JP5080371B2 - Defect detection method for press parts - Google Patents

Description

  The present invention relates to a defect detection method for a pressed part formed by pressing a steel plate such as a car body of an automobile, for example.

In the manufacturing process of automobiles, particularly in the process of manufacturing a car body (body part), a thin steel plate having a thickness of about 0.6 to 3.0 mm is used as a material. ) Is often formed.
The press working as described above is a working method in which a steel plate disposed between a pair of dies is pressed with both dies using a press device having a pair of dies (male type, female type), and the object to be processed It is possible to shorten the time required for processing per part to be as short as about 5 seconds. For this reason, it is suitable as a processing method for mass-producing parts having the same shape such as a vehicle body. Note that the above-described pair of molds are formed so that the surfaces facing each other have a shape that follows the outer shape of a target component (such as a vehicle body).

  However, in the manufacture of pressed parts by press working (hereinafter referred to as “pressed parts”), the press is performed for reasons such as variations that occur in the material properties of the steel sheet that is the material and fluctuations that occur in the pressing conditions of the pressing device. Defects such as cracks may occur in the parts. In recent years, due to the improvement of material quality and the progress of press working technology, defects occurring in pressed parts have been greatly reduced, but it is difficult to completely prevent the occurrence.

  In addition, when press parts are continuously produced at high speed by pressing, heat (hereinafter referred to as “processing heat”) is generated in the pressed parts and the mold used for the pressing. When processing heat is generated, there is an increased possibility that a pressed part will be defective due to deterioration of the lubricant applied to the surface of the mold or material, thermal deformation generated in the mold, or the like. Since the generation factor of the processing heat changes depending on the season, processing conditions, etc., it is difficult to predict whether or not a defect will occur in the pressed part.

By the way, at parts production sites of automobile manufacturers, it is impossible to use this press part as a product even if the defect that occurred in the press part is about several millimeters. Is an important issue. In particular, when a defect occurs in a pressed part continuously due to a material defect (variation in material characteristics, etc.) and the part is used as it is in the assembly of a vehicle body in a subsequent process, the work cost, material cost, etc. The loss, that is, the monetary loss is very large.
For this reason, the press part which the defect by press work tends to generate | occur | produce may perform the process of test | inspecting separately by an inspector's visual inspection with respect to a press part.

  However, press parts formed by press working require a short time of about 5 seconds for processing per piece. For example, all of the press parts produced for about 500 pieces per hour are visually inspected. It is very difficult. In addition, it is difficult to secure the labor of the inspector at the work site where the number of workers is reduced by introducing mechanization greatly. Further, as long as the inspection by visual inspection, that is, the inspection of the defect by the human eye, is performed, there is a possibility that the defect generated in the pressed part is missed.

As a countermeasure against these problems, for example, a method of detecting defects occurring in a pressed part with reference to the following two technologies can be considered.
As described in Patent Document 1, the first technique irradiates thermal energy to a portion to be inspected composed of a joint including a heat conducting member, and infrared rays emitted from the portion to be inspected are transmitted by an infrared camera. This is an inspection method for receiving light and inspecting an electronic component. In this inspection method, a laser beam is used as thermal energy, and the inspection target portion is irradiated with the laser beam vertically. Further, the laser irradiation diameter is set so as to cover the entire inspection target part.

The second technique is a turbine blade film defect inspection apparatus including a shield, a film defect inspection monitoring unit, a controller, and an infrared video signal processing unit, as described in Patent Document 2. is there.
In the technique described in Patent Document 2, the shield surrounds the turbine blade in a state of being implanted in the turbine shaft. In addition, the coating defect inspection monitoring unit can be placed on a rotating table that is rotatably supported by a frame with respect to the turbine blade, and can move forward and backward on each of the X, Y, and Z axes. And can be tilted. Further, the controller controls the forward / backward movement and tilt movement of the film defect inspection / monitoring unit, and the infrared video signal processing unit images the temperature distribution of the film based on the signal from the film defect inspection / monitoring unit.

However, in the techniques described in Patent Document 1 and Patent Document 2 described above, heat from an external heating device is applied to a component to be inspected (a joint portion including a heat conducting member in Patent Document 1 and a turbine blade in Patent Document 2). It is necessary to give energy. For this reason, it is necessary to further install a heating device on the part production line in which the press device is already installed, and there is a problem that the space efficiency is lowered.
In addition, in the techniques described in Patent Document 1 and Patent Document 2, a heating process using a heating device is added to the inspection process for the component to be inspected, so that even if the defect detection accuracy can be improved, There is a problem that it is difficult to improve the productivity of parts.

On the other hand, as a defect detection method for a continuously processed product, for example, there is a defect detection method for a sheet-like member described in Patent Document 3.
The defect detection method for a sheet-like member described in Patent Document 3 irradiates a sheet-like component that is continuously fed with light, detects an optical image in which a change in brightness is caused by the light irradiation, and detects the image. The type of defect is specified by analyzing the above.

However, in the technique described in Patent Document 3, it is an essential condition that the part to be inspected is flat over the entire surface, and there is a problem that it is difficult to apply to an automobile press part having a complicated shape.
In addition to the techniques described in Patent Document 1 to Patent Document 3 described above, as a method for detecting defects in a resin-made flexible tube filled with a gel-like substance such as toothpaste, cosmetics, paint, food, etc., for example, a patent There is a method described in Document 4.

The method described in Patent Document 4 is a method of detecting a defect by transmitting infrared light through a translucent inspection object and analyzing the transmitted image.
Further, as a method for inspecting defects such as scratches on a semiconductor thin plate or a substrate, foreign matter contamination, cracks, etc., there is a technique described in Patent Document 5, for example.
The technique described in Patent Document 5 irradiates the inspection object with the transmission illumination means, captures the illumination light transmitted through the inspection object as a transmission image, and analyzes the image to determine whether the inspection object is good or bad. This is a method of making a determination.

However, in the techniques described in Patent Document 4 and Patent Document 5 described above, the object to be measured is a semi-transparent material that can transmit infrared light, or a device that can transmit light such as a semiconductor substrate. Yes. For this reason, it is difficult to apply to a pressed part made of a steel plate that cannot transmit light.
In order to solve these problems, for example, a defect detection method for a pressed part described in Patent Document 6 has been proposed.

  In the defect detection method for a pressed part described in Patent Document 6, first, a temperature distribution on the surface of the pressed part is imaged with an infrared camera in the production line of the part equipped with the pressing device. Then, the temperature distribution on the surface of the part changed by the processing heat is analyzed, and the analyzed temperature distribution is compared with a preset reference temperature distribution to detect defects generated in the pressed part.

  Specifically, a reference table in which the non-defective product temperature distribution is histogrammed is stored in advance in the memory. Then, the thermal image captured by the infrared camera is processed, and the temperature distribution information of the pressed part is analyzed as a histogram, and the analyzed temperature distribution information is compared with a reference table to determine the defect generated in the pressed part. To do. In the temperature distribution analysis, the temperature distributions expressed on the surface of the pressed part may be directly compared without using the above-described histogram formation.

