JP2712962B2 - Work identification method and device - 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 identifying a workpiece, and more particularly, to a method for identifying whether or not a workpiece is a target workpiece based on an image of the workpiece obtained by an image detection unit. And its device.

[0002]

2. Description of the Related Art Conventionally, it has been proposed to use a feature value to identify whether or not a work is a desired work. Here, as feature values that may be generally adopted,
Number of holes, number of corners, area, total area, maximum hole area, minimum hole area,
Thirteen types of characteristic values are known, such as a perimeter, a maximum radius, a minimum radius, a major axis length of an inertia ellipse, a minor axis length of an inertia ellipse, a circumscribed rectangle width, and a circumscribed rectangle length. Therefore, by obtaining variances for these 13 types of characteristic values and identifying whether or not the work is a desired work, it is possible to achieve a work identification with extremely high accuracy.

[0003]

However, obtaining the variance for each of the 13 types of characteristic values causes a significant increase in the processing load, and the time required for work identification is greatly increased. There are inconveniences. Therefore, due to the tact time, it cannot be put to practical use in the field of component assembling where improvement in processing speed is strongly required.

In order to solve such inconveniences, the present inventors set the feature values to be preferentially selected in advance, instead of equally weighting the 13 types of feature values. Only when the work identification accuracy based on the set feature values is not high, the feature value with the highest work identification accuracy is selected from the extracted feature values, and the average value and variance of the selected feature values are selected. Based on the above, a method of obtaining the upper and lower thresholds and identifying the target work was considered, and the work was actually identified. As a result, the time required for work identification can be greatly reduced, and a considerably high work identification accuracy can be achieved. However, in this method, since the discrete value feature value and the continuous amount feature value are handled in the same way without any distinction, the discrete value feature value is not selected as the workpiece identification feature value in most cases. There is. That is, the discrete feature value generally shows a distribution with a wide tail. For example, if attention is paid to the number of holes as the discrete feature value, the work having the number of holes n and the work of n + 1 are high. Accuracy can be identified. However, since the tail tends to show a wide distribution, when selecting the feature value with the largest variance ratio, the possibility of selecting a discrete-valued feature value is significantly reduced, and instead, high-precision identification is achieved. A continuous quantity feature value that is difficult to select is selected. As a result, loads such as calculation of a threshold value for work identification and discrimination of the accuracy of work identification become large, and the time required for work identification becomes long.

[0005]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and can reduce a calculation load for work identification, reduce the number of characteristic values for work identification, and reduce work identification. It is an object of the present invention to provide a work identification method and a device capable of significantly improving accuracy.

[0006]

According to a first aspect of the present invention, there is provided a work identification method for extracting a feature value of a plurality of teaching works, and extracting an area value and a feature value of the extracted feature values. Select a value that defines the surrounding contour, and set the area value and the value that defines the surrounding contour when the accuracy of identifying the workpiece by the area value corresponding to the type of the teaching work and the value that defines the surrounding contour is high. If the work value is selected as the work identification feature value and the work identification accuracy based on the area value and the value defining the surrounding contour is low, the feature value with the highest work identification accuracy is selected from the discrete value feature values, and the discrete value is selected. It is determined whether or not the ranges based on the upper and lower thresholds for each work obtained based on the average value and the variance of the feature values selected from the value feature values are overlapped. , Continuous quantity Work identification accuracy among Chochi selects a feature value is the largest, is a method for identifying a target work based on the lower threshold finally selected feature values and corresponding.

