG. 16B are calculated. Specifically, a straight line LE1, orthogonal to the profile measurement line LA1, and passing through the center points e1, e2.
The equation of LE2 is obtained, and X on the straight lines LE1 and LE2.
Profiles are taken with points whose coordinates are smaller and larger than the X-coordinates of the center points e1 and e2 of the pins by ４ of the lateral length of the chip component as start and end points. Then, a value smaller than the maximum value Dmax and larger than the minimum value Dmin of the density value in this profile is set as a threshold value, and based on a comparison between the threshold value and the density value, the left edge point of each of the above pins is determined using linear interpolation. The coordinates of f1 and f2 are obtained in sub-pixel units. At this time, when the number of pins on the screen is small, in order to increase the number of data for calculating the position and inclination of the component center of gravity and increase the accuracy, a plurality of thresholds are set, for example, Dmin + (Dmax−Dmin) · 1 / 3, Dmin + (Dmax-
It is preferable to set three types of thresholds, ie, Dmin) ・, Dmin + (Dmax−Dmin) ・ 、, and obtain the left edge coordinates of each pin for each.

In step S107, the position of the center point g1 of the pin array on the left side is calculated. Specifically, the coordinates of the center point g1 are obtained from the coordinates of the left edge points f1 and f2 of each pin. At this time, the left edge points f1 and f2 of each pin are If each is obtained, the center point g1 for each of the three types is obtained.
Find the coordinates of.

In steps S108 to S111,
By the same processing as steps S104 to S107, a profile is taken along the profile measurement line LA2 passing through the upper and lower corner candidate points NE2 and SE1 on the right side,
The center points e3 and e4 of the right side pins are calculated from this profile, the right edge points f3 and f4 of the right side pins corresponding to the center points e3 and e4 are detected, and based on these edge points f3 and f4. The center point g2 of the right side pin row is calculated.

In step S112, the position of the center of gravity w and the inclination of the chip component are calculated. Specifically, the midpoints of these are determined from the center point g1 of the left-side pin row and the center point g2 of the right-side pin row, and the slope thereof is calculated from the equation of a straight line LG passing through the center points g1 and g2 of both pin rows. These are the center of gravity w and the inclination of the chip component. At this time, when the three types of center point g1 of the left side pin row and the center point g2 of the right side pin row are obtained as described above, the center point is obtained for each of the three types, and these are averaged. Is the center of gravity w, and the center points g1,
The inclination of the straight line LG passing through g2 is obtained, and the average value is defined as the inclination of the chip component.

[Processing for Electrode Vertical Arrangement] In steps S113 to S116, a profile is taken along a profile measurement line passing through the left and right corner candidate points on the upper side, and the center point of each upper side pin is calculated from this profile. , And detects the upper edge points of the upper pins corresponding to these center points, and calculates the center point of the upper pin row based on these edge points. Step S117
In S120, a profile is taken along the profile measurement line passing through the left and right corner candidate points on the lower side,
The center point of each lower side pin is calculated from the profile, the lower edge point of each lower side pin corresponding to these center points is detected, and the center point of the lower side pin row is calculated based on these edge points. In step S121, the center of gravity and the inclination of the chip component are calculated based on the center point of the upper pin row and the center point of the lower pin row. The processing in steps S113 to S121 is performed by replacing the left side with the upper side and the right side with the lower side in the processing in the case of the electrode left-right arrangement.
This is the same as steps S104 to S112.

According to the third example of the recognition method as described above, the corners of the pins at both ends in the left and right or upper and lower sides are easily and reliably detected by oblique scanning, and based on this, the pin The profile measurement lines LA1 and LA2 in the direction traversing the columns are properly determined, and the profile measurement lines LA1 and LA2 do not vary and the accuracy of profile measurement is increased. From the profiles along the profile measurement lines LA1 and LA2, the center point of each pin on this line is obtained, and the edge point of each pin and the pin row center point are sequentially obtained. The center of gravity and inclination of the part are calculated. In this way, the center of gravity and the inclination of the chip component such as the mini-transistor 2D can be obtained with high accuracy.

(4) Fourth Example of Recognition Method FIGS. 17 and 18 show a case in which a chip component having a plurality of pins on at least one side is to be recognized, and in particular, one side of the power transistor 2E and the like. A fourth example of a recognition method suitable for a case where a chip component having a different number of pins and an arrangement of pins on an opposite side portion is set as a recognition target is shown in a flowchart, and this method will be described with reference to FIG.