With such a defect detection method for a pressed part, it is not necessary to install a heating device on the production line for the pressed part, so that it is possible to suppress a decrease in space efficiency. Moreover, since it is not necessary to add a heating process to the inspection process for the press part to be inspected, it is possible to suppress a reduction in productivity of the press part.
Japanese Patent Laid-Open No. 5-52785 JP 2003-98134 A JP-A-9-153299 JP 2006-64389 A JP 2007-309679 A JP 2006-177892 A JP 2007-309679 A

In the defect detection method for a pressed part described in Patent Document 6, when the state of pressing is stable and the processing heat is stable, the pressed part is based on the difference between the analyzed temperature distribution and the reference temperature distribution. This makes it easy to detect defects that have occurred.
However, the press part defect detection method described in Patent Document 6 is difficult to put into practical use due to the following problems.

  In general press work sites, mold replacement work and trial placement work are often performed before press work. After these work is performed, continuous production of press parts is performed. There are many cases. In addition, even during continuous production, continuous production may be interrupted unsteadily due to a material supply operation for supplying the material for the pressed parts, a loading operation for loading the pressed parts on the conveying means, or the like.

In such a case, the processing heat generated by pressing is in a state close to the ambient temperature immediately after the start of continuous production, and increases every time pressing is repeated. Then, when such a certain amount of processing heat is applied, a rise in the processing heat almost stops and stabilizes.
At this time, when continuous production is interrupted due to material supply work, loading work, or the like, the stable processing heat rapidly decreases. When continuous production is resumed, the reduced processing heat will rise again. Therefore, there is a possibility that a change in processing heat may occur in a practical press working site.

In addition, since the processing heat is greatly affected by the ambient temperature, especially when there is a large difference in the ambient temperature, such as in summer and winter, there is a large difference in the temperature distribution of the processing heat on the part surface. There is a fear.
Therefore, at the work site where the press work is actually performed, if a defect is detected based on the difference between the analyzed temperature distribution and the reference temperature distribution as in the defect detection method for a pressed part described in Patent Document 6, There is a risk that problems such as oversight or false detection occur. In addition, said false detection is determining with the defect, although the defect has not generate | occur | produced.

In addition, as a method for automatically detecting a characteristic part such as a defect from image data, a method called second derivative processing is applied to temperature or image data, and the temperature or image brightness changes. A method of extracting only a part and detecting a defect from the data is generally known.
According to this method, there is an advantage that the influence of the entire temperature and image brightness can be eliminated. As an example, the technique described in Patent Document 7 proposes a method for automatically and accurately determining a semiconductor thin plate or substrate scratch, contamination, cracks, or the like using this second-order differential processing.

  However, in an actual press production line, there are various disturbances, and there is a problem that erroneous detection occurs only by extracting a temperature change portion. For example, in a press line, a steel plate coated with rust preventive oil or a steel plate that has undergone a process of washing with washing oil before pressing is pressed. At this time, the oil may adhere unevenly in the form of polka dots, and this oil film has the effect of increasing the reflectivity, so that when this image is taken with an infrared camera, the temperature is detected only at this site.

If this image is subjected to second-order differential processing by the above method, it is determined as a defect and erroneous detection occurs. In addition, due to changes in illumination and sunlight, a high temperature is detected only in a certain part, which may cause false detection.
In addition, regarding the direct comparison of temperature distributions represented on the surface of the component and the histogram formation performed by the defect detection method for a pressed component described in Patent Document 6, it is difficult to put it to practical use due to the problems described below. is there.

At the work site where the press work is actually performed, the press part is taken out from the mold by a robot or the like and placed on a conveying means such as an inspection jig or a conveyor. This placement work is a work during high-speed production. Therefore, it is difficult to place the pressed part at a desired position.
For this reason, the position of the placed press part may be displaced by about several tens of millimeters with respect to a desired position. On the other hand, since the infrared camera that captures a thermal image is fixed at a predetermined position, the image data captured by the infrared camera may be displaced from a desired position.
Therefore, when the press part after the placement work is displaced with respect to a desired position, an error occurs in the comparison between the image data captured by the infrared camera and the reference image, and a defect is overlooked or erroneously detected. Problems may arise.

Patent Document 4 compares the appearance image of an inspection product obtained by transmitting infrared rays through a semi-transparent inspection object and the characteristic portion of a reference appearance image recorded in advance in relation to the positional deviation of the inspection object. Thus, a method for calculating the amount of coordinate movement and correcting the positional deviation has been proposed. However, the feature part detection method described in Patent Document 4 is calculated based on the density of a transmission image and cannot be applied to a pressed part that cannot obtain a transmission image.
The present invention has been made paying attention to the above-described problems, and it is an object of the present invention to provide a defect detection method for a pressed part that can suppress oversight of a defect occurring in the pressed part and false detection. And

In order to solve the above-mentioned problem, the invention described in claim 1 of the present invention is a press part defect detection method for detecting defects generated in a press part after press working,
A surface temperature distribution detecting step for detecting a surface temperature distribution of the pressed part after press working, and a defect detecting step for detecting a defect generated in the pressed part based on the surface temperature distribution detected in the surface temperature distribution detecting step. And having
The defect detection step includes a temperature sudden change region detection step of detecting a temperature sudden change region in which the temperature has changed from a peripheral region by a predetermined temperature difference or more, out of the surface temperature distribution detected in the surface temperature distribution detection step, and the temperature A defect determination step for determining whether or not a defect has occurred in the pressed part, based on the temperature sudden change region detected in the sudden change region detection step,
In the temperature sudden change region detection step, the temperature sudden change region is detected by a spatial second derivative process for the surface temperature distribution detected in the surface temperature distribution detection step,
In the defect determination step, a temperature difference between the surface temperature of the press part in the temperature sudden change region detected in the temperature sudden change region detection step and the surface temperature of the press part around the temperature sudden change region detected in the temperature sudden change region detection step is calculated. If the detected temperature difference is equal to or less than the defect internal / external temperature difference threshold value set in advance for each pressed part having at least one of a different shape and material, a defect occurs in the pressed part. It is characterized by determining that it is.

  According to the present invention, in the temperature sudden change region detection step, among the surface temperature distribution of the pressed part after the press working, the temperature sudden change region, which is a region where the temperature changes from the surrounding region by a predetermined temperature difference or more, is obtained. Detection is performed by spatial second-order differential processing for the surface temperature distribution detected in the detection step. Further, in the defect determination step, a defect occurring in the pressed part is detected based on the temperature sudden change region detected in the temperature sudden change region detection step.

Specifically, a temperature difference between the surface temperature of the press part in the temperature sudden change region and the surface temperature of the press part around the temperature sudden change region is detected, and the detected temperature difference is a predetermined defect inside / outside temperature difference threshold value. It is determined whether or not: And when the detected temperature difference is below a defect part inside / outside temperature difference threshold value, it determines with the defect having generate | occur | produced in press parts. The above-mentioned “predetermined defect portion inside / outside temperature difference threshold” is set to a unique value for each pressed part having a different shape or material, for example.
For this reason, it becomes possible to detect the defect which generate | occur | produces in a press part, without receiving the influence by the change of the process heat which generate | occur | produces in the press part and the metal mold | die used for press work, and the temperature change of a press part periphery.