According to a second aspect of the present invention, there is provided a work identification apparatus for extracting a characteristic value of a plurality of teaching works, and selecting a value defining an area value and a surrounding contour from the extracted characteristic values. First selecting means, and first determining means for determining whether or not the identification accuracy of the work based on the area value corresponding to the type of the teaching work and the value defining the surrounding contour is high;
When the identification accuracy of the workpiece by the value defining the area value and the surrounding contour is high, the value defining the area value and the surrounding contour is selected as the feature value for work identification, and the value defining the area value and the surrounding contour is selected. A second selecting unit that selects a feature value having the maximum work identification accuracy from the discrete value feature values when the work identification accuracy is low; an average value of the feature values selected from the discrete value feature values; Second determining means for determining whether or not ranges based on upper and lower thresholds for each work obtained based on the variance overlap each other; The method includes a third selection unit that selects a feature value having the highest accuracy, and a target work identification unit that identifies a target work based on the finally selected feature value and the corresponding upper and lower thresholds.

[0008]

According to the work identification method of the present invention, a value defining an area value and a peripheral contour is first selected from a plurality of feature values of a teaching work as feature values for work identification.
If the area value and the value defining the surrounding contour can achieve a sufficient work identification accuracy, the work identification can be achieved with high accuracy using these characteristic values as work identification characteristic values. Also, if the area value and the value defining the surrounding contour cannot achieve sufficient workpiece identification accuracy,
Select the feature value with the largest variance ratio from the discrete value feature values,
If the selected discrete value feature value can achieve sufficient work identification accuracy, the selected discrete value feature value can be used as a work identification feature value to achieve high-precision work identification.
Further, if the selected discrete value feature value does not achieve sufficient work identification accuracy, the feature value having the largest dispersion ratio is selected from the continuous value feature values, and the selected continuous value feature value is selected. Is used as a feature value for work identification.

Therefore, the possibility that the feature value having the largest variance ratio must be selected from the continuous-quantity feature values is significantly reduced, and the processing load for work identification can be reduced. In addition, since discrete value feature values that are easy to determine the validity of work identification are selected in preference to continuous quantity feature values,
The accuracy of work identification can also be improved, and the types of characteristic values required for work identification can be significantly reduced.

According to the work identification device of the present invention, the characteristic values of a plurality of teaching works are extracted by the characteristic value extracting means, and among the extracted characteristic values, the values defining the area value and the surrounding contour are extracted. Selection is made by the first selection means, and whether or not the identification accuracy of each of the workpieces is high is determined by the first determination means. If the discrimination accuracy of the work is not high as a result of the discrimination, the feature value having the largest variance ratio is selected from the discrete value feature values by the second selecting means, and whether or not the discrimination accuracy of the work is high is determined. The determination is made by the second determination means. As a result of this determination, if the work identification accuracy is not high, the feature value having the largest variance ratio is selected from the continuous quantity feature values by the third selecting means. In addition, as a result of the above determination, when the identification accuracy of the work is high, the corresponding feature value is used. Otherwise, the target work is highly accurately identified by the target work identification unit using the selected continuous amount feature value. it can.

[0011] Of the above feature values, the discrete value feature value is:
Unlike the continuous feature value, it is not necessary to set upper and lower thresholds based on the average value and the variance. For example, it is only necessary to select the maximum value and the minimum value, and the process for setting the upper and lower threshold values is significantly simplified. And the process of determining the validity of the work identification based on the set upper and lower thresholds can be significantly simplified, so that the overall process can be simplified,
Moreover, the feature value for identifying the work can be reduced, and the work can be identified with sufficient accuracy.

[0012]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. FIG. 7 is a diagram schematically showing the configuration of a component supply system to which the work identification method of the present invention is applied, in which a part of the component transport track 11a of the bowl feeder 11 is formed of a translucent plate 11b. Lighting devices 12 above and below the translucent plate 11b, respectively.
a and 12b are arranged. And the plate 1 with translucency
1b and a lower mirror 12b between the lower lighting device 12b.
c, the light from the translucent plate 11b (the silhouette light of the work by the illumination device 12a and the illumination device 12
The image detection unit 13 is arranged so that the reflected light of the work by the light source b is reflected by the half mirror 12c and received by the image detection unit 13 including a CCD camera or the like. Further, a suction head 14a for gripping a work is mounted on the industrial robot 14. Further, the image detecting unit 13
An image processing unit 13a that performs a binarization process or the like by using an image signal captured from the computer as an input, and a robot controller 14b that receives a processing result signal from the image processing unit 13a and gives an operation command to the industrial robot 14. doing.