First, in step S201, a threshold value for detecting a corner portion is set in the same manner as in step S1 in the first example.

Then, the corner portion is detected, and processing for obtaining the position of the center of gravity and the inclination of the chip component is performed based on the corner part. In this example, prior to these processings, the electrodes are arranged on the left or right or up and down. It is determined whether the electrodes are arranged (step S202). If the electrodes are arranged on the left and right, it is determined whether the number of pins on either side of the left or right side is large (step S203), and the electrodes are arranged vertically. If yes, it is determined which of the upper and lower sides has the larger number of pins (step S204). If the electrodes are arranged in the left and right direction and the number of pins on the left side is large, the corner portion detection process of step S205 and the step S206
Steps S212 to S212 are performed, and when the electrodes are arranged in the left and right direction and the number of pins on the right side is large, the corner part detection processing in step S213 and steps S214 to S220 based thereon are performed.
Are performed. When the electrodes are arranged vertically and the number of pins on the upper side is large, the corner portion detection processing of step S221 and the respective processing of steps S222 to S228 based thereon are performed, and the electrodes are arranged vertically and the number of pins on the lower side is large. In this case, the corner portion detection process in step S229 and the processes in steps S230 to S236 based thereon are performed.

[Processing when the number of pins on the left side is large in electrode left-right arrangement] In step S205, the process of corner detection is as follows.
As shown in FIG. 19A, 45 ° with respect to the X axis and the Y axis.
The outer corners of the pins at the upper and lower ends of the pin row on the left side are detected by scanning with the inclined scanning line Ls, and the corner candidate points NW1, NW2, SW1, SW of these corners are detected.
Ask for 2.

Next, in step S206, the profile of the density value is obtained for the left side along the profile measurement line LA set based on the above-mentioned corner detection. That is, the start point ST is set so that the profile measurement line LA passes through the upper and lower corner candidate points NW1 and SW2.
And the coordinates of the end point EN are set by the above equations (1) to (4) (see FIG. 19A), and a profile is taken along the line LA.

In step S207, the position of the center point h1, h2, h3 of each pin on the left side is calculated based on the profile. Specifically, a value smaller than the maximum value Dmax and larger than the minimum value Dmin of the density value in the profile, for example, Dmin + (Dmax-Dmin)） 1 /
3 as a threshold value, based on a comparison between the threshold value and the density value, the upper and lower edges of each of the pins are obtained in sub-pixel units by linear interpolation, and the coordinates of the center point are defined as the center points h1, h2, and h2 of the pins. h3 (FIG. 19)
(B)).

In step S208, the left edge points i1, i2, i3 of each pin as shown in FIG. 19B are calculated. Specifically, the profile measurement line LA
And the equations of straight lines LH1, LH2, LH3 passing through the center points h1, h2, h3 are obtained, and the straight line LH
1, LH2, LH3, the X coordinate is the center point h1,
1 / one of the horizontal length of the chip component than the X coordinate of h2 and h3
The profile is taken on a line starting and ending with a point smaller and larger by two. Then, the minimum value D is smaller than the maximum value Dmax of the density value in this profile.
A value larger than min, for example, Dmin + (Dmax-Dmin)
Using 1/2 as a threshold value, based on a comparison between the threshold value and the density value, the left edge points i1, i2, and i3 of the respective pins are calculated in sub-pixel units using linear interpolation.

In step S209, the inclination of the chip component is calculated based on the arrangement of the left edge points i1, i2 and i3 of the respective pins. Specifically, each of the edge points i1,
For all combinations of two points out of i2 and i3 (in the example shown in FIG. 19, each combination of i1 and i2, i1 and i3, and i2 and i3), the slope of a straight line orthogonal to a straight line passing through the two points is obtained. The average value is defined as the inclination of the chip component.

In step S210, the center point j1 of the left side pin row is determined based on the left side edge points i1, i2, i3.
, Ie, these edge points i1, i2, i3
Are the coordinates of the center point j1 of the pin row.

In step S211, the right edge point j2 of the right side pin is determined based on the profile of the direction of the center line obtained from the inclination and the center point j1 of the left side pin row.
Is detected (see FIG. 19C). Specifically, an equation of a straight line (center line) LJ which passes through the center point j1 of the pin array on the left side and has the same inclination as the inclination of the chip component is obtained, and the X coordinate on the straight line LJ is the center point of the pin array on the left side. A profile is taken with a point larger than the X coordinate of j1 by 3/2 of the horizontal length of the chip component as a start point ST and a center point j1 of the left side pin row as an end point EN. Then, a value smaller than the maximum value Dmax and larger than the minimum value Dmin, for example, Dmin + (Dmax-Dmin) ・ of the density value in this profile is set as a threshold value, and based on a comparison between the threshold value and the density value, The right edge point j2 of the right side pin on the straight line is obtained for each sub-pixel by linear interpolation.