Next, the invention described in claim 2 is the invention described in claim 1, wherein the press part is predicted to have a defect in the surface temperature distribution detected in the surface temperature distribution detection step. A defect monitoring area storing step for storing the defect monitoring area in advance;
The defect monitoring area storage process is a pre-process of the defect determination process,
In the defect determination step, when the temperature sudden change region detected in the temperature sudden change region detection step is present in the defect monitor region based on the defect monitor region stored in the defect monitor region storage step, the press part It is characterized by determining that a defect has occurred.

According to the present invention, a defect monitoring area storing step for storing in advance a defect monitoring area, which is an area predicted to have a defect in a pressed part, is performed as a preceding process of the defect determination process. Further, in the defect determination step, a defect that occurs in the pressed part is detected based on a defect monitoring area that is an area predicted to have a defect in the pressed part in the surface temperature distribution of the pressed part.
For this reason, when there is a sudden temperature change region outside the defect monitoring region, that is, at a position where no defect occurs, such as a gap in the pressed part, the detected sudden temperature change region is determined from the determination target of the defect occurring in the pressed part. It can be excluded.

Next, the invention described in claim 3 is the invention described in claim 2, wherein, in the surface temperature distribution detection step, the surface of the pressed part placed at a predetermined surface temperature distribution detection position after pressing. Detect the placement surface temperature distribution, which is the temperature distribution,
In the defect detection step, based on a reference surface temperature distribution that is a surface temperature distribution of a pressed part when ideally arranged at the surface temperature distribution detection position, the displacement of the placement surface temperature distribution relative to the reference surface temperature distribution is performed. Based on the temperature distribution displacement amount detection step for detecting the amount and the displacement amount detected in the temperature distribution displacement amount detection step, the defect monitoring region of the placement surface temperature distribution is approximate to the defect monitoring region of the reference surface temperature distribution And a defect monitoring region position correcting step for correcting the position of the surface temperature distribution described above,
In the defect determination step, the temperature sudden change region detected in the temperature sudden change region detection step is present in the defect monitor region based on the defect monitor region of the surface temperature distribution described above corrected in the defect monitor region position correction step. In this case, it is determined that a defect has occurred in the pressed part.

According to the present invention, when the position of the press part is displaced from the surface temperature distribution detection position in the defect monitoring area position correction step, the position of the defect monitoring area of the mounting surface temperature distribution is determined based on the amount of displacement. The position of the placement surface temperature distribution is corrected so as to approximate the position of the defect monitoring region of the reference surface temperature distribution.
For this reason, even if the position of the mounting surface temperature distribution is displaced from the reference surface temperature distribution, the position of the defect monitoring area of the mounting surface temperature distribution is determined based on the amount of displacement. It is possible to correct the position.

Next, the invention described in claim 4 is the invention described in claim 2 or 3, wherein the defect monitoring area is divided into a plurality of defect monitoring areas,
Dividing the surface of the pressed part into a plurality of surface temperature distribution detection regions corresponding to the plurality of defect monitoring regions,
In the surface temperature distribution detection step, the surface temperature distribution of the pressed part is detected for each of the plurality of surface temperature distribution detection regions,
In the temperature sudden change region detection step, the temperature sudden change region is detected in the plurality of surface temperature distributions by a spatial second-order differential process for the surface temperature distribution detected for each of the plurality of surface temperature distribution detection regions in the surface temperature distribution detection step. Detect for each area,
In the defect determination step, based on the temperature sudden change region detected for each of the plurality of surface temperature distribution detection regions in the temperature sudden change region detection step, it is determined whether or not a defect has occurred in the pressed part. The determination is made for each distribution detection region.

According to the present invention, the defect monitoring area is divided into a plurality of defect monitoring areas, and the surface of the pressed part is divided into a plurality of surface temperature distribution detection areas corresponding to the plurality of defect monitoring areas. Further, in the temperature sudden change region detection step, the temperature sudden change region is detected for each of the plurality of surface temperature distribution detection regions by the spatial second-order differential processing for the surface temperature distribution detected for each of the plurality of surface temperature distribution detection regions. Moreover, in a defect determination process, it is determined for every some surface temperature distribution detection area | region whether the defect generate | occur | produced in press parts.
For this reason, it is possible to divide an area predicted to have a defect in a pressed part into a plurality of ranges according to the frequency of occurrence of the defect, and detect a defect occurring in the pressed part for each of the divided areas. It becomes possible.

  According to the present invention, it is possible to detect a defect occurring in a pressed part without being affected by a change in processing heat or a change in ambient temperature. And false detection can be suppressed.

Next, an embodiment of the present invention (hereinafter referred to as “the present embodiment”) will be described with reference to the drawings.
First, with reference to FIG. 1, the structure of the equipment which applies the defect detection method (henceforth "defect detection method") of the press part of this embodiment is demonstrated.
FIG. 1 is a diagram showing a production line of a pressed part P to which the defect detection method is applied. In the present embodiment, a case where the press part P is a car body (body part) of an automobile will be described.

As shown in FIG. 1, the production line of the pressed part P to which the defect detection method is applied includes a surface temperature distribution detection unit 1, a component information storage unit 2, a defect monitoring area storage unit 4, and a defect detection unit 6. It has.
The surface temperature distribution detection means 1 includes one or a plurality of infrared cameras 8 capable of taking a thermal image of the press part P, detects the surface temperature distribution of the press part P, and detects the defect monitoring area storage means 4. Output to. In the present embodiment, as an example, a case will be described in which the surface temperature distribution detecting means 1 includes two infrared cameras 8 as shown in the drawing.

Similar to the defect monitoring area storage unit 4, the component information storage unit 2 is formed by using a storage unit such as a memory provided in a computer, for example.
Further, the part information storage means 2 includes a shape of the press part P ideally arranged at the surface temperature distribution detection position, a threshold value used for a binarization process described later, a harmful defect size, and a predetermined defect portion inside / outside temperature difference. The threshold value is stored.

The “surface temperature distribution detection position” is an ideal position where the press part P is arranged when detecting the surface temperature distribution. The setting of the surface temperature distribution detection position is performed according to, for example, the shape of the press part P and the performance (imaging range, etc.) of the infrared camera 8.
The “hazardous defect size” is a size inherent to the pressed part P, and is set in advance according to the purpose of use of the pressed part P, for example.

Further, the “predetermined defect portion internal / external temperature difference threshold” is set to a unique value for each pressed part P having a different shape or material, for example.
The defect monitoring area storage unit 4 is formed using a storage unit such as a memory provided in a computer, for example, and it is predicted that a defect will occur in the pressed part P among the surface temperature distribution detected by the surface temperature distribution detection unit 1. The defect monitoring area that is the area that has been processed is stored in advance.

  Here, the defect monitoring area is an area peculiar to the press part P having the same shape. For example, when the shape of the press part P is a shape having a gap part, the gap part is used as the press part P. It is set as a position where no defect occurs. In addition, for example, when the pressed part P has a shape that has a part that is more likely to cause defects than a flat part, such as a draw-formed part, this part is set as a position at which a defect is likely to occur in the pressed part P. May be.

The defect monitoring area is divided into a plurality of defect monitoring areas. When setting a plurality of defect monitoring areas, for example, it is set according to the shape of the entire press part P.
The defect detection means 6 is formed using, for example, a computer having a storage means such as a memory, and includes a temperature distribution displacement amount detection means 10, a defect monitoring area position correction means 12, a temperature sudden change area detection means 14, Defect determining means 16 is provided.

