JPH034376A - Shape recognizing method - Google Patents

Abstract

PURPOSE:To recognize the specific part of an object to be recognized from roundness by forming light cutting plane shape on the object to be recognized by irradiating an almost cylindrical shape object to be recognized with light in a direction intersecting orthogonally to the longitudinal direction of the object, and finding the roundness by finding a circular arc approximated to the light cutting plane shape. CONSTITUTION:When a seedling 3 as the object to be recognized is illuminated with slit light 5 from a slit light source 4 arranged on a placing board 2, the light 5 is formed in the light cutting plane shape along the shape of the seedling 3. When the cutting plane shape is image-picked up with an image pickup device 6, an image signal from the device 6 is sent to an image processing device 10, and is A/D-converted 11, then, it is stored in an image memory 12. Next, a pre-processing part 14 performs the smoothing and thinning processings of image data, etc., with the command of a main control part 13. And the circular arc approximated to each of the light cutting plane shape S1, S2,... Sn of obtained cutting plane shape is found with a circular arc matching part 15, and the roundness of the circular arc is found with a roundness calculating part 16, then, the shape of the specific part of the seedling 3 can be obtained from the roundness.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a shape recognition method for recognizing a cross section of a cylindrical object, such as a plant market.

(Prior Art) Such shape recognition is applied to, for example, selecting the size of asparagus as an object to be recognized.

The size of the asparagus is sorted by placing an ITV (industrial television) camera above the conveyor conveying the asparagus to image the asparagus being conveyed, and using the image data obtained by this imaging. From this, the maximum diameters of the asparagus tip and trunk are determined, the ratio of these maximum diameters is determined, and the length of the asparagus is determined. Then, the size of asparagus is determined by comparing the ratio and length of these maximum diameters with a reference value.

When this method is used for shape recognition of seedlings, etc., each of the above methods obtains image data using one camera, so this image data becomes two-dimensional information about the seedlings.

Therefore, although it is possible to obtain two-dimensional width and length information of the stem of a seedling, it is not possible to obtain three-dimensional shape change information on the image surface.

Therefore, it is difficult to recognize the parts of the seedling where the techniques are divided or the parts covered with leaves from information about three-dimensional shape changes on the image surface.

(Problems to be Solved by the Invention) As described above, it is difficult to recognize specific shaped parts such as cross sections of objects to be recognized, parts where techniques are divided, and parts covered with leaves.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a shape recognition method that can recognize a specific shape portion of an object to be recognized.

[Structure of the Invention] (Means and Effects for Solving the Problems) The present invention provides a method for detecting one object by irradiating one-dimensional light in a direction intersecting the longitudinal direction of a substantially cylindrical object. A light section shape is formed on the top, an arc that approximates this light section shape is found, the circularity of the light section shape with respect to the arc is determined, and a specific shaped part of the object to be recognized is determined from this roundness. This is a shape recognition method that recognizes shapes.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of a shape recognition device to which the shape recognition method of the present invention is applied. A mounting table 2 is provided on the XY child table, and a plant seedling 3 as an object to be recognized is placed on this mounting table 2. Note that the seedlings 3 are arranged so that the direction in which they extend coincides with the X direction. On the other hand, a slit light source 4 is arranged above the mounting table 2, and this slit light source 4 emits slit light 5 in the X direction, that is, a direction substantially perpendicular to the direction in which the seedlings 3 extend, onto the surface of the mounting table 2. The beam is irradiated at an angle of ψ degrees. Further, an imaging device 6 is placed directly above the city 3 where the slit light 5 is irradiated to take an image of the seedling 3. By the way, X
The Y child table receives a drive control signal from the movement control section 7, and the X table moves in the X direction at predetermined intervals.

The image signal output from the imaging device 6 is sent to a recognition processing device 10. This recognition processing device 10 has an A/D
An (analog/digital) converter 11 and an image memory 12 are provided, and the image signal from the imaging device 6 is converted into a digital image signal by the A/D converter 11 and stored in the image memory 12 as image data. It has become.

Further, this recognition processing device 10 is equipped with a main control section 13, and this main control section 13 includes an image memory 12, a preprocessing section 14,
An arc matching section 15, a roundness calculating section 16, a trunk determining section 17, and an output section 18 are connected.

The preprocessing unit 14 has a function of performing smoothing processing, -th order differential processing, binarization processing, and thinning processing on the image data stored in the image memory 12. The arc matching section 15 has a function of finding the arc that most closely approximates the shape of the optical cross section of the image data stored in the image memory 12 by the method of least squares, and the circularity calculating section 16
has the function of determining the circularity of the optical section shape with respect to the arc as an evaluation function. In addition, the trunk determining unit 17 determines the specific shaped portion of the seedling 3, that is, the stem of the seedling 3, based on the roundness.
It has the function of identifying branched parts and parts covered with leaves.

Next, the operation of the apparatus configured as described above will be explained. When the seedling 3 is placed on the table 2 and the slit light 5 is irradiated onto the seedling 3 from the slit light source 4, the slit light 5 is formed into a light section shape that follows the shape of the seedling 3. Then, when this light section shape is imaged by the imaging device 6,
The image signal output from this imaging device 6 is sent to an image processing device 10. This image processing device 10 converts the image signal into A
The /D converter 11 converts the signal into a digital image signal and stores it in the image memory 12 as image data.