In step S212, step S2
10 and the center point j1 of the left side pin row
The midpoint between the right edge pin j2 and the right edge point j2 of the right pin determined in 211 is calculated as the center of gravity w of the chip component.

[Processing when the Number of Pins on the Right Side is Large in Electrode Left-Right Arrangement] In step S213, the outer corners of the pins at the upper and lower ends of the pin array on the right side are detected by oblique scanning at 45 °. The profile on the right side is obtained based on the corner detection, and the position of the center point of each pin on the right side is calculated from the profile in step S215, and the position of the right edge point of each pin corresponding to the center point is calculated in step S216. Is detected. Further, step S217
Then, the inclination of the chip component is calculated from the arrangement of the respective right edge points. In step S218, the position of the center point of the right side pin row is calculated from the respective right edge points.
In step S2, the left edge point of the left pin is detected.
In step 20, the position of the center of gravity of the chip component is calculated from the center point of the right side pin row and the left edge point of the left side pin.

The processes in steps S213 to S220 are the same as steps S205 to S212 described above, except that the left side and right side in the process when the number of pins on the left side is large are replaced.

[Processing when the number of pins on the upper side is large in electrode vertical arrangement] In step S221, the outer corners of the pins at the left and right ends of the upper side pin row are detected by oblique scanning at 45 °. The profile of the upper side is obtained based on the corner detection, the position of the center point of each pin on the upper side is calculated from the profile in step S223, and the position of the upper edge point of each pin corresponding to each center point is calculated in step S224. Is detected. Further, step S225
Then, the inclination of the chip component is calculated from the arrangement of the respective upper edge points. In step S226, the position of the center point of the upper side pin row is calculated from the respective upper edge points.
In step S2, the lower edge point of the lower pin is detected.
In step 28, the position of the center of gravity of the chip component is calculated from the center point of the upper pin row and the lower edge point of the lower pin.

The processes of steps S221 to S228 are the same as steps S205 to S212 described above, except that the left side is replaced by the upper side and the right side is replaced by the lower side in the processing when the number of pins on the left side is large in the left-right electrode arrangement.

[Processing when the number of pins on the lower side is large due to the vertical arrangement of the electrodes] In step S229, the outer corners of the pins on the left and right ends of the lower row of pins are detected by oblique scanning at 45 °. The profile of the lower side is obtained based on the corner detection, the position of the center point of each lower side pin is calculated from the profile in step S231, and the lower edge point of each pin corresponding to the respective center point is calculated in step S232. Detect the position. Further, step S233
Then, the inclination of the chip component is calculated from the arrangement of the lower edge points. In step S234, the position of the center point of the lower pin row is calculated from the lower edge points.
In step S2, the upper edge point of the upper pin is detected.
At 36, the position of the center of gravity of the chip component is calculated from the center point of the lower pin row and the upper edge point of the upper pin.

The processing in steps S229 to S236 is the same as the processing in steps S205 to S212 except that the left side is replaced with the lower side and the right side is replaced with the upper side in the processing when the number of pins on the left side is large in the left-right electrode arrangement.

According to the fourth example of the recognition method described above, the corners of the pins at both ends in the side portion having a large number of pins are easily and reliably detected by scanning in the oblique direction. The profile measurement line LA in the direction crossing the pin array on the side portion is properly determined, and the profile measurement line is not varied and the accuracy of the profile measurement is increased. Then, the center point of each pin on this line is determined from the profile along the profile measurement line, the edge points of each pin are further determined, and the inclination of the chip component is determined from the arrangement of these edge points, The pin row center point on the side edge, the edge point on the opposite side on the center line, and the center of gravity of the chip component are sequentially obtained. Thus, the center of gravity and the inclination of the chip component such as the power transistor 2E can be obtained with high accuracy.

(5) Others Other Examples of Discriminating Corners Among the recognition methods shown in the above-described specific examples, an angle of 45 ° is used.
In the example shown in FIG. 10 described above, the process of detecting a corner portion by the scanning of the corner candidate point indicates that the density of the target pixel is equal to or higher than the threshold value and the density of the pixel on the center side is higher than the threshold value. However, in order to reliably prevent an erroneous determination when a large noise exists, it is more preferable to determine a corner candidate point by a method using tracking as described below.