The temperature distribution displacement amount detection means 10 includes a reference surface temperature distribution storage unit 18 and a displacement amount detection unit 20.
The reference surface temperature distribution storage unit 18 stores in advance a reference surface temperature distribution that is a surface temperature distribution of the pressed part P ideally arranged at the surface temperature distribution detection position.
Specifically, the shape of the press part P ideally arranged at the surface temperature distribution detection position from the part information storage unit 2 and the surface temperature distribution detected by the surface temperature distribution detection unit 1 for the press part P Is acquired in advance. Then, the above-described reference surface temperature distribution is calculated and stored from the acquired shape of the pressed part P and the surface temperature distribution.

Here, the reference surface temperature distribution is divided into a plurality according to a plurality of surface temperature distribution detection regions. That is, the reference surface temperature distribution storage unit 18 stores a reference surface temperature distribution divided into a plurality of parts. In the following description, only one reference surface temperature distribution among the reference surface temperature distributions divided into a plurality of parts will be described, but the other reference surface temperature distributions have the same configuration.
The displacement amount detection unit 20 detects the displacement amount of the placement surface temperature distribution, which is the surface temperature distribution of the pressed part P placed toward the surface temperature distribution detection position with respect to the above-described reference surface temperature distribution.

Here, similarly to the reference surface temperature distribution, the placement surface temperature distribution is divided into a plurality of parts according to a plurality of surface temperature distribution detection regions. In addition, although the following description demonstrates one mounting surface temperature distribution among the mounting surface temperature distribution divided | segmented into plurality, it is the same also about other mounting surface temperature distribution.
Hereinafter, the process which the displacement amount detection part 20 performs is demonstrated concretely.
First, based on the shape of the press part P acquired from the part information storage means 2 and the placement surface temperature distribution, the air gap that the press part P has, that is, the air where the air is located in the placement surface temperature distribution. Detect the temperature detection area.

After detecting the air temperature detection region, based on the placement surface temperature distribution and the air temperature detection region, the temperature in the air temperature detection region of the placement surface temperature distribution, that is, the temperature of the air in the placement surface temperature distribution (hereinafter, “ Air temperature Tair ”) is analyzed.
After analyzing the air temperature Tair, binarization processing is performed on the placement surface temperature distribution using the air temperature Tair as a threshold value. In this binarization process, when the temperature at each pixel of the placement surface temperature distribution is the temperature T (x, y), the pixel of T (x, y) ≧ Tair is converted to “1”. This is a process of converting a pixel of T (x, y) <Tair to “0”. The above (x, y) is each pixel position in the placement surface temperature distribution.

In addition, the binarization process is also performed on the reference surface temperature distribution by the same procedure as the binarization process on the placement surface temperature distribution described above. Since the binarization processing procedure for the reference surface temperature distribution is the same as the binarization processing procedure for the mounting surface temperature distribution described above, the description thereof is omitted.
Then, a selected portion is extracted from the entire reference surface temperature distribution subjected to the binarization process, and a reference shape image is formed using the extracted temperature distribution. This reference shape image includes most of the air temperature detection region or characteristic portions such as corners and curved portions.

Next, the reference shape image is scanned in the X direction and the Y direction with respect to the entire placement surface temperature distribution subjected to the binarization process. Thereby, the area | region which corresponds with a reference | standard shape image among the whole mounting surface temperature distribution is searched.
In searching for a region that matches the reference shape image with respect to the entire placement surface temperature distribution, the coordinates (x, y) of the reference shape image with respect to the placement surface temperature distribution are sequentially detected and accumulated.

  Then, based on the accumulated coordinates, an area that most closely matches the reference shape image is calculated in the entire placement surface temperature distribution, and an area that most closely matches the reference shape image is detected in the entire placement surface temperature distribution. . Further, a displacement amount with respect to the region of the reference shape image in the reference surface temperature distribution of this region is detected, and based on this displacement amount, a displacement amount of the mounting surface temperature distribution with respect to the reference surface temperature distribution is detected.

Here, the placement surface temperature distribution is divided according to a plurality of reference surface temperature distributions. That is, the displacement amount detection unit 20 detects the displacement amounts of the plurality of placement surface temperature distributions with respect to the plurality of reference surface temperature distributions.
As described above, the temperature distribution displacement amount detection means 10 detects the displacement amount of the placement surface temperature distribution with respect to the reference surface temperature distribution.

The defect monitoring area position correcting means 12 is mounted so that the defect monitoring area of the mounting surface temperature distribution approximates the defect monitoring area of the reference surface temperature distribution based on the displacement detected by the temperature distribution displacement detecting means 10. Correct the position of the surface temperature distribution.
Specifically, the defect monitoring area of the placement surface temperature distribution is corrected by displacing the position of the placement surface temperature distribution according to the displacement amount detected by the temperature distribution displacement amount detection means 10, and the reference surface Approximate defect monitoring area of temperature distribution.

Here, the temperature distribution displacement amount detection means 10 detects displacement amounts of a plurality of placement surface temperature distributions. That is, the defect monitoring area position correcting unit 12 corrects the defect monitoring areas of a plurality of placement surface temperature distributions.
The sudden temperature change region detection means 14 includes a spatial second-order differential processing unit 22 and a determination region extraction unit 24.
The spatial secondary differential processing unit 22 performs spatial secondary differential processing represented by the following expression (1) on each of the plurality of placement surface temperature distributions whose defect monitoring area is corrected by the defect monitoring area position correcting unit 12. The mounting surface temperature distribution is converted into a temperature sudden change region distribution that is a distribution of the temperature sudden change region.
W (x, y) = d ² T (x, y) / dx ² + d ² T (x, y) / dy ² (1)

Here, the temperature sudden change region is a region in the placement surface temperature distribution in which the temperature changes with a predetermined temperature difference or more from the surrounding region. The predetermined temperature is set according to the shape of the press part P, for example.
The determination region extraction unit 24 performs binarization similar to the binarization processing on the placement surface temperature distribution for each of the plurality of temperature sudden change region distributions converted from the placement surface temperature distribution by the spatial secondary differentiation processing unit 22. Process. The threshold value used for this binarization processing is set for each defect monitoring area and acquired from the component information storage means 2. The threshold used for the binarization process of the temperature sudden change region distribution is set based on the surface temperature of the press part P, for example.

Then, a temperature sudden change region is detected from each of the plurality of temperature sudden change region distributions subjected to the binarization process. Further, from the plurality of temperature sudden change region distributions subjected to the binarization process, a portion including the temperature sudden change region and its peripheral portion is extracted as a determination region.
As described above, the temperature sudden change region detection unit 14 converts the temperature sudden change region into a plurality of surface temperatures by the spatial second-order differential processing on the mounting surface temperature distribution detected by the surface temperature distribution detection unit 1 for each of the plurality of surface temperature distribution detection regions. Detect for each distribution detection area.