FIGS. 1 to 3 are flowcharts showing an embodiment of a work identification method according to the present invention. FIG. 1 shows a teaching process for a work to be identified, FIG. 2 shows a feature value selection process for work identification, and FIG. FIG. 3 shows a work identification process based on the selected feature value. In the flowchart of FIG. 1, in step SP11, an image of a work of the same type of n samples is captured by an image detection unit 13 such as a CCD camera, and in step SP12, a feature value of the corresponding work is extracted based on the captured image. In step SP13, the area value and the circumscribed rectangle length are selected from the extracted characteristic values, and the respective variances (for example, standard deviation σ) and the average value xav are calculated in step SP14, and the THu is calculated in step SP15.
p = xav + kσ and THdn = xav-kσ (where k is a constant of about 1 to 3) to calculate the area value,
Upper and lower thresholds THup and TH of the circumscribed rectangle length
dn is calculated.

Note that, from step SP11 to step SP1
A series of processing up to 5 is performed for each type of work. In the flowcharts of FIGS. 2 and 3, step SP2
In 1, it is determined whether or not ranges determined based on the upper and lower threshold values THup and THdn of the respective feature values for the two types of workpieces overlap each other. If it is determined that the ranges overlap, step SP22 is performed. In the two types of works, the dispersion ratio F0 of discrete feature values (for example, the number of holes, the number of images in the case of front / back discrimination, etc.)
Is selected in step SP23, and the maximum value and the minimum value of the discrete value characteristic values selected in step SP24 are selected as the upper threshold value THup and the lower threshold value THdn, respectively.
At step SP25, it is determined whether or not the ranges of the upper and lower thresholds overlap each other. If it is determined that the ranges overlap, at step SP26, whether or not the processing for all the discrete value feature values has been completed Is determined, and if not completed, the discrete value feature value having the next largest dispersion ratio is selected in step SP27, and the process in step SP24 is performed again. If it is determined in step SP26 that the processing for all the discrete value feature values has been completed, the variance ratio F0 of each continuous quantity feature value is calculated for the two types of workpieces in step SP28, and the variance ratio is maximized in step SP29. Is selected, the upper threshold THup and the lower threshold THdn are calculated based on the variance and the average of the continuous quantity feature values selected in step SP30, and the ranges of the upper and lower thresholds overlap in step SP31. It is determined whether or not they overlap, and if it is determined that they overlap, it is determined in step SP32 whether or not one range is completely included in the other range. If it is determined in step SP21, SP25 or SP31 that they do not overlap, step SP21
At 33, the corresponding feature value is adopted as the work identification feature value, and the corresponding upper and lower thresholds are adopted.
If it is determined in step SP32 that one range is not completely included in the other range, the upper and lower thresholds are changed in step SP34 to eliminate the overlap, and the corresponding continuous amount feature value is changed. It is adopted as a feature value for work identification. Conversely, step SP3
If it is determined in step 2 that one range is completely included in the other range, in step SP35, it is instructed that the width of the upper and lower thresholds should be reduced and re-teaching should be performed.

In the flowchart of FIG. 4, in step SP41, an image of the work to be identified is fetched, and in step SP42, the characteristic value selected by the processing of FIG. 2 is obtained. Then, in a step SP43, it is determined whether or not the area value is a value between the upper threshold and the lower threshold, and in a step SP44, it is determined whether or not the circumscribed rectangle length is a value between the upper threshold and the lower threshold. If it is determined in both steps SP43 and SP44 that the value is between the upper limit threshold and the lower limit threshold, the upper limit threshold and the lower limit threshold are calculated in addition to the area and the circumscribed rectangle length in step SP45. It is determined whether or not there is a characteristic value present, and if it is determined that there is a characteristic value, it is determined in step SP46 whether or not the corresponding characteristic value is a value between the calculated upper threshold and lower threshold. Is determined.

