JP2009244108A - Method and apparatus for inspecting screw - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of appropriately determining the quality of screws by extracting only defects of thread parts of the screws. <P>SOLUTION: This invention is provided as a screw inspection method. The method is provided with a process for acquiring a luminance distribution of the surface of a screw; a process for specifying the position of a point of inflection of the luminance distribution along the axial direction of the screw; a process for positioning mask data to the luminance distribution according to the position of a specified point of inflection; a process for extracting a luminance distribution of a thread part of the screw on the basis of the luminance distribution of the surface of the screw by the use of positioned mask data; and a process for determining the quality of the screw on the basis of the luminance distribution of the thread part of the screw. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a method and apparatus for inspecting screws.

  Conventionally, when a screw is inspected, a visual inspection is performed by an operator. The visual inspection by the workers has a problem that the workload of the workers is high and individual inspection results vary for each worker. In addition, when inspecting the external thread, when inspecting the internal thread, it was not possible to inspect the quality of the thread appropriately by visual inspection.

  Therefore, techniques for automating screw inspection have been developed. Patent Document 1 discloses a technique for determining whether a screw is good or bad based on image data obtained by imaging the surface of the screw.

JP 59-141008 A

  In the technique of Patent Document 1, the quality of a screw is determined according to the overall size of the defect without distinguishing between a defect at the crest of the screw and a defect at the trough. Therefore, in the technique of Patent Document 1, even when the thread crest is healthy and a defect exists in the screw trough, the screw is determined to be defective. However, even if there are some defects in the valley of the screw, the screw can often be used without any problem if the thread of the screw that receives the load of the screw is healthy. Conversely, even if the thread valley is healthy, the screw cannot be used if there is a defect in the thread peak. Therefore, there is a need for a technique for extracting only the defects at the thread ridges and determining the quality of the screws according to the size of the thread ridge defects.

  The present invention solves the above problems. The present invention provides a technique capable of appropriately determining whether or not a screw is good by extracting only a defect in a screw thread.

  The present invention is embodied as a method for inspecting a screw. The method includes obtaining a luminance distribution on the surface of the screw, identifying a position of the inflection point of the luminance distribution along the axial direction of the screw, and masking according to the position of the identified inflection point. A step of aligning the data with the luminance distribution, a step of extracting the luminance distribution of the screw ridge from the luminance distribution of the surface of the screw using the aligned mask data, and the luminance distribution of the screw ridge A step of determining whether the screw is good or bad is provided.

  In the above method, the brightness distribution of the thread crest is extracted from the brightness distribution of the screw surface using mask data. In general, in the brightness distribution of the screw surface, the brightness of the thread ridge and the brightness of the thread valley are higher than the brightness of the slope of the screw. It is difficult to determine the valley of the screw. Therefore, in the above method, the peak portion and the valley portion of the screw are discriminated by paying attention to the position of the inflection point of the luminance distribution along the axial direction of the screw. This makes it possible to properly align the mask data and accurately extract the luminance distribution of the thread ridges. Based on the extracted luminance distribution of the thread ridge, it is possible to determine whether or not the screw is good by determining whether or not the thread is missing.

  The present invention can also be embodied as a device for inspecting screws. The apparatus includes a means for acquiring a luminance distribution on the surface of the screw, a means for specifying the position of the inflection point of the luminance distribution along the axial direction of the screw, and a mask according to the position of the specified inflection point. Means for aligning the data with the luminance distribution, means for extracting the luminance distribution of the screw ridge from the luminance distribution of the screw surface using the aligned mask data, and based on the luminance distribution of the screw ridge Means for determining whether the screw is good or bad is provided.

  According to the apparatus and the method of the present invention, it is possible to appropriately determine the quality of the screw by extracting only the defect of the screw thread.

The main features of the embodiments described below are listed.
(Feature 1) The method and apparatus of the present invention inspects an internal thread formed on the inner surface of a screw hole.