This method is applied to the corner candidate points NW and N of the pins 201 and 202 of the chip component 20 as shown in FIG.
A case where E, SW, and SE are determined will be described as an example.

In the case of detecting the upper left corner of the pin 201, when a determination target point (a pixel whose density exceeds the threshold value) is found by scanning at an angle of 45 °, first, as a first condition, the target point as shown in FIG. Surrounding pixels ("1", "2", "3", "4", "5", "6", "7"
It is checked that the density values of the pixels “2”, “3”, and “4” among the pixels “0”) are equal to or smaller than the threshold value. When this condition is satisfied, the upper side is tracked starting from the target point. Specifically, the three pixels “1”, “0”, and “7” on the right side are examined in this order. If the density exceeds the threshold, the starting point is moved to the pixel and the three pixels on the right side are examined again. Such tracking processing is repeated until the density of all of the three pixels (“1”, “0”, “7”) becomes equal to or less than the threshold. Therefore, if the discrimination target point is a correct corner candidate point NW, tracking is performed from this target point to the right end of the upper side of the pin 201.

Then, as a second condition for determination, the number of repetitions of the tracking is compared with the pin width equivalent value given by the library data, and it is determined whether or not the condition that the deviation between the two is within a predetermined allowable range is satisfied. Is checked, and if this is satisfied, the above-mentioned determination target point is set as a corner candidate point NW. Note that when the first condition is not satisfied, or when the second condition is not satisfied even if the first condition is satisfied,
The scanning at 5 ° is continuously performed to search for another determination target point.

When the upper right corner of the pin 201 is detected, it is determined that the density values of the pixels “0”, “1”, and “2” among the pixels around the determination target point are equal to or smaller than the threshold value. As a first condition, when this condition is satisfied, the three pixels "3", "4", and "5" on the left side of the starting point are examined in this order, and tracking is repeated until the density of the three pixels becomes equal to or less than the threshold. When detecting the lower left corner of the pin 202, “4” and “5” are selected from the pixels around the determination target point.
The first condition is that the density value of the pixel “6” is equal to or less than the threshold value, and when this condition is satisfied, “7” on the right side of the starting point
The three pixels “0” and “1” are examined in this order, and the tracking is repeated until the density of the three pixels becomes equal to or less than the threshold.
When the lower right corner of the pin 202 is detected, the first condition is that the density values of the pixels “6”, “7”, and “0” among the pixels around the determination target point are equal to or less than the threshold value. When this condition is satisfied, the three pixels "5", "4", and "3" on the left side of the starting point are examined in this order, and tracking is repeated until the density of the three pixels becomes equal to or less than the threshold.

In each of the upper right corner, lower left corner, and lower right corner, if the deviation between the number of times of repetition of tracking and the pin width equivalent value is within a predetermined allowable range, the point to be determined is a corner candidate. Points NE, SW, and SE are assumed.

According to such a corner portion discrimination method, if the discrimination target point is noise, the number of repetitions is smaller than the pin width equivalent value and the deviation falls outside the allowable range. Erroneous determination can be reliably prevented. Also, if the determination target point is a corner portion NW ', NE', S of the main body 200 of the chip component 20,
In the case of W ′ and SE ′, the deviation becomes outside the allowable range due to the increase in the number of repetitions.
Even with the component 20 shown as 0, the corners of the pins 201 and 202 and the corners of the main body 200 are not confused, and the corners can be accurately determined.

Another Example of Method of Calculating Edge Points Based on Profiles Among the specific examples of the above-described recognition methods, there is a process of calculating edge points in sub-pixel units by interpolation based on a profile. In the description of the example, one-dimensional interpolation (see FIG. 12 and the above equation (5)) is performed.
In order to increase the accuracy of the calculation, it is even more desirable to perform two-dimensional interpolation as described below. The two-dimensional interpolation method of the present example is described as y = ax + b (a <
A case will be described in which a profile is taken on the straight line represented by 1) and coordinates of a point exceeding the threshold value are first obtained.