If it is determined in step SP45 that there is no characteristic value, or if it is determined in step SP46 that the corresponding characteristic value is a value between the calculated upper and lower thresholds, Step SP
At 47, the corresponding work is identified as the work to be processed. Conversely, step SP43, step SP44
Alternatively, if it is determined in step SP46 that the value is not between the upper threshold and the lower threshold, the process proceeds to step SP4
In step 8, the corresponding work is identified as a disturbance work.

More specifically, when the distribution shown in FIG. 5A is obtained for the work A and the work B as a result of performing the processing based on the flowchart of FIG. 1, both distributions overlap. Since no part exists, for example, for work A, as shown in FIG. 5, the upper threshold THup and the lower threshold THd
By setting n and determining whether or not the characteristic value obtained for the work to be processed in the flowchart of FIG. 3 is within the range of the upper threshold value THup and the lower threshold value THdn, highly accurate workpiece identification can be achieved. Conversely, FIG.
Work A resulting from processing based on the flowchart of
When the ranges of the upper and lower thresholds calculated based on the distributions obtained for the work B and the work B overlap each other, the discrete value feature values are selected in order from the one with the largest dispersion ratio F0. Then, based on the selected feature value, if the range between the upper threshold value THup and the lower threshold value THdn does not overlap as shown in FIG. 4B, the corresponding discrete value feature value and the corresponding upper and lower threshold value are set. It is possible to achieve high-precision workpiece identification by using this. Conversely, if the ranges of the upper and lower thresholds overlap in any of the discrete value feature values, the continuous amount feature value with the largest variance ratio F0 is selected and corresponds to the selected continuous amount feature value. If the ranges of the upper and lower thresholds do not overlap with each other, the work can be identified with high accuracy using the corresponding continuous amount feature value and the corresponding upper and lower thresholds. Conversely, when the distribution shown in FIG. 4C is obtained for the selected continuous amount feature value, there is a considerable overlap between the ranges of the upper and lower threshold values THup and THdn determined based on both distributions. Therefore, even if the processing of the flowchart of FIG. 3 is performed, highly accurate workpiece identification cannot be achieved. Therefore, in this case, the upper and lower thresholds are changed to eliminate the overlap (FIG. 4).
(C) Refer to the dashed line in the middle) A highly accurate work identification can be achieved. However, if one range is completely included in the other range, it is impossible to identify the work. Therefore, the width of the upper and lower thresholds is reduced, and teaching is performed again. To obtain the optimal feature values and the corresponding upper and lower thresholds.

That is, when a work is to be identified,
Unlike characters, figures, and the like, it is often possible to achieve high-precision work identification based on the area value and the circumscribed rectangle length. Therefore, there is no need to select another feature value based on the flowchart of FIG. Therefore, another characteristic value may be selected for only some of the workpieces based on the flowchart of FIG. 2, so that the processing load can be reduced as a whole. Also in the processing of FIG. 2, the upper and lower thresholds are extremely easy to set, and the discrete value feature value that is extremely easy to determine the validity of the work identification is preferentially selected. Even if sufficient work identification accuracy cannot be achieved depending on the length of the rectangle, it is quite often possible to achieve sufficient work identification based on discrete value feature values (for example, discrimination of different types of work based on silhouette images, reflection, etc.). As a result, the processing load can be further reduced as a whole. In addition, when there is an overlap between the ranges of the upper and lower threshold values THup and THdn determined based on the distribution of the continuous amount feature value, the upper and lower threshold values are changed to eliminate the overlap (the one-dot chain line in FIG. 4C). ) Can achieve highly accurate workpiece identification.

When there are three or more types of workpieces, the processing of the flowchart of FIG. 2 may be performed for each workpiece. In addition, instead of the circumscribed rectangle length, the circumscribed rectangle width and the inertia ellipse major axis length are used. Or the inertia elliptical minor axis length can be used, and various other design changes can be made without departing from the scope of the present invention.