The defect determination means 16 includes a temperature sudden change region position determination unit 26, a temperature sudden change region size determination unit 28, and a temperature sudden change region inside / outside temperature difference determination unit 30.
The temperature sudden change region position determination unit 26 compares the positions of the determination regions extracted from the plurality of temperature sudden change region distributions by the determination region extraction unit 24 with the positions of the corresponding defect monitoring regions, respectively. Then, for example, as shown in FIG. 2, for each of a plurality of defect monitoring areas, it is determined whether or not the corresponding temperature sudden change area exists in the defect monitoring area, and the temperature sudden change area exists in the defect monitoring area. When it is determined, the determination result is output to the temperature sudden change region size determination unit 28. Note that FIG. 2 is a diagram illustrating a state in which, in one defect monitoring area, it is determined whether or not a sudden temperature change area exists in the defect monitoring area. Further, in FIG. 2, the temperature sudden change region is shown with a symbol E.

  For example, as shown in FIG. 3, the temperature sudden change region size determination unit 28 compares the size of the temperature sudden change region with the harmful defect size acquired from the component information storage unit 2, and the temperature sudden change region is larger than the harmful defect size. If it is, the determination result is output to the temperature difference determining part 30 inside and outside the temperature sudden change region. FIG. 3 is a diagram showing a state in which the size of the temperature sudden change region is compared with the size of the harmful defect. Further, in FIG. 3, as in FIG. 2, the temperature sudden change region is indicated by a symbol E.

  The sudden temperature change region inside / outside temperature difference determination unit 30 detects a temperature difference between the surface temperature of the pressed part P in the sudden temperature change region and the surface temperature of the pressed part P around the sudden temperature change region. Then, for example, as shown in FIG. 4, it is determined whether or not the detected temperature difference is equal to or less than a predetermined defect portion inside / outside temperature difference threshold acquired from the component information storage unit 2. And when the detected temperature difference is below a defect part inside / outside temperature difference threshold value, it determines with the defect having generate | occur | produced in press parts. FIG. 4 is a diagram showing a state in which the detected temperature difference is compared with the defect internal / external temperature difference threshold. Further, in FIG. 4, similarly to FIGS. 2 and 3, the temperature sudden change region is indicated with a symbol E.

As described above, the defect determination unit 16 determines whether or not a defect has occurred in the pressed part P based on the temperature sudden change region detected by the temperature sudden change region detection unit 14 for each of the plurality of surface temperature distribution detection regions. The determination is made for each temperature distribution detection region.
Therefore, the defect detection means 6 described above detects a defect generated in the pressed part P based on the surface temperature distribution detected by the surface temperature distribution detection means 1.

(Defect detection method)
Next, a defect detection method performed by the above-described facility will be described with reference to FIGS. 1 to 4 and FIGS.
The defect detection method includes a surface temperature distribution detection step performed by the above-described surface temperature distribution detection unit 1 and a defect detection step performed by the above-described defect detection unit 6.

The surface temperature distribution detection step is a step of detecting the surface temperature distribution of the pressed part P for each of a plurality of surface temperature distribution detection regions.
The defect detection step is a step of detecting defects generated in the pressed part P based on the surface temperature distribution detected for each of the plurality of surface temperature distribution detection regions in the surface temperature distribution detection step.
The defect detection step includes a temperature distribution displacement amount detection step, a defect monitoring region position correction step, a temperature sudden change region detection step, and a defect determination step.

The temperature distribution displacement amount detection step is a step of detecting the displacement amount of the mounting surface temperature distribution described above with respect to the reference surface temperature distribution described above.
The defect monitoring area position correction step is performed so that the defect monitoring area of the mounting surface temperature distribution approximates the defect monitoring area of the reference surface temperature distribution based on the displacement detected in the temperature distribution displacement amount detecting process. This is a step of correcting the position of the temperature distribution.

The temperature sudden change region detection step is a step of detecting the temperature sudden change region for each of the plurality of surface temperature distribution detection regions by spatial second-order differential processing for the placement surface temperature distribution detected in the surface temperature distribution detection step.
The defect determination step determines whether a defect has occurred in the pressed part P based on the temperature sudden change region detected for each of the plurality of surface temperature distribution detection regions in the temperature sudden change region detection step. It is the process of determining to.

Here, the defect detection method has a defect monitoring area storage process as a pre-process of the surface defect determination process.
The defect monitoring region storage step is a step of storing in advance a defect monitoring region which is a region predicted to have a defect in the pressed part P among the surface temperature distribution detected in the surface temperature distribution detection step.

Hereinafter, the surface temperature distribution detection process, the temperature distribution displacement amount detection process, and the defect monitoring region position correction process will be described with reference to FIG. 1 and FIG. 5 and FIG.
FIG. 5 is a flowchart showing the contents of the pre-process of the surface temperature distribution detection process, the surface temperature distribution detection process, the temperature distribution displacement amount detection process, and the defect monitoring region position correction process. FIG. 6 is a diagram illustrating an example of processing performed by the surface temperature distribution detection unit 1 and the defect detection unit 6 in the surface temperature distribution detection step, the temperature distribution displacement amount detection step, and the defect monitoring region position correction step.

As shown in FIG. 5, when starting the defect detection method (step S100), first, the pre-process of the surface temperature distribution detection process is performed.
In the pre-process of the surface temperature distribution detection process, the part information of the pressed part P is acquired from the part information storage means 2, and the process proceeds to the surface temperature distribution detection process (step S101). This part information includes the shape of the press part P ideally arranged at the surface temperature distribution detection position and the surface temperature distribution detected by the surface temperature distribution detection means 1 for the press part P.

Next, the surface temperature distribution detection process will be described.
In the surface temperature distribution detection step, a thermal image of the pressed part P is captured by each infrared camera 8, and the captured thermal image is acquired (step S102). Then, the acquired thermal image is converted into a placement surface temperature distribution for each of the surface temperature distribution detection areas divided into a plurality of parts, and then the process proceeds to a temperature distribution displacement amount detection step.

Next, the temperature distribution displacement amount detection step will be described.
In the temperature distribution displacement amount detection step, first, in the surface temperature distribution detection step, a plurality of placement surface temperature distributions detected for each of the divided surface temperature distribution detection regions are acquired. As shown in FIG. 6A, the acquired placement surface temperature distribution represents an area where the surface temperature is different with a visible image. FIG. 6A shows only one placement surface temperature distribution among a plurality of placement surface temperature distributions.

As shown in FIG. 6A, the placement surface temperature distribution of the press part P has a region that rises to about 50 ° C. This region is, for example, a region showing a portion where stress acting at the time of press working is high or a portion where sliding with the mold is intense.
Further, as shown in FIG. 6A, the placement surface temperature distribution of the pressed part P has a low temperature (about 29 ° C.) region as compared with other regions. This region is a void portion of the press part P, and is a region having a temperature substantially equal to the temperature of the atmosphere.

  After obtaining the plurality of placement surface temperature distributions, the above-described placement surface temperature distributions and the component information of the press part P obtained in step S101 are respectively described above for the plurality of placement surface temperature distributions. The detected air temperature detection area is detected. And based on this air temperature detection area | region and mounting surface temperature distribution, the air temperature Tair mentioned above is analyzed (step S103). In addition, as shown in FIG. 6A, the air temperature detection region is a lower temperature region than the other regions in the placement surface temperature distribution.