  A screw inspection system 100 according to an embodiment of the present invention will be described with reference to the drawings. As shown in FIG. 1, the screw inspection system 100 includes a measuring device 102 and an inspection control device 104. The inspection control device 104 is a general general-purpose computer device that includes a control device 110, an input device 112, an output device 114, a storage device 116, and a calculation device 118. The storage device 116 includes a program storage unit 120 and a data storage unit 122. The program storage unit 120 stores a program for causing the screw inspection system 100 to execute an operation described later. The data storage unit 122 stores data required when the screw inspection system 100 operates, data input from the input device 112 and the measuring device 102 during the operation of the screw inspection system 100, and operation of the screw inspection system 100. The generated data is stored in it. The control device 110 sequentially reads out the program stored in the program storage unit 120 and executes various arithmetic processes using the arithmetic device 118, whereby the screw inspection system 100 realizes operations described later. FIG. 1 shows only the components of the measuring device 102 that have a signal input / output relationship with the inspection control device 104.

  FIG. 2 shows the mechanical configuration of the measuring apparatus 102. As shown in FIG. 2, the measuring device 102 includes a probe 202 and an arm 204 that supports the probe 202. The probe 202 is a cylindrical member, and accommodates an LED 210 as a light emitting element and a CCD 212 as a light receiving element at the tip thereof. The arm 204 is driven by the actuator 208 to insert the probe 202 into the screw hole H. When the LED 210 emits light while the probe 202 is inserted into the screw hole H, light is irradiated to the side of the probe 202 and the light reflected by the inner surface of the screw hole H is received by the CCD 212. The operations of the actuator 208, the LED 210, and the CCD 212 are controlled by the inspection control device 104. The CCD 212 detects the luminance of the received reflected light and outputs it to the inspection control device 104.

  The probe 202 is rotatably supported by the arm 204 in the circumferential direction. By driving the motor 206 mounted on the arm 204, the probe 202 rotates in the circumferential direction with respect to the arm 204. The operation of the motor 206 is controlled by the inspection control device 104. By continuously performing the light irradiation by the LED 210 and the luminance detection by the CCD 212 while rotating the probe 202, the luminance distribution in the circumferential direction of the inner surface of the screw hole H can be acquired for one line.

  When the luminance distribution in the circumferential direction of the inner surface of the screw hole H is acquired for one line, the actuator 208 is driven to pull the probe 202 upward by one line, and the LED 210 is irradiated with light while rotating the probe 202 again. Luminance detection by the CCD 212 is performed. By repeating this, the luminance distribution in the circumferential direction is acquired in order from the deepest portion of the screw hole H upward, and the luminance distribution of the entire inner surface of the screw hole H can be acquired.

  The inspection control device 104 can calculate the insertion depth of the tip of the probe 202 based on the control history of the actuator 208. The motor 206 has a built-in encoder and outputs the rotation angle of the probe 202 to the inspection control device 104. The inspection control device 104 stores the luminance detected by the CCD 212 in the data storage unit 122 in association with the insertion depth of the tip of the probe 202 and the rotation angle of the probe 202 at that time. Finally, the data memory 122 stores the luminance distribution of the entire inner surface of the screw hole H as original image data.

  FIG. 3 shows the relationship between the shape of the screw formed on the inner surface of the screw hole H and the brightness of the reflected light at each part. Since the thread crest 302 and the trough 304 are smooth surfaces perpendicular to the incident light from the probe 202, the brightness of the reflected light is high. In the case of a general screw, since the width of the crest 302 is wider than the width of the trough 304, the brightness of the reflected light from the crest 302 is higher than the brightness of the reflected light from the trough 304. Further, since the inclined surface portion 306 of the screw is inclined with respect to the incident light from the probe 202, the brightness of the reflected light is lower than that of the peak portion 302 and the valley portion 304. Although not shown in FIG. 3, when a defect exists in the thread crest 302 or the trough 304, the part often has an inclination or unevenness, and the brightness of the reflected light is a healthy mountain. It becomes lower than the brightness | luminance of the reflected light in the part 302 and the trough part 304. FIG.

  Hereinafter, an inspection process for screws formed on the inner surface of the screw hole H by the screw inspection system 100 will be described in detail with reference to FIG.

  In step S <b> 402, the entire inner surface of the screw hole H is imaged by the measuring device 102, and original image data indicating the luminance distribution of the entire inner surface of the screw hole H is acquired. By the process in step S <b> 402, original image data obtained by imaging the entire inner surface of the screw hole H is stored in the data storage unit 122. FIG. 5 illustrates the original image data 500 acquired in step S402.

  In step S404 in FIG. 4, a defect extraction process is performed based on the original image data stored in the data storage unit 122. Details of this processing will be described later. By the processing in step S404, defect image data that identifies the part where the defect exists and the other part is stored in the data storage unit 122.

