KR102000907B1 - Appearance vision inspection method for ferrite part - Google Patents

Abstract

The present invention relates to a method for inspecting exterior vision of ferrite by using an inspection system comprising first to third exterior inspecting units (100 to 300) each having a lighting unit and a photographing unit, a control unit (400), and a sorting unit (500), the method comprising the steps of: a presetting step of controlling the inspection system and inputting setting data according to the type of a ferrite model through the control unit (400) (S10); a first exterior inspecting step of loading the ferrite on the first exterior inspecting unit (100) and generating photographic data on the left region (Z1) and the right region (Z2) (S20); a second exterior inspecting step of transferring the ferrite to the second exterior inspecting unit (200) and generating photographic data on an upper center region (Z3), an upper left region (Z4), and an upper right region (Z5) (S30); a third exterior inspecting step of transferring the ferrite to the third exterior inspecting unit (300) and generating photographic data on a lower left region (Z6), a lower right region (Z7), and a lower center region (Z8) (S40); a reading step of receiving the photographic data on the regions and generating reading data by machine vision image processing to detect defects (S50); and a sorting step of transferring the ferrite to the sorting unit (500) and selectively discharging the ferrite according to the reading data (S60). Thus, defects such as fine cracks can be accurately and efficiently detected in the production of the ferrite.

Description

FIELD OF THE INVENTION [0001] The present invention relates to a ferrite outer appearance inspection method,

The present invention relates to a ferrite external vision inspection method, and more particularly, to a ferrite external appearance inspection method for ferrite (hereinafter referred to as " ferrite ") which uses machine vision to accurately and quickly detect defects And an appearance inspection method.

Ferrite, which is an oxide-based magnetic material obtained by mixing and sintering iron oxide with manganese, nickel, and zinc, is used in various industrial fields. For example, a component that requires a heat resistance characteristic, such as an automobile exhaust system passage, is mainly composed of a ferrite-based alloy excellent in heat resistance and oxidation resistance characteristics in a temperature range of hundreds of degrees Celsius. In addition, components made of a ferrite magnetic material are mounted on various electronic products requiring magnetic properties.

On the other hand, since ferrite is formed by using an alloy material made of an iron oxide compound, a large amount of various impurities remain even though the refining process is performed. Therefore, ferrite has a problem such that cracks or fine silks are generated in the manufacturing process, or appearance damage or deformation occurs in the course of processing such as rolling.

When the ferrite is attached to a device in a state where defects such as cracks are generated, it may cause deterioration of the performance and lifetime of the device. Therefore, when the ferrite is manufactured, the process of inspecting the appearance is performed to select good products and defective products Work is being done.

In general, a method for inspecting the appearance of a ferrite is generally a method in which a test stand is provided on one side of a conveyor for conveying ferrite and a worker directly observes the appearance of the part with the naked eye. Recently, the ferrite being transferred is photographed with a camera, A technique for detecting an abnormality is applied to an inspection method in which an operator confirms an image.

As an example of a known technique, Korean Patent Laid-Open No. 10-1997-0058518 discloses a method of visually inspecting the appearance of a semi-cylindrical ferrite core, comprising a main controller for checking the appearance of the ferrite core to judge a positive / A frame / grabber section for modifying the inspection position of the ferrite core; and a frame / grabber section for detecting the position of the ferrite core in the infrared sensor, And a camera for capturing an image of the ferrite core at a plurality of positions.

As another example, Korean Patent Registration No. 10-0188400 discloses a ferrite core comprising a core conveying portion provided at one side of a conveyor belt to which a ferrite core is fed and installed so as to sandwich and convey ferrite cores, a camera fixing plate And a CCD camera supported and fixed to the upper center of the camera fixing plate on the upper side of the core holding plate, and a ferrite core And constitutes a testing apparatus including a core inspecting section installed for carrying out the work.

Korean Patent Laid-Open No. 10- 1997-0058518 (July 31, 1997) Korean Patent No. 10 - 0158400 (December 15, 1998) Korean Patent Publication No. 10-1994-0048527 (July 29, 1997) Korean Patent Publication No. 10- 1997- 0058491 (July 31, 1997)

Conventionally, it is a common practice for a worker to conduct inspection in such a manner as to visually check the appearance of the product when the ferrite is manufactured. However, such a method has a disadvantage in that not only the reliability is low, but also productivity and efficiency are low due to the generation of labor costs and the delay of the process time because the inspection quality is different according to the skill level of the operator or the judgment standard.

Due to such a problem, recently, a ferrite being conveyed is photographed with a digital camera and an inspection method in which an operator confirms an image through a monitor screen is applied. However, the inspection method using a general digital camera has a limitation in that it is possible to detect defects only at a recognizable level from a photographed image.

On the other hand, there is an example in which a 2D line scan camera or a 3D line scan camera is applied to a conventional ferrite inspection instead of a general digital camera. However, as known, the 2D line scan camera is susceptible to generation of noise because the photographing is performed while linearly moving the inspected object or the camera, and it is very difficult to distinguish cracks due to the dark color and surface texture particular to ferrite. Therefore, in the case of micro cracks, there is a disadvantage that not only the detection rate is small but also the detection efficiency by the software image processing is low, so that the hardware such as the shutter speed and the FPS requires high quality. In addition, 3D cameras are expensive because of the high cost of the equipment itself, and it takes about 10 to 20 seconds for each unit of the inspected object, and the amount of captured data is so large that the processing time is also long. There is a disadvantage in that the efficiency is considerably low for application to the field.