After the air temperature Tair is analyzed in step S103, the binarization process described above is performed on each of the plurality of placement surface temperature distributions using the analyzed air temperature Tair as a threshold value (step S104).
As shown in FIG. 6B, the placement surface temperature distribution subjected to the binarization process in step S104 has only two types of regions, that is, the region where the press part P exists and the air temperature detection region. Distribution. FIG. 6B shows only one placement surface temperature distribution among the plurality of placement surface temperature distributions subjected to the binarization process.

After performing binarization processing on the plurality of placement surface temperature distributions in step S104, binarization processing is performed on the plurality of reference surface temperature distributions in the same procedure as the binarization processing on the placement surface temperature distributions. Do. Then, a part including a characteristic part is extracted from each of the plurality of reference surface temperature distributions subjected to the binarization process, and a plurality of reference shape images are formed using the extracted temperature distributions ( Step S105).
As shown in FIG. 6C, the reference shape image formed in step S105 includes an area where the pressed part P exists and an air temperature detection area in the reference surface temperature distribution subjected to the binarization process. FIG. 6C shows only one reference shape image among the plurality of reference shape images.

  After forming a plurality of reference shape images in step S105, one entire mounting surface temperature distribution (in FIG. 5 and FIG. 6, “binarization” is performed among the plurality of reference surface temperature distributions subjected to binarization processing. The corresponding reference shape image is scanned in the X direction and the Y direction. Thereby, an area that matches (matches) the reference shape image is searched for in the entire placement surface temperature distribution (step S106). In this embodiment, the entire reference surface temperature distribution is scanned with the corresponding reference shape image in the X direction and the Y direction, but the processing in step S106 is not limited to this. That is, the entire placement surface temperature distribution may be scanned in the X direction and the Y direction while the reference shape image is rotated at an arbitrary angle.

  In step S106, when searching for a region that matches the reference shape image in the entire placement surface temperature distribution, as shown in FIG. 6D, the placement surface temperature distribution subjected to binarization processing is used. The corresponding reference shape image is scanned in the X direction and the Y direction with respect to the whole. FIG. 6D shows only one placement surface temperature distribution among a plurality of placement surface temperature distributions, and shows a reference shape image corresponding to the placement surface temperature distribution.

In step S106, when searching for an area matching the reference shape image for the entire placement surface temperature distribution, it is determined whether there is an area matching the reference shape image for the entire placement surface temperature distribution. Determination is made (step S107).
In step S107, if an area matching the reference shape image is not detected in the entire placement surface temperature distribution, that is, if it is determined that there is no area matching the reference shape image in the entire placement surface temperature distribution, an alarm is issued. Activating means (not shown). As a result, the alarm lamp is turned on, the alarm sound is output from the alarm speaker, the determination result is displayed on the display, and the interruption instruction signal for interrupting the operation is output to the press device or the like. alarm").

  On the other hand, when it is determined in step S107 that there is a region that matches the reference shape image in the entire placement surface temperature distribution, the displacement amount of the reference shape image in the reference surface temperature distribution of this region is detected (“Heat Image displacement analysis ”). And based on this displacement amount, the displacement amount of the mounting surface temperature distribution with respect to the reference surface temperature distribution is detected, and the process proceeds to the defect monitoring region position correcting step (step S108). FIG. 6E shows a state in which, in step S108, an area that most closely matches the reference shape image in the entire placement surface temperature distribution is determined. FIG. 6E shows only one placement surface temperature distribution among a plurality of placement surface temperature distributions, and shows a reference shape image corresponding to the placement surface temperature distribution.

Next, the defect monitoring area position correction process will be described.
In the defect monitoring region position correction step, the position of the placement surface temperature distribution is displaced according to the displacement amount of the placement surface temperature distribution with respect to the reference surface temperature distribution detected in step S100. Thereby, the defect monitoring area of the mounting surface temperature distribution is corrected (“thermal image deviation correction”) and approximated to the defect monitoring area of the reference surface temperature distribution (step S109). FIG. 6F shows a state in which the position of the placement surface temperature distribution is displaced in step S109 according to the amount of displacement detected in step S100. FIG. 6F shows only one placement surface temperature distribution among a plurality of placement surface temperature distributions, and a reference shape image corresponding to the placement surface temperature distribution.

Hereinafter, the temperature sudden change region detection step and the defect determination step will be described with reference to FIG. 1 and FIG. 7 and FIG.
FIG. 7 is a flowchart showing the contents of the pre-step of the temperature sudden change region detection step, the temperature sudden change region detection step, and the defect determination step. FIG. 8 is a diagram showing an example of processing performed by the defect detection means 6 in the temperature sudden change region detection step and the defect determination step.

As shown in FIG. 7, before performing the temperature sudden change region detection step, first, the pre-step of the temperature sudden change region detection step is started (step S200).
In the previous step of the temperature sudden change region detection step, the defect monitoring region is acquired from the defect monitoring region storage unit 4 and the part information of the pressed part P is acquired from the component information storage unit 2 to detect the temperature sudden change region detection step. (Step S201). The component information acquired from the component information storage unit 2 includes the threshold value used for the binarization process for the above-described temperature change region distribution, the above-described harmful defect size, and the defect internal / external temperature difference threshold value.

Next, the temperature sudden change region detection step will be described.
In the temperature sudden change region detection step, a plurality of placement surface temperature distributions obtained by correcting the position of the defect monitoring region in the defect monitoring region position correction step are acquired from the defect monitoring region position correction unit 12 (step S202).
The placement surface temperature distribution acquired in step S202 is an image displaying a defect monitoring area as shown in FIG. FIG. 8A shows only one placement surface temperature distribution among a plurality of placement surface temperature distributions.
As shown in FIG. 8A, in the placement surface temperature distribution acquired in step S202, the defect monitoring region is a region that does not include the void portion of the pressed part P.

In step S202, after acquiring a plurality of placement surface temperature distributions in which the positions of the defect monitoring areas are corrected, the spatial second-order differential processing represented by the above equation (1) is performed on each placement surface temperature distribution. (Step S203). Thereby, the plurality of placement surface temperature distributions are converted into temperature sudden change region distributions, which are distributions of temperature sudden change regions, respectively.
The placement surface temperature distribution subjected to the spatial second-order differential processing in step S203 becomes a temperature sudden change region distribution as shown in FIG. 8B. FIG. 8B shows only one placement surface temperature distribution among the plurality of placement surface temperature distributions subjected to the spatial second-order differential processing.

In step S203, a plurality of placement surface temperature distributions are converted into temperature sudden change region distributions, respectively, and one temperature sudden change region distribution is selected from the plurality of temperature sudden change region distributions (step S204).
In step S204, after selecting one temperature sudden change region distribution from the plurality of temperature sudden change region distributions, the binarization process described above is performed on the selected temperature sudden change region distribution (step S205). The threshold used for the binarization process in step S205 is the threshold acquired from the component information storage unit 2 in step S201.

  As shown in FIG. 8B, the temperature sudden change region distribution subjected to binarization processing in step S205 is a distribution in which only two types of regions, the temperature sudden change region and other regions, exist. Further, as shown in FIG. 8B, in the temperature sudden change region distribution subjected to the binarization process, in addition to the defect D occurring in the press part P, the end portion 32 of the press part P also has the temperature sudden change region. Is displayed.