  In step S406, a thread pattern extraction process is performed based on the original image data stored in the data storage unit 122. Details of this processing will be described later. Through the processing in step S406, the thread pattern image data obtained by extracting the thread thread pattern is stored in the data storage unit 122.

  In step S408, the quality of the screw hole H is determined based on the defect image data generated in step S404 and the thread pattern image data generated in step S406. Details of this processing will be described later.

  Hereinafter, the defect extraction process performed in step S404 in FIG. 4 will be described with reference to FIG.

  In step S <b> 602, original image data is read from the data storage unit 122.

  In step S604, a moving average process is performed on the luminance distribution represented by the original image data read in step S602. As shown in FIG. 5, in the original image data 500, there is a repetitive pattern in which a screw crest is bright, a subsequent slope is dark, a following trough is bright, and a subsequent slope is dark. . By performing the moving average process in step S604 of FIG. 6, the bright and dark repetitive patterns as described above are leveled. When this moving average process is performed, an average value including a peak or valley having high luminance is calculated for a healthy portion where no defect is present in the peak or valley of the screw, so that the luminance becomes slightly bright. On the other hand, the average value that does not include a peak or valley having a high luminance is calculated for a portion where a defect exists in the peak or valley of the screw, so that the luminance is slightly dark.

  In step S606, binarization processing is performed on the luminance distribution after the moving average processing is performed in step S604. By performing this binarization process, a luminance distribution is obtained in which a healthy part in which no defect exists in a peak or a valley and a defective part in which a defect exists in a peak or a valley are clearly distinguished.

  In step S608, the luminance distribution after the binarization process in step S606 is stored in the data storage unit 122 as defect image data.

  FIG. 7 illustrates defect image data generated by the defect extraction process. By performing the defect extraction process of FIG. 6 on the original image data 500 of FIG. 5, there is a healthy part 704 in which no defect exists in the peaks and valleys and a defect in the peaks and valleys as shown in FIG. Defect image data 700 identifying the defect portion 702 is obtained. In the defect image data 700, the healthy part 704 is expressed as a bright part, and the defective part 702 is expressed as a dark part.

  Hereinafter, the thread pattern extraction process performed in step S406 of FIG. 4 will be described with reference to FIG.

  In step S <b> 802, original image data is read from the data storage unit 122.

  In step S804, the luminance of one line along the axial direction of the screw (vertical direction in the original image data 500 in FIG. 5) is determined from the original image data read in step S802, that is, the luminance distribution of the entire inner surface of the screw hole H. Extract the distribution. FIG. 9A illustrates the luminance distribution extracted in step S804.

  In step S806, the position of the inflection point is specified for the luminance distribution for one line extracted in step S804. The position of the inflection point is specified by calculating a second-order differential value for the luminance distribution for one line shown in FIG. 9A and obtaining a point where the second-order differential value is zero. As shown in (a) of FIG. 9, the position of the inflection point P in the luminance distribution for one line has a luminance peak corresponding to the thread peak and a luminance peak corresponding to the thread valley. Each position is specified to be sandwiched between both sides. The inspection control device 104 according to the present embodiment generates an inflection point waveform shown in FIG. 9B according to the position of the specified inflection point. This inflection point waveform is a binary waveform whose value is switched at the specified inflection point, and the range corresponding to the peak or valley of the screw exists between the inflection points before and after. Takes a value of “1” and takes a value of “0” in a nonexistent range.

  In step S808, the mask data is aligned according to the position of the inflection point specified in step S806. In this embodiment, the reference inflection point waveform shown in FIG. 9C is stored in advance in the data storage unit 122. The reference inflection point waveform is an inflection point waveform which is obtained in advance from a luminance distribution in the case of an ideal screw shape having no manufacturing tolerance and is set in advance from the position of the inflection point in the luminance distribution. The sum of the difference between the inflection point waveform shown in (b) of FIG. 9 and the reference inflection point waveform shown in (c) of FIG. 9 is calculated, the offset amount that minimizes the sum is calculated, and the mask data Perform position alignment.