First, in the profile stage, when calculating the density value at each point where the x coordinate is an integer, a value that does not perform interpolation in the y direction according to the one-dimensional method (the integer value closest to the value for the x coordinate) , The density value at the position of the y-coordinate, and the density value indicated by a black circle in FIG. 22) are sequentially stored in a one-dimensional array.
In the method of the present example, the value of the y coordinate is set to a small number (x in FIG. 22).
Think at. Then, the density value F (x ₀ ) at the position x ₀ is

[0105]

F (x ₀ ) = F (x ₀ , [ax ₀ + b]) {1− (ax ₀ − [ax ₀ ])} + F (x ₀ , [ax ₀ + b] +1) (ax ₀ − [Ax ₀ ]) are sequentially stored in a one-dimensional array. However, in this equation, [ax ₀ + b] and [ax ₀ ] are respectively ax ₀ +
b, representing the maximum integer that does not exceed the value of ax _0. Also, F
(X ₀ , [ax ₀ + b]), F (x ₀ , [ax ₀ + b] +
1) are (x ₀ , [ax ₀ + b]), (x ₀ , [a
x ₀ + b] +1) is the density value of the pixel at each coordinate position.

Next, this one-dimensional array is sequentially examined, and i
Th first time When exceeding the threshold value θ in element d _i, obtains the coordinates (x _1, y ₁₎ as follows.

[0107]

Equation 7] _{x 1 = x s + i- (} d i -θ) / (d i -d i-1) y 1 = ax 1 + b However, x _s is the x coordinate of the start point of the profile.

As described above, in the method of the present embodiment, since the two-dimensional interpolation including the interpolation at the profile stage is performed, the process of obtaining the edge points from the profile in subpixel units can be performed with higher accuracy. Can be.

Another example of setting the start point and end point of profile measurement line S1 of steps S4, S6 in the first example of the recognition method
6, The setting method of the start point and the end point of the measurement line of the profile measurement line in the S18 processing is as described in the above (1) to (4).
Instead of the expression

[0110]

Xs = NW2x- (SW1x-NW2x) + offset Ys = NW2y- (SW1y-NW2y) Xe = SW1x + (SW1x-NW2x) + offset Ye = SW1y + (SW1y-NW2y); ) Instead of

[0111]

Xs = NE1x- (SE2x-NE1x) -offset Ys = NE1y- (SE2y-NE1y) Xe = SE2x + (SE2x-NE1x) -offset Ye = SE2y + (SE2y-NE1y); Instead of equation (13)

[0112]

Xs = NW1x- (NE2x-NW2x) Ys = NW1y- (NE2y-NW1y) + offset Xe = NE2x + (NE2x-NW1x) Ye = NE2y + (NE2y-NW1y) + offset of (NE2y-NW1y) + offset ) Instead of

[0113]

Xs = SW2x− (SE1x−SW2x) Ys = SW2y− (SE1y−SW2y) −offset Xe = SE1x + (SE1x−SW2x) Ye = SE1y + (SE1y−SW2y) −offset. Here, offset is a value registered in the library data.

[0114]

As described above, according to the present invention, the scanning line for scanning the inside of the processing window representing the image of the chip component is inclined in the directions inclined with respect to the X-axis direction and the Y-axis direction of the processing window. Is set, particularly preferably set to 45 °, and a corner of the chip component is detected by performing an oblique scanning along the scanning line while moving the scanning line in parallel, and based on the detection data, The position and inclination of are determined. For this reason, even when the image of the chip component is small, the corner can be easily and reliably detected, and the position and inclination of the chip component can be accurately obtained based on the corner.

In this method, when a chip component whose image has a rectangular shape or a similar shape is to be recognized, corner portions are detected by oblique scanning in order at four corner portions of the chip component, and the detection is performed. From the profile of the density value along the side profile measurement line set based on the data, the center point of the chip component on both sides is determined, and the midpoint of both side edge points on the center line passing through the two side center points By calculating the inclination of the center line and the center of gravity and inclination of the chip component, the position and inclination of this type of chip component can be obtained with high accuracy. Alternatively, two straight lines that intersect with the side profile measurement line and pass through a specific point are set, the midpoint of both side edge points on each straight line is obtained, and the center of gravity of the chip component and Accuracy can also be improved by obtaining the inclination.

When a chip component having a plurality of pins on both sides is to be recognized, the detection of a corner portion by oblique scanning is performed for the outer corners at both ends of the pin row on both sides of the chip component. From the profile of the density value along the profile measurement line set based on the detected data, the position of the center point of each pin on both sides is obtained, and the edge point of each pin corresponding to the center point of each pin is obtained. Based on these edge points, the centers of the pin rows on both sides are determined, and the center of the two pin rows and the slope of the straight line connecting the centers of the two pin rows are defined as the center of gravity and the slope of the chip component. By calculating, the position and inclination of this kind of chip component can be obtained with high accuracy.