[0020]

[Embodiment 2] FIG. 6 is a block diagram showing an embodiment of a work identifying apparatus according to the present invention, wherein an image detecting section 13 for taking an image of a work and image taking by the image detecting section 13 are repeated a predetermined number of times. A first iteration control unit 2, a feature value extraction unit 1a for extracting a feature value based on a captured image, and selecting an area value and a circumscribed rectangle length from among the extracted feature values for all works. A first calculator 3 for calculating the standard deviation σ and the average value xav by using the standard deviation σ and the average value xav calculated by the first calculator 3.
Hup = xav + kσ and THdn = xav−kσ
(Where k is a constant of about 1 to 3), a first threshold value calculation unit 3a that calculates an upper threshold value THup and a lower threshold value THdn, and a range of upper and lower threshold values corresponding to the area value and the circumscribed rectangle length. A feature value determining unit 4 for determining whether or not these are appropriate as feature values based on whether or not they overlap each other; and determining that the area value and the circumscribed rectangle length are inappropriate as feature values. A discrete value feature value selecting unit 5 for selecting discrete value feature values in order from the one with the largest variance ratio on the condition that the dispersion ratio is the largest, and setting the maximum value and the minimum value related to the selected discrete value feature value to an upper threshold THup and a lower threshold THdn. And a discrete value feature value determination unit that determines whether or not the adopted upper and lower threshold ranges are appropriate as feature values based on whether or not the ranges of the adopted upper and lower threshold values overlap each other. In response to the discrimination result from the discrete value feature value discriminating unit 6 indicating that they overlap, the discrete value feature value selection is performed on condition that there is a discrete value feature value that is not selected. A repetition control unit 6a for repetitively operating the unit 5, and a continuous amount feature value having a maximum dispersion ratio in response to a control signal from the repetition control unit 6a indicating that there is no unselected discrete value feature value. The continuous quantity feature value selection unit 7 to be selected, and the selected continuous quantity feature value standard deviation σ
And a second calculating unit 7a for calculating an average value xav,
The standard deviation σ and the average value x calculated by the calculation unit 7a
THup = xav + kσ and THdn based on av
= Xav-kσ, to calculate an upper threshold THup and a lower threshold THdn, a second threshold calculator 7b,
A continuity feature value discriminating unit 8 for discriminating whether or not ranges of upper and lower threshold values corresponding to the area value and the circumscribed rectangle length are appropriate as feature values based on whether or not the ranges overlap; When the ranges of the upper and lower thresholds calculated by the two threshold calculator 7b overlap each other, a threshold changing unit 8a that changes the upper and lower thresholds to eliminate the overlap, and an image captured by the image detector 1. A feature value output unit 9 that obtains and outputs an area value, a circumscribed rectangle length, and a selected feature value when applicable, a feature value output unit, a first threshold value calculation unit 3a, and a second threshold value calculation Part 7
Upper threshold, lower threshold, threshold setting unit 5 calculated by b
A signal indicating that the workpiece is a processing target by determining the magnitude of the upper threshold, the lower threshold, or the upper threshold and the lower threshold changed by the threshold changing unit 8a. Or a work discriminating unit 10 that outputs a signal indicating that the work is a disturbance work. Further, it is needless to say that the first calculation unit 3 and the second calculation unit 7a may be used together, and the first threshold value calculation unit 3a and the second threshold value calculation unit 7b may be used together.

The operation of the work identification device having the above configuration is as follows. First, an image of a work with n samples is fetched by the image detecting unit 13 for each type of work, and the standard calculation unit 3 calculates the standard deviation σ and the average value of the area value and the circumscribed rectangle length for each work type by the first calculating unit 3. xav is calculated, and based on the calculated standard deviation σ and the average value xav, the first threshold value calculation unit 3a calculates THup = xav +
The calculation of kσ and THdn = xav−kσ is performed to calculate the upper threshold THup and the lower threshold THdn. Then, the characteristic value determination unit 4 determines whether the area value and the circumscribed rectangle length are appropriate as the characteristic values, that is, whether the ranges of the upper and lower thresholds overlap each other.