  FIG. 9D shows the mask data handled in this embodiment. The mask data is binary data used to extract the luminance distribution of only the thread crest from the luminance distribution of the screw. The range in which the mask data takes a value of “1” is set to a range that includes the threaded portion of the screw and does not include the threaded valley portion in the case of an ideal screw shape with no manufacturing tolerance. . In other ranges, the mask data takes a value of “0”. In the present embodiment, the mask data of FIG. 9D is represented based on the offset amount between the inflection point waveform shown in FIG. 9B and the reference inflection point waveform of FIG. Positioning is performed with respect to the luminance distribution 9 (a).

  In step S810, the luminance distribution for one line extracted in step S804 (shown in FIG. 9A) is multiplied by the mask data (shown in FIG. 9D) aligned in step S808. The luminance distribution for one line obtained by extracting the thread ridges shown in FIG.

  In step S812, it is determined whether or not the above extraction process has been completed for all lines along the axial direction of the screw in the original image data read in step S802. If there is a line that has not been subjected to extraction processing (NO in step S812), the processing returns to step S804, and the processing from step S804 to step S810 is executed for the line that has not yet undergone extraction processing. If the extraction process has been completed for all lines (YES in step S812), the process proceeds to step S814.

  In step S814, the extracted screw thread pattern is corrected. At the time of step S814, the thread pattern for the entire inner surface of the screw hole H has already been extracted. In this extracted thread pattern, if there is a missing part in the threaded part of the screw hole H, the thread pattern in which that part is missing is extracted. In step S814, the extracted thread pattern is approximated by a straight line having a uniform width to perform processing for correcting such a missing thread pattern.

  In step S816, the thread pattern corrected in step S814 is stored in the data storage unit 122 as thread pattern image data.

  FIG. 10 illustrates the thread pattern image data 1000 obtained by the thread extraction process described above. By performing the thread pattern extraction process of FIG. 8 on the original image data shown in FIG. 5, the thread pattern image data 1000 in which the thread thread portion 1002 and the other portion 1004 are distinguished as shown in FIG. 10 is obtained. . In the screw thread pattern image data 1000, the width of the screw thread portion in the actual screw hole H and the pitch of the thread thread are accurately reflected. By using this thread pattern image data 1000, a more realistic thread pattern can be set as compared with the case where the thread pattern is set from the nominal dimensions of the screw.

  Hereinafter, the quality determination process for the screw hole H performed in step S408 of FIG. 4 will be described with reference to FIG.

  In step S1102, the defect image data stored in the data storage unit 122 is read.

  In step S1104, the thread pattern image data stored in the data storage unit 122 is read.

  In step S1106, the area of the missing portion of the thread is calculated based on the defect image data read in step S1102 and the thread pattern image data read in step S1104. As shown in FIG. 12, an area 1202 that is identified as a defective part in the defect image data and an area 1206 that overlaps an area 1204 that is identified as a threaded part in the screw thread pattern image data are shown in FIG. It is judged as a missing part. By calculating the area of this overlapping region 1206, the area of the thread missing portion can be acquired.

  In step S1108, the area of the healthy part of the thread is calculated based on the thread pattern image data read in step S1104 and the area of the missing part of the thread calculated in step S1106. The thread represented in the thread pattern image data represents an ideal thread when it is assumed that there is no missing thread in the screw hole H. Therefore, by calculating the total area of the region identified as the screw thread portion in the screw thread pattern image data and subtracting the area of the screw thread missing portion calculated in step S1106, The area of the healthy part can be calculated.

In step S1110, the effective screw length of the screw hole H is calculated based on the area of the missing part of the thread calculated in step S1106 and the area of the healthy part of the thread calculated in step S1108. The effective screw length is calculated by the following formula.
Effective screw length = axial length of the threaded portion of the screw hole H x area of the threaded healthy part / (area of the threaded healthy part + area of the threaded missing part)

  In step S1112, the effective screw length calculated in step S1010 is compared with the necessary screw length required for the screw hole H. The required screw length is defined such that sufficient strength is secured against the fastening torque applied to the screw hole H. If the effective screw length is greater than or equal to the required screw length (YES in step S1112), the screw hole H is determined to be good in step S1114. If the effective screw length is less than the required screw length (NO in step S1112), the screw hole H is determined to be defective in step S1116. The result of the quality determination for the screw hole H is output to the output device 114 in FIG.

  As described above, according to the screw inspection system 100 of the present embodiment, it is possible to appropriately inspect the quality of the screw formed on the inner surface of the screw hole H without performing a visual inspection by an operator.