When a chip component having a plurality of pins on at least one side is to be recognized, the detection of a corner by scanning in an oblique direction is performed outside the both ends of the pin row on one side of the chip component. The position of the center point of each pin on one side is obtained from the profile of the density value along the profile measurement line set based on the detection data, and the position of the corresponding edge point of each pin is determined for the corner. Then, based on these edge points, the inclination of the chip component is determined, and the center point of the pin row at one side edge and the corresponding edge point on the other side are determined to obtain the chip component. By determining the center of gravity, the position and inclination of this type of chip component can be accurately determined.

[Brief description of the drawings]

FIG. 1 is a schematic view showing an embodiment of a chip component recognition apparatus according to the present invention.

FIG. 2A is a perspective view of a chip resistor as a specific example of a chip component, and FIG. 2B is a diagram showing an image of the chip resistor.

FIG. 3 is a diagram showing an image of a tantalum capacitor as a specific example of a chip component.

FIG. 4 is a view showing an image of an aluminum capacitor as a specific example of a chip component.

FIG. 5A is a perspective view of a mini-transistor which is a specific example of a chip component, and FIG. 5B is a view showing an image of the mini-transistor.

FIG. 6 is a diagram showing an image of a power transistor as a specific example of a chip component.

FIG. 7 is a flowchart illustrating a first example of a recognition method.

FIG. 8 is an explanatory diagram showing a processing window.

FIG. 9 is an explanatory diagram showing a specific example of a method of scanning the inside of a processing window in an oblique direction.

FIG. 10 is an explanatory diagram showing a specific example of how to detect a corner portion.

FIGS. 11 (a), (b) and (c) are explanatory diagrams showing processing performed sequentially in the first example of the recognition method.

FIG. 12 is an explanatory diagram showing a profile of a density value and a method of subpixel formation.

FIG. 13 is an explanatory diagram showing profile measurement lines in a case where electrodes are vertically arranged in the first example of the recognition method.

FIG. 14 is an explanatory diagram schematically showing a second example of the recognition method.

FIG. 15 is a flowchart illustrating a third example of the recognition method.

FIGS. 16 (a) and (b) are explanatory views showing processing sequentially performed in the third example of the recognition method.

FIG. 17 is a part of a flowchart showing a fourth example of the recognition method.

FIG. 18 is another part of the flowchart showing the fourth example of the recognition method.

FIGS. 19 (a), (b) and (c) are explanatory views showing processing sequentially performed in the fourth example of the recognition method.

FIG. 20 is an explanatory diagram showing another example of determining a corner portion.

FIG. 21 is an explanatory diagram showing a tracking method used in the example shown in FIG. 20;

FIG. 22 is an explanatory diagram showing another example of a method of calculating an edge point based on a profile.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Component mounting head 2 Chip component 3 Camera 10 Image processing unit 12 Image memory 13 Processing means 14 Corner detection means 15 Calculation means

Claims (7)