When the image data is thus stored in the image memory 12, the main control section 13 issues an operation command to the preprocessing section 14. The preprocessing section 14 performs smoothing processing, -th order differential processing, binarization processing, and thinning processing on the image data to obtain the image data shown in FIG. 2.

Next, the main control unit 13 determines whether image capture has been completed for the entire seedling 3, and if it has not been completed, the main control unit 13 sends an
Issues a command to move the table. As a result, the X table of the XY child table moves by a predetermined distance. Hereinafter, as shown in FIG. 3, three-dimensional information of the light section shapes Sl, S2...S, at different positions in the seedling 3 is obtained in the same way as the action explained above, and finally the seedling 3 The shape of the optical cross section at each position in the entire area is obtained.

When the optical section shape of the entire seedling 3 is obtained in this way, the main control section 13 gives an operation command to the arc matching section 15. In response to this command, the arc matching section 15
is each light section shape Sl.

The most approximate circular arc for each of S2...S is determined by the least squares method. Here, how to obtain the circular arc will be explained with reference to FIG. 4, taking the optical section shape Sl as an example. Here, the center position of the arc Q is (x, y) and the radius is r. Further, let R be the distance from the center position (X, Y) of the circular arc Q to the end position (Xi, Yi) in the optical section shape S.

However, the distance R is It is represented by R-(Xi-x)'+(Yi-y)2.

Then, E-Σ(r-R)2 is set, and Δr1Δx1Δy is solved as follows. That is, by using a mathematical method to find a solution to a nonlinear equation, let XsY be the solution. As a result, the radius r of the arc Q and its center position (x, y) can be found.

Next, the main control section 13 gives an operation command to the roundness calculation section 16. The circularity calculation unit 16 calculates an evaluation function H representing the circularity of the optical section shape S1 from the arc Q and the optical section shape S. That is, this evaluation function H is expressed as H-(r-R)2/r2. If the evaluation function H has a small value, it indicates that the outline of the light section shape SI is close to a perfect circle, and if the value has a large value, it indicates that the outline of the light section shape S1 is far from a perfect circle.

Once the evaluation function H is determined in this way, the main control section 13 gives an operation command to the executive determination section 17. This executive judgment part 1
7 compares the evaluation function H with a reference value for determining the stem of the seedling 3 to determine the stem, divided portion, and portion covered by leaves in the seedling 3, respectively. That is, the trunk determining unit 17 determines that the light-cut shape S is formed on one stem when the roundness is large, and determines that the light-cut shape S is formed on a single stem when the roundness is small. It is determined that it is formed on a branched part or a part covered by leaves.

However, the above operation is performed for each light cutting shape S2. S,
. . . It is performed every S7 to obtain the evaluation function H. Figure 5 shows the results of determining the evaluation function H for the entire seedling 3. From this result, each node 1mA, B, and C with a large value of the evaluation function H is found to be covered by the dividing part of W3 or leaves. become a part.

In this way, in the above embodiment, a slit light is irradiated onto the seedling 3 to form a light section shape, and an arc Q that approximates this light section shape is determined to determine the light section shape. The roundness of the seedling 3 with respect to the arc Q is calculated, and the stem, branching parts, and parts covered by leaves of the seedling 3 are recognized from this roundness. It is possible to determine the branching parts and parts covered by leaves and extract their positions. Furthermore, the diameter at each position of the seedling 3 can be determined.

Note that the present invention is not limited to the above-mentioned embodiment, and may be modified without departing from the spirit thereof. In the above-mentioned embodiment, a case has been described in which the present invention is applied to the recognition of the stem of Ichi 3, but the present invention may also be applied to the case where the object to be recognized is asparagus and the diameter of the tip and stem is recognized. Moreover, soldering may be applied to detection of defects and welding defects as objects to be recognized. Furthermore, other means may be used to determine the roundness.

[Effects of the Invention] As described in detail above, according to the present invention, it is possible to provide a shape recognition method that can recognize a specific shape portion of an object to be recognized.

[Brief explanation of drawings]

1 to 5 are diagrams for explaining the shape recognition method according to the present invention, in which FIG. 1 is a block diagram of a shape recognition device to which this method is applied, and FIG. 2 is a preprocessing diagram. FIG. 3 is a schematic diagram of the later image data, FIG. 3 is a schematic diagram of the shape of a light section, FIG. 4 is a schematic diagram for explaining calculation of an evaluation function, and FIG. 1...XY child table 2...Placement stand, 3...Seedling,
4... Slit light source, 6... Imaging device, 7... Movement control unit, 10... Recognition processing unit, 11... A/D converter, 12... Image memory, 13...・Main control unit, 1
4... Pre-processing section, 15... Arc matching section, 16
... Roundness calculation section, 17... Executive judgment section.

Claims (1)

[Claims] A one-dimensional light is irradiated in a direction intersecting the longitudinal direction of a substantially cylindrical object to be recognized to form a light section shape on the object to be recognized, and an approximation is made to this light section shape. A shape recognition method characterized by determining a circular arc, determining the circularity of the optical section shape with respect to the circular arc, and recognizing a specific shaped portion of the object to be recognized from this circularity.