  In the above-described embodiment, the width of the thread ridge and the pitch of the thread can be measured from the thread pattern image data generated by the thread pattern extraction process of FIG. Not only the quality determination focusing on the missing thread, but also the quality determination focusing on the processing accuracy and required tolerances of the screw thread can be performed.

  In the above-described embodiment, by changing the mask data in the screw thread pattern extraction process of FIG. 8, the screw valley pattern can be extracted instead of extracting the screw thread pattern. By extracting the screw valley pattern, the width of the screw valley can be measured. It is also possible to make a pass / fail judgment focusing on the machining accuracy of the thread valley and the required tolerance.

  In the above embodiment, the case where the internal thread formed on the inner surface of the screw hole H is inspected has been described. However, it is also possible to inspect the thread of the external thread based on the same principle.

  In the above-described embodiment, a region where a region identified as a defective portion in the defect image data and a region identified as a screw thread portion in the screw thread pattern image data overlap is identified as a missing portion of the screw thread. Explained the case. In addition to this, for example, without correcting the thread pattern in step S814 in FIG. 8, the thread missing part and the sound part directly from the thread pattern extracted by the processing from step S802 to step S812 are performed. May be identified.

  In the above embodiment, the case where the luminance distribution on the inner surface of the screw hole H is acquired by the LED 210 and the CCD 212 provided at the tip of the probe 202 has been described. In addition to this, the luminance distribution of the inner surface of the screw hole H may be acquired by another optical system, for example, using a combination of a semiconductor laser and an optical fiber.

Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above.
The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology illustrated in the present specification or the drawings achieves a plurality of objects at the same time, and has technical utility by achieving one of the objects.

1 is a diagram illustrating a configuration of a screw inspection system 100. FIG. 2 is a diagram illustrating a mechanical configuration of a measuring device 102. FIG. It is a figure which shows typically the relationship between the shape of a screw, and reflected luminance. 5 is a flowchart for explaining screw inspection processing by the screw inspection system 100. It is a figure which illustrates the original image data 500 which the screw inspection system 100 handles. 5 is a flowchart for explaining defect extraction processing by the screw inspection system 100. It is a figure which illustrates the defect image data 700 which the screw inspection system 100 produces | generates. 5 is a flowchart for explaining thread pattern extraction processing by the screw inspection system 100. It is a figure explaining signs that screw inspection system 100 extracts a screw thread pattern. It is a figure which illustrates the screw thread pattern image data 1000 which the screw inspection system 100 produces | generates. 5 is a flowchart for explaining quality determination processing of a screw hole H by the screw inspection system 100. It is a figure explaining signs that screw inspection system 100 distinguishes a missing part and a healthy part of a screw thread.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Inspection system 102 Measuring apparatus 104 Inspection control apparatus 110 Control apparatus 112 Input apparatus 114 Output apparatus 116 Storage apparatus 118 Operation apparatus 120 Program storage part 122 Data storage part 202 Probe 204 Arm 206 Motor 208 Actuator 210 LED
212 CCD
302 Screw ridge 304 Screw valley 306 Screw slope 500 Original image data 700 Defect image data 702 Defect portion 704 Healthy portion 1000 Thread pattern image data 1002 Thread ridge 1004 Portion other than screw ridge 1202 Defect Area 1204 identified as a part Area 1206 identified as a screw thread part Area identified as a screw thread missing part

Claims (2)

A method for inspecting screws,
Obtaining a luminance distribution of the surface of the screw;
Identifying the position of the inflection point of the luminance distribution along the axial direction of the screw;
Aligning the mask data with the luminance distribution according to the position of the specified inflection point;
Extracting the brightness distribution of the thread ridge from the brightness distribution of the surface of the screw using the aligned mask data;
A method comprising a step of determining whether a screw is good or bad based on a luminance distribution of a screw thread. A device for inspecting screws,
Means for obtaining the luminance distribution of the surface of the screw;
Means for identifying the position of the inflection point of the luminance distribution along the axial direction of the screw;
Means for aligning the mask data with the luminance distribution according to the position of the specified inflection point;
Means for extracting the brightness distribution of the thread crest from the brightness distribution of the screw surface using the aligned mask data;
An apparatus comprising means for determining whether a screw is good or bad based on a luminance distribution of a screw thread.

Images