(57) [Claims] An image of a chip component picked up by a component mounting head of a mounting machine is taken, the image is displayed in a processing window, and a position and a tilt of the chip component are determined based on scanning in the processing window. In the method for recognizing parts, a scanning line for scanning inside the processing window is set in a direction inclined with respect to the X-axis direction and the Y-axis direction of the processing window, and the scanning line is moved in parallel with the scanning line. The image data is determined, and the above scanning line is located at the corner of the chip component.
When the two sides sandwiching the vertex are crossed , each position on the two sides is detected by comparing the density value with a predetermined threshold value to detect a corner of the chip component, and the detected data of the corner is detected. A chip component recognition method, wherein a position and a tilt of the chip component are obtained based on the following. 2. A method for detecting a corner portion of the chip component, comprising :
Two sides where the scanning line sandwiches the apex of the corner of the chip component
To detect each position on the two sides when crossing
In addition, at the center side of the chip component from each position on the two sides
And two positions in a direction perpendicular to the direction of the scanning line.
2. The method according to claim 1, wherein the recognition is performed by determining the image data . Wherein the image is a recognition target chip components to be rectangular or analogous thereto shape, upper left corner definitive detection of the corner portions to the four corners on the processing window of the chip component and the lower-right corner To
Set the scanning line direction to the upper right and diagonally
The direction of the scanning line to the corner and upper right corner is lower right
To carry out sequentially set to gully obliquely, as the position and the process of obtaining the inclination of the chip components, each co
Among the detected data of the two positions of the Na portion, Contact to the side portion opposite to the one side portion on the upper side of the chip component and this
The left side profile measurement line based on each left detection data and the right side profile based on each right detection data
Section profile measurement line and these left,
Connects the center points of the profile measurement lines on the right side
From the core wire and the center point of this center line, the center of gravity of the chip component
And inclination, or the second place in each corner
Among the detected data location, and the upper side portion profile measuring line and one side portion on the left side of the chip component and which on the basis of the above the detection data definitive in the side portion opposite to each
Lower side profile measurement based on lower side detection data
The line and the calculated upper and lower side profile
And the center line connecting the center points of the measurement lines.
The method for recognizing a chip component according to claim 1 or 2, wherein the center of gravity and the inclination of the chip component are obtained from the center point of (1). Wherein the image is a recognition target chip components to be rectangular or analogous thereto shape, upper left corner definitive detection of the corner portions to the four corners on the processing window of the chip component and the lower-right corner To
Set the scanning line direction to the upper right and diagonally
The direction of the scanning line to the corner and upper right corner is lower right
To carry out sequentially set to gully obliquely, as the position and the process of obtaining the inclination of the chip components, each co
Among the detected data of the two positions of the Na portion, Contact to the side portion opposite to the one side portion on the upper side of the chip component and this
The left side profile measurement line based on each left detection data and the right side profile based on each right detection data
Seeking a part profile measuring line, a process of determining the specific point of the left and the right side portion profile on the measurement line by two points offset by a predetermined amount in the vertical <br/> sides than the respective center points, the left , a straight line passing through the specific points of the respective upper and intersects the right side portion profile measurement line, it
Set a straight line that passes through the specific point below each, and
The midpoint of both side edge points on the straight line is obtained, and the center of gravity and the inclination of the chip component are obtained based on each midpoint , or
Of the detection data at the two positions of the corner portion, the tip portion
One side on the left side of the product and the opposite side
Upper side profile based on the upper detection data
Based on the measurement line and the detection data of each lower side.
Find the side profile measurement line and the upper and lower sides
Left of each center point on the section profile measurement line
Two specific points offset by a predetermined amount on both right sides
Process and the top and bottom side profile measurement lines
A straight line that crosses the
Set a straight line passing through the specific point on the right side of each
Find the midpoint of both side edge points on both straight lines, and
3. The method for recognizing a chip component according to claim 1, wherein the center of gravity and the inclination of the chip component are obtained based on the obtained value. 5. When to be recognized a chip component having a plurality of pins on both upper and lower sides sides, the outer corner of the left end pin of the pin array of the detection of the corner portion of the upper side portion of the chip component and The rightmost pin of the pin row on the lower side
The direction of the scanning line with respect to the outer corners
And set the right end pin of the pin row on the upper side of the chip component.
Left end of pin row at outer corner and lower side of pin
Lower scan line direction to outer corner of pin
To perform set obliquely, as the position and obtains the inclination processing of the chip component, the right end in the upper side portion Pi
Of the detected data at the above two positions at the outer corner of the
Above the detection data on the right and the outer corner of the left pin
The detection data on the left side of the detection data at the two positions
Then, the profile measurement line in the upper pin row direction is obtained based on the above, and the position of the center point of each of the right and left ends of the pin on the obtained upper pin row direction profile measurement line is obtained.
Both are the outer corners of the right end pin on the lower side
Of the detection data at the right side of the detection data at the above two positions.
Data of the above two positions at the outer corners of the data and the left end pin.
Data based on the detection data on the left side of the
Find the profile measurement line in the pin row direction,
Left and right on the profile measurement line in the lower pin row direction
Processing to determine the position of the center point of each pin at the end,
Seeking the position of the edge point of each pin corresponding to the center point of each pin in the upper right end portions to determine the center of the pin array in the upper side portion on the basis of the edge point of each pin, the lower