In step S205, after binarization processing is performed on one selected temperature sudden change region distribution, a temperature sudden change region is detected from the temperature sudden change region distribution subjected to binarization processing. Further, from the temperature sudden change region distribution subjected to the binarization process, a portion including the temperature sudden change region and its peripheral portion is extracted as a determination region, and the process proceeds to the defect determination step (step S206).
The determination area extracted in step S206 includes a defect D occurring in the pressed part P, an end 32 of the pressed part P, and a peripheral part of both, as shown in FIG.

Next, the defect determination process will be described.
In the defect determination step, the position of the determination region extracted in step S206 is compared with the position of the corresponding defect monitoring region, and it is determined whether or not the temperature sudden change region included in the determination region exists in the defect monitoring region ( Step S207).
As shown in FIG. 8D, the defect D that has occurred in the pressed part P in the two temperature sudden change regions included in the determination region extracted in step S206 exists in the defect monitoring region. On the other hand, the end portion 32 of the pressed part P does not exist in the defect monitoring area.

Therefore, the defect D generated in the press part P in the two temperature sudden change regions is determined as a defect generated in the press part P. Moreover, the end part 32 of the press part P is not determined as a defect generated in the press part P in the two temperature sudden change regions.
If it is determined in step S207 that the temperature sudden change region included in the determination region exists in the defect monitoring region, the size of the temperature sudden change region is compared with the harmful defect size acquired from the component information storage unit 2 in step S201. Then, it is determined whether or not the size of the temperature sudden change region is equal to or larger than the harmful defect size (step S208).

On the other hand, if it is determined in step S207 that the temperature sudden change region included in the determination region does not exist in the defect monitoring region, the temperature sudden change region distribution other than the temperature sudden change region distribution selected in step S204 is selected from the plurality of temperature sudden change region distributions. Is selected (step S211). And it transfers to the process of step S204 mentioned above.
If it is determined in step S208 that the size of the temperature sudden change region is equal to or larger than the harmful defect size, a temperature difference (inside / outside temperature difference) between the surface temperature in the temperature sudden change region and the surface temperature around the temperature sudden change region is detected. Then, the detected internal / external temperature difference is compared with the defect internal / external temperature difference threshold acquired from the component information storage unit 2 in step S201 (step S209). Then, it is determined whether or not the detected temperature difference between the inside and outside of the area is equal to or less than the defect inside and outside temperature difference threshold.

On the other hand, if it is determined in step S208 that the size of the temperature sudden change region is smaller than the harmful defect size, it is determined that no defect has occurred in the position of the temperature sudden change region, and the process proceeds to the above-described step S211.
If it is determined in step S209 that the temperature difference between the inside and outside of the region is equal to or less than the defect inside / outside temperature difference threshold, it is determined that a defect has occurred at the position of the temperature sudden change portion (step S210). FIG. 8E shows a state in which a defect is generated in the pressed part P.

  In step S211, the process of selecting one temperature sudden change region distribution from a plurality of temperature sudden change region distributions is repeated. And as a result of performing the process of step S207, step S208, and step S209 with respect to all the temperature sudden change area | region distribution, when it transfers to the process of step S211, since the defect has not generate | occur | produced in the press part P, press part It is determined that P is a non-defective product (non-defective product determination), and the process is terminated (step S212).

(effect)
Therefore, according to the defect detection method of the present embodiment, in the defect determination step, a defect occurring in the pressed part P is detected based on the temperature sudden change region in the surface temperature distribution of the pressed part P. Further, in the temperature sudden change region detection step, the temperature sudden change region is detected by a spatial second-order differential process with respect to the surface temperature distribution. Further, in the defect determination step, a defect occurring in the pressed part P is detected based on the temperature sudden change region detected in the temperature sudden change region detection step.

For this reason, it is not affected by changes in the processing heat generated in the press part P and the mold and the temperature change around the press part P, and further, disturbances such as oil film of oil used for press processing are eliminated, and the press part It becomes possible to detect defects occurring in P.
As a result, it is possible to suppress oversight and erroneous detection of defects occurring in the pressed part P, and it is possible to improve the detection accuracy of defects occurring in the pressed part P.

Moreover, if it is the defect detection method of this embodiment, a defect monitoring area | region memory | storage process will be performed as a pre-process of a defect determination process. Further, in the defect determination step, a defect occurring in the pressed part P is detected based on the defect monitoring area in the surface temperature distribution of the pressed part P.
For this reason, when a temperature sudden change region exists outside the defect monitoring region, that is, at a position where no defect occurs, such as a gap of the press part P, the detected temperature sudden change region is determined as a defect occurring in the press part P. It can be excluded from the target.

As a result, it is possible to suppress erroneous detection of defects generated in the pressed part P, and it is possible to improve the detection accuracy of defects generated in the pressed part P.
Further, in the defect detection method of the present embodiment, when the position of the press part P is displaced from the surface temperature distribution detection position in the defect monitoring region position correction step, the placement surface is based on the amount of displacement. Correct the position of the temperature distribution.

For this reason, even if the position of the mounting surface temperature distribution is displaced from the reference surface temperature distribution, the position of the defect monitoring area of the mounting surface temperature distribution is determined based on the amount of displacement. It is possible to correct the position.
As a result, since it becomes possible to correct the defect monitoring area to an appropriate position, it becomes possible to suppress erroneous detection of defects occurring in the pressed part P, and to improve the detection accuracy of defects occurring in the pressed part P. It becomes possible to make it. In addition, it is possible to cope with an increase in the speed of the production line for the press part P, and it is possible to improve the production efficiency of the press part P.

  In the defect detection method of the present embodiment, the defect monitoring area is divided into a plurality of defect monitoring areas, and the surface of the press part P is divided into a plurality of surface temperature distribution detection areas corresponding to the plurality of defect monitoring areas. To do. Further, in the temperature sudden change region detection step, the temperature sudden change region is detected for each of the plurality of surface temperature distribution detection regions by the spatial second-order differential processing for the surface temperature distribution detected for each of the plurality of surface temperature distribution detection regions. Moreover, in a defect determination process, it is determined for every some surface temperature distribution detection area | region whether the defect generate | occur | produced in the press parts P or not.

For this reason, the region predicted to have a defect in the pressed part P is divided into a plurality of ranges in accordance with the occurrence frequency of the defect, and the defect generated in the pressed part P is detected for each of the divided regions. It becomes possible.
As a result, since it becomes possible to detect defects occurring in the pressed part P with high accuracy for each part of the pressed part P, it is possible to suppress oversight and erroneous detection of defects generated in the pressed part P. Thus, it becomes possible to improve the detection accuracy of defects generated in the pressed part P.

(Application examples)
Moreover, although the defect detection method of this embodiment includes the defect monitoring area | region storage process which memorize | stores a defect monitoring area | region previously, it is not limited to this. In other words, for example, when the thermal image captured by the infrared camera is a portion of the pressed part that does not have a gap, such as when the entire thermal image captured by the infrared camera corresponds to the defect monitoring area, It is good also as a process which does not include a defect monitoring area | region memory | storage process.

  Further, in the defect detection method of the present embodiment, the defect detection process includes a temperature distribution displacement amount detection process and a defect monitoring region position correction process. However, the present invention is not limited to this. In other words, if it is possible to ideally place the pressed parts after the press working at the surface temperature distribution detection position, the defect detection process includes a temperature distribution displacement amount detection process and a defect monitoring area position correction process. There may be no process.