When the characteristic value discriminating unit 4 determines that the area value and the circumscribed rectangle length are appropriate as the characteristic values, the image detecting unit 13 captures an image of the work to be identified, and the characteristic value output unit 9 outputs. The area value and the circumscribed rectangle length are obtained and output, and the work discriminating unit 10 calculates the magnitude of the characteristic value output from the characteristic value output unit 9 and the upper and lower thresholds calculated by the first threshold calculator 3a. A signal indicating that it is a work to be processed or a disturbance work is output. That is, when both output feature values are within the range of the corresponding upper threshold value and lower limit threshold value, a signal indicating that the work to be identified is the work to be processed is output, and at least one feature value is within the above range. If it is outside, a signal indicating that the identification target work is a disturbance work is output.

Conversely, when the characteristic value discriminating unit 4 determines that the area value and the circumscribed rectangle length are not appropriate as the characteristic values, the discrete value characteristic value selecting unit 5 selects the discrete value characteristic having the maximum dispersion ratio. The threshold value setting unit 5a selects a maximum value and a minimum value of the selected discrete value feature value as an upper threshold value THup and a lower threshold value THdn, and the upper and lower limits adopted by the discrete value feature value determination unit 6. It is determined whether or not these threshold values are appropriate as feature values based on whether or not the threshold ranges overlap. If the discrete value feature value discriminating unit 6 determines that there is no overlap, the iterative control unit 6a selects a discrete value feature value on the condition that there exists a discrete value feature value that is not selected. The unit 5 is repeatedly operated. That is, it is determined whether or not the ranges of the upper and lower thresholds overlap in order from the discrete value feature value having the highest variance ratio. As a result of this iterative operation, the processing based on all the discrete value feature values is performed, and if it is determined that the ranges of the upper and lower thresholds overlap with each other, the continuous amount feature value selection unit 7, the continuous quantity feature value having the maximum dispersion ratio is selected. Then, the second calculator 7a calculates the standard deviation σ and the average value xav of the selected continuous amount feature value, and the second threshold calculator 7b calculates the standard deviation σ and the average calculated by the second calculator 7a. THup = xav + kσ and THdn = xav-kσ based on the value xav
To calculate the upper threshold THup and the lower threshold TH
dn is calculated. Then, the continuous amount feature value discriminating unit 8 determines whether or not the ranges of the upper and lower thresholds calculated by the second threshold calculating unit 7b overlap each other, and the upper and lower limits calculated by the second threshold calculating unit 7b. When it is determined that the ranges of the thresholds overlap each other, the threshold changing unit 8a changes the upper and lower thresholds to eliminate the overlap. Thereafter, the image of the work to be identified is captured by the image detection unit 1, and the feature value output unit 9 selects the area value, the circumscribed rectangle length and the discrete value feature value selection unit 5, and the feature value selected by the continuous quantity feature value selection unit 7. In the work discriminating unit 10, the characteristic value output from the characteristic value output unit 9 and the upper threshold value, the lower threshold value calculated in correspondence with the characteristic value, or the threshold value correcting unit 8a. A difference between the changed upper and lower thresholds is determined, and a signal indicating the work to be processed or a signal indicating the disturbance work is output. That is, when all the output feature values are within the range of the corresponding upper threshold and lower threshold, respectively, a signal indicating that the work to be identified is the work to be processed is output, and at least one feature value is equal to or smaller than the above. When it is out of the range, a signal indicating that the identification target work is a disturbance work is output.

Although not particularly shown, when one of the upper and lower threshold ranges based on the continuous amount feature value is completely included in the other range, it is possible to obtain an effective threshold even if the threshold is changed. Since this is not possible, the interval between the upper and lower threshold values may be set narrower and the above series of processing may be performed again.