Corresponding to the center point of each pin at the left side and the bottom right side
Find the position of the edge point of the pin, and find the edge point of each pin
Of Finding the Center of the Pin Row in the Lower Side Based on
And the center of the pin row and the lower side of these upper sides.
A slope of a straight line connecting the middle point and the two pin rows center and the center of definitive pin rows, performs a process of determining a centroid and the inclination of the chip component, on the other hand, a chip having a plurality of pins on the left and right both side portions Parts
If the target is to be recognized, use the chip to detect the corner
Outer corner of the top pin of the pin row on the left side of the part
The outer core of the pin row at the
The scanning line direction to the corner
Outside the lower end pin of the pin row on the left side of the chip component
Outside the upper end pin of the pin row at the corner and right side
Set the scanning line to the right corner
And determine the position and inclination of the chip component.
The process is performed outside the lower end pin on the left side.
On the lower side of the detection data of the above two positions of the corner,
Detection data and the above two positions of the outer corner of the upper end pin
Based on the upper detection data of the detection data of
To find the profile measurement line in the left pin row direction,
Determined profile measurement line in left pin row direction
Find the position of the center point of each pin at the top and bottom,
The outer corner of the lower end pin on the right side
Upper and lower detection data of two positions
The detection data at the above two positions at the outer corners of the end pin
The pin array on the right side based on the detection data on the upper side
Direction profile measurement line, and find the right side
Upper and lower pins on the profile measurement line in the pin row direction
Process to determine the position of the center point of the
Of the edge point of each pin corresponding to the center point of each pin in
Find the position, and based on the edge point of each pin, the left side
Find the center of the pin row in the
The edge of each pin corresponding to the center point of each pin at the lower right corner
The edge point of each pin
Processing to find the center of the pin row on the right side
The center of the pin row on the left side and the pin on the right side.
A straight line connecting the center of the pin row and the center of these two pin rows
Find the slope of the line as the center of gravity and slope of the chip component
3. The method for recognizing a chip component according to claim 1, wherein the process is performed . 6. A case in which a chip component having a plurality of pins on at least one of the upper and lower sides is to be recognized.
It is run against the detection of the corner portion on the outer corner of the left end pin of the pin array on the one side portion of the chip component
Set the direction of the check line to diagonally upward or diagonally right
And, the direction of the scan lines against the outer corner portion of the right end pin
The processing is performed by setting the angle to the lower right or the upper right .
The above 2 of the outer corner portion of the right end pin in the side portion
Among the position detection data, the detection data on the right side and the left end
Of the detection data at the above two positions at the outer corner of the pin,
The profile measurement line in the pin row direction is obtained based on the detection data on the left side of the above, the position of the center point of each pin on the left and right ends on the obtained profile measurement line in the pin row direction is obtained, and the center of each pin is obtained. A process of obtaining the position of the edge point of each pin corresponding to the point; a process of obtaining the inclination of the chip component based on the arrangement of the edge point of each pin; From the density profile on the center line obtained based on the center of the pin array at the edge of one side and the inclination of the chip component, the other side on the center line is obtained. detecting an edge point parts side performs the processing for obtaining the center and the center line of the center of gravity of the chip components from an edge point of the other side portion of the pin rows at the edge of the one side portion, while at least A plurality of pins on one side portion which is one of the right
If the chip component with the
Detection of the tip part at the one side of the chip component.
Scan line direction with respect to the outer corners of the top row of pins
Is set to diagonally upward or diagonally downward,
The direction of the scanning line with respect to the outer corner of
Or, perform the operation by setting the angle to the upper right.
The process for determining the position and inclination of the product
Detection data of the above two positions at the outer corners of the lower end pin
Data on the lower side of the data and the outer
Above the detection data at the above two positions
Pin row direction profile measurement based on detected data
Find the line and measure the profile in the direction of the pin row.
Find the positions of the center points of the upper and lower pins on the
The position of the edge point of each pin corresponding to the center point of each pin
Based on the desired process and the location of the edge points for each pin
Processing to determine the inclination of chip components and the edge of each pin
Center of pin row at one side edge based on point
And the center of the pin row at this one side edge
The darkness on the center line obtained based on the inclination of the chip component
From the profile of the degree value, the other side on this center line
An edge point is detected, and a pin at the edge of the one side portion is detected.
From the center of the row and the edge point on the other side on the center line
3. The method for recognizing a chip component according to claim 1 , further comprising: performing a process of obtaining a center of gravity of the chip component. 7. An image pickup means for picking up an image of a chip component adsorbed on a component mounting head of a mounting machine, an image storage means for storing an image of the chip component picked up by the image pickup means, and an image from the image storage means. And processing means for scanning the processing window representing the image and performing arithmetic processing based on the scanning line. The processing means scans the processing window for the X-axis direction and the Y-axis direction of the processing window. The scanning line is set to be inclined with respect to the direction, the scanning line is moved in parallel, the density value of the image on the scanning line is determined, and the scanning line sandwiches the vertex of the corner of the chip component
On the two sides where the density value changes by crossing the two sides
Each position of the corner detecting means for detecting and comparing the density value and a predetermined threshold value, it has a calculating means for calculating the centroid position and inclination of the chip component on the basis of the detection data of the corner portion An apparatus for recognizing chip components.