(Example)
Hereinafter, referring to FIG. 1 to FIG. 8, using FIG. 9 and FIG. 10, a press using a defect detection method similar to the defect detection method of the present embodiment installed in the production line for the pressed part P is used. The example which detected the defect which generate | occur | produced in the component P is shown.
FIG. 9 is a diagram showing a placement surface temperature distribution.
As shown in FIG. 9, by using the defect detection method, an area that best matches the reference shape image in the entire placement surface temperature distribution of the pressed part P (shown as “reference image matching position” in FIG. 9). Range) can be detected.

FIG. 10 is a diagram illustrating a state in which the position of the placement surface temperature distribution is displaced according to the amount of displacement of the placement surface temperature distribution with respect to the reference surface temperature distribution in the defect monitoring region position correction step.
As shown in FIG. 10, by using the defect detection method, the displacement amount of the mounting surface temperature distribution with respect to the reference surface temperature distribution is “0” in the X direction and “−2” in the Y direction. Is detected. Based on the detected displacement amount (shown as “panel displacement correction amount” in FIG. 10), the defect generated in the press part P by displacing the position of the placement surface temperature distribution in the defect monitoring region position correction step. D (denoted as “automatically recognized defect” in FIG. 10) can be automatically detected.
Therefore, it is confirmed that it is possible to improve the production efficiency in the production line of the press part P by detecting the defect generated in the press part P using the same defect detection method as the defect detection method of the present embodiment. did.

It is a figure which shows the defect detection method of this invention. It is a figure which shows the state which determines whether the temperature sudden change area | region exists in a defect monitoring area | region in one defect monitoring area | region. It is a figure which shows the state which compares the magnitude | size of a rapid temperature change area | region, and a harmful defect size. It is a figure which shows the state which compares the detected temperature difference and a defect part inside / outside temperature difference threshold value. It is a flowchart which shows the content of the pre-process of a surface temperature distribution detection process, a surface temperature distribution detection process, a temperature distribution displacement amount detection process, and a defect monitoring area | region position correction process. It is a figure which shows the example of the process which a surface temperature distribution detection means and a defect detection means perform in a surface temperature distribution detection process, a temperature distribution displacement amount detection process, and a defect monitoring area | region position correction process. It is a flowchart which shows the content of the pre-process of a temperature sudden change area | region detection process, a temperature sudden change area | region detection process, and a defect determination process. It is a figure which shows the example of the process which a defect detection means performs in a temperature sudden change area | region detection process and a defect determination process. It is a figure which shows the mounting surface temperature distribution in the Example of this invention. It is a figure which shows the state which displaced the position of mounting surface temperature distribution according to the displacement amount of mounting surface temperature distribution with respect to reference | standard surface temperature distribution in the defect monitoring area | region position correction process in the Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Surface temperature distribution detection means 2 Parts information storage means 4 Defect monitoring area | region storage means 6 Defect detection means 8 Infrared camera 10 Temperature distribution displacement amount detection means 12 Defect monitoring area position correction means 14 Temperature sudden change area detection means 16 Defect determination means 18 Reference | standard Surface temperature distribution storage unit 20 Displacement amount detection unit 22 Spatial second derivative processing unit 24 Determination region extraction unit 26 Temperature sudden change region position determination unit 28 Temperature sudden change region size determination unit 30 Temperature sudden change region internal / external temperature difference determination unit 32 End of press part Part P Press parts E Temperature change region D Defect

Claims (4)

A method for detecting a defect in a pressed part that detects a defect generated in a pressed part after pressing,
A surface temperature distribution detecting step for detecting a surface temperature distribution of the pressed part after press working, and a defect detecting step for detecting a defect generated in the pressed part based on the surface temperature distribution detected in the surface temperature distribution detecting step. And having
The defect detection step includes a temperature sudden change region detection step of detecting a temperature sudden change region in which the temperature has changed from a peripheral region by a predetermined temperature difference or more, out of the surface temperature distribution detected in the surface temperature distribution detection step, and the temperature A defect determination step for determining whether or not a defect has occurred in the pressed part, based on the temperature sudden change region detected in the sudden change region detection step,
In the temperature sudden change region detection step, the temperature sudden change region is detected by a spatial second derivative process for the surface temperature distribution detected in the surface temperature distribution detection step ,
In the defect determination step, a temperature difference between the surface temperature of the press part in the temperature sudden change region detected in the temperature sudden change region detection step and the surface temperature of the press part around the temperature sudden change region detected in the temperature sudden change region detection step is calculated. If the detected temperature difference is equal to or less than the defect internal / external temperature difference threshold value set in advance for each pressed part having at least one of a different shape and material, a defect occurs in the pressed part. A method for detecting a defect of a pressed part, characterized in that it is determined that the pressed part is defective. A defect monitoring area storing step of preliminarily storing a defect monitoring area which is an area predicted to have a defect in the pressed part among the surface temperature distribution detected in the surface temperature distribution detecting process;
The defect monitoring area storage process is a pre-process of the defect determination process,
In the defect determination step, when the temperature sudden change region detected in the temperature sudden change region detection step is present in the defect monitor region based on the defect monitor region stored in the defect monitor region storage step, the press part 2. The method for detecting a defect in a pressed part according to claim 1, wherein it is determined that a defect has occurred. In the surface temperature distribution detection step, a placement surface temperature distribution that is a surface temperature distribution of the pressed part placed at a predetermined surface temperature distribution detection position after pressing is detected,
In the defect detection step, based on a reference surface temperature distribution that is a surface temperature distribution of a pressed part when ideally arranged at the surface temperature distribution detection position, the displacement of the placement surface temperature distribution relative to the reference surface temperature distribution is performed. Based on the temperature distribution displacement amount detection step for detecting the amount and the displacement amount detected in the temperature distribution displacement amount detection step, the defect monitoring region of the placement surface temperature distribution is approximate to the defect monitoring region of the reference surface temperature distribution And a defect monitoring region position correcting step for correcting the position of the surface temperature distribution described above,
In the defect determination step, the temperature sudden change region detected in the temperature sudden change region detection step is present in the defect monitor region based on the defect monitor region of the surface temperature distribution described above corrected in the defect monitor region position correction step. 3. The press part defect detection method according to claim 2, wherein when the press part is determined, it is determined that a defect has occurred in the press part. Dividing the defect monitoring area into a plurality of defect monitoring areas;
Dividing the surface of the pressed part into a plurality of surface temperature distribution detection regions corresponding to the plurality of defect monitoring regions,
In the surface temperature distribution detection step, the surface temperature distribution of the pressed part is detected for each of the plurality of surface temperature distribution detection regions,
In the temperature sudden change region detection step, the temperature sudden change region is detected in the plurality of surface temperature distributions by a spatial second-order differential process for the surface temperature distribution detected for each of the plurality of surface temperature distribution detection regions in the surface temperature distribution detection step. Detect for each area,
In the defect determination step, based on the temperature sudden change region detected for each of the plurality of surface temperature distribution detection regions in the temperature sudden change region detection step, it is determined whether or not a defect has occurred in the pressed part. 4. The pressed part defect detection method according to claim 2, wherein the determination is made for each distribution detection region.