[0025]

As described above, according to the first aspect of the present invention, it is possible to greatly reduce the frequency of selecting a continuous quantity feature value other than an area value and a value defining a peripheral contour as a work identification feature value. As a result, it is possible to remarkably reduce the processing load for work identification, and to achieve a unique effect that high-precision work identification can be achieved for various types of works to be identified.

According to the second aspect of the present invention, it is possible to greatly reduce the frequency of selecting a continuous quantity feature value other than an area value and a value defining a surrounding contour as a work identification feature value. In this case, the processing load required for the identification can be significantly reduced, and a high-precision workpiece identification can be achieved for various types of workpieces to be identified.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a process of teaching a work to be identified in the work identification method of the present invention.

FIG. 2 is a flowchart showing a main part of a feature value selection process for work identification in the work identification method of the present invention.

FIG. 3 is a flowchart showing a work identification process based on a selected characteristic value in the work identification method of the present invention.

FIG. 4 is a flowchart illustrating a remaining part of a feature value selection process for work identification;

FIG. 5 is a diagram schematically showing a case where upper and lower thresholds corresponding to a feature value can be applied as they are and a case where they cannot be applied as they are.

FIG. 6 is a schematic diagram showing an upper threshold and a lower threshold of a feature value.

FIG. 7 is a block diagram showing one embodiment of the work identification device of the present invention.

FIG. 8 is a diagram schematically showing a configuration of a component supply system to which the work identification method of the present invention is applied.

[Explanation of symbols]

1a Feature value extraction unit 3 First calculation unit 3a First
Threshold calculator 4 Feature value discriminator 5 Discrete value feature value selector 5a
Threshold setting unit 6 Discrete value feature value discriminating unit 7 Continuous amount feature value selecting unit
7a Second calculation unit 9 Feature value output unit 10 Work discrimination unit

──────────────────────────────────────────────────続 き Continuation of the front page (51) Int.Cl. ⁶ Identification number Agency reference number FI Technical display location G01B 11/00 G01B 11/00 Z 11/26 11/26 H

Claims (2)

(57) [Claims] 1. A feature value of a plurality of teaching works is extracted, and an area value and a value defining a surrounding contour are selected from the extracted feature values, and an area value corresponding to a type of the teaching work is selected. In the case where the identification accuracy of the workpiece by the value defining the surrounding contour is high, the area value and the value defining the surrounding contour are selected as the feature value for workpiece identification, and the work value is determined by the value defining the area value and the surrounding contour. When the identification accuracy is low, a feature value with the highest workpiece identification accuracy is selected from the discrete value feature values, and the feature value is obtained based on the average value and variance of the feature values selected from the discrete value feature values. It is determined whether or not the ranges based on the upper and lower thresholds for each work overlap, and in the case where the ranges overlap, the feature value with the highest work identification accuracy is selected from the continuous amount feature values, and the final value is selected. Selected A work to be identified based on the characteristic value and a corresponding upper and lower threshold value. 2. A feature value extracting means (1a) for extracting feature values of a plurality of teaching works;
First selecting means (3) for selecting an area value and a value defining a surrounding contour, and whether or not the work identification accuracy by the value defining the area value and the surrounding contour corresponding to the type of the teaching work is high; (3a) (4)
In the case where the accuracy of identifying the workpiece by the value defining the area value and the surrounding contour is high, the value defining the area value and the surrounding contour is selected as the feature value for workpiece identification, and the area value and the surrounding contour are defined. A second selecting unit (5) for selecting a feature value having the maximum work identification accuracy from the discrete value feature values when the workpiece identification accuracy by the value is low; and a feature selected from the discrete value feature values. The second determination means (5a) and (6) for determining whether or not the ranges based on the upper and lower thresholds for each workpiece obtained based on the average value and the variance of the values overlap each other; Third selecting means (7) for selecting a feature value having the highest work identification accuracy from the continuous quantity feature values, and selecting a target work based on the finally selected feature value and the corresponding upper and lower threshold values. Object to be identified Click identification unit (9) (10) and the workpiece identification device, which comprises a.