Accordingly, the present invention has been made to solve the above-mentioned problems of the prior art,

A method for inspecting a ferrite outer appearance vision using an inspection system comprising a first outer appearance inspection unit to a third outer appearance inspection unit (100-300), a control unit (400), and a selection unit (500)

A presetting step (S10) of controlling the inspection system and inputting the setting data according to the type of the ferrite model through the control unit (400);

A first appearance inspection step (S20) of loading ferrite in the first appearance inspection unit (100) and generating photographic data of the left area (Z1) and the right area (Z2);

A second visual inspection step S30 for transferring ferrite to the second visual inspection unit 200 and generating photographic data of the upper central region Z3, the upper left region Z4 and the upper right region Z5;

A third visual inspection step (S40) of transferring ferrite to the third visual inspection unit (300) and generating photographic data of the lower left zone (Z6), the lower right zone (Z7) and the lower central zone (Z8);

A reading step (S50) for receiving photographic data for the areas in the control unit (400) and generating read data by machine vision image processing to detect defects; And

And a selection step (S60) of transferring ferrite to the sorting unit (500) and selectively discharging ferrite according to the read data. This makes it possible to accurately and efficiently detect defects such as micro cracks It is possible to achieve.

The present invention provides an appearance inspection method using machine vision in order to efficiently detect defects such as micro-cracks that occur during the manufacture and processing of ferrite.

That is, the present invention obtains high-resolution photographed data for each region of ferrite, and makes it possible to quickly and precisely conduct a visual inspection by machine-running image processing based on the obtained high-resolution photographed data.

Accordingly, the present invention significantly increases the productivity and efficiency compared to the conventional visual inspection method or image inspection method using a line scanner camera, and also allows setting and inspection standards of an inspection system according to different specifications and shapes for each model of ferrite And can be easily reflected in the inspection process. In particular, there is an advantage that the inspection accuracy and distinguishing power by deep running can be remarkably improved.

The present invention has the effect of minimizing the detection error through automation of the entire ferrite appearance inspection work, significantly improving the reliability of the inspection result, and further enabling unattended appearance inspection work.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a flow chart of a ferrite external vision inspection method according to the present invention; FIG.
2 is a schematic block diagram of an inspection system used in a ferrite external vision inspection method according to the present invention.
3 is a plan view showing a process flow of the first to third external inspection steps and the selection step of the ferrite external vision inspection method according to the present invention on an inspection system.
FIGS. 4 to 6 are diagrams showing an embodiment of the first external appearance inspection step to the third external appearance inspection step of the ferrite external vision inspection method according to the present invention. FIG.
7 to 12 are views showing an embodiment of a reading step of a ferrite external vision inspection method according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments of the invention are shown. In the following description, a detailed description of parts that can be easily implemented by those skilled in the art may be omitted. It is to be understood that the present invention is not limited to the following embodiments, and various modifications may be made without departing from the scope of the present invention. will be.

2 is a schematic block diagram of an inspection system used in a ferrite external vision inspection method according to the present invention. FIG. 3 is a schematic view of a ferrite external vision inspection method according to the present invention. FIG. 4 is a plan view showing the process flow of the first external inspection step to the third external inspection step and the selection step of the ferrite external appearance inspection method according to the present invention. FIG. 7 to FIG. 12 are images showing an embodiment of the reading step of the ferrite external vision inspection method according to the present invention.

The ferrite external vision inspection method to which the technique of the present invention is applied relates to a visual inspection method using a machine vision, which is capable of more accurately detecting defects such as micro-cracks, which are hard to be visually recognized during manufacture of ferrite components (hereinafter referred to as ferrite) I know.

The ferrite external vision inspection method of the present invention includes the steps of constructing an inspection system in the latter part of the ferrite manufacturing system to load a ferrite taken out in a state in which impurities are removed through a firing unit and a cleaning unit to remove fine silks, To check whether or not. Hereinafter, the present invention is applied to a ferrite core manufacturing system widely used for automotive parts, and a series of inspection processes for detecting defects in ferrite as an inspected object will be described as an example.

The inspection system used in the inspection method of the present invention includes a first exterior inspection unit to a third exterior inspection unit (100 to 300), and a control unit (400) and a screening unit (500). Ferrite is transferred to the first to third external inspection units 100 to 300 so as to inspect each area. The control unit 400 is made up of a computer and a computer program stored in the medium so as to control the entire inspection process. The sorting unit (500) is interlocked with the control unit (400) so as to selectively discharge ferrite as a good product, a defective product, and a retention product according to a read result.

In the ferrite external vision inspection method according to the present invention, the outer appearance of a three-dimensional ferrite having a curved surface is divided into a left region Z1, a right region Z2, an upper central region Z3, The upper left region Z4, the upper right region Z5, the lower left region Z6, the lower right region Z7, and the lower central region Z8 to examine each region. As shown in FIG. 1 It is also possible to include a preset step S10, a first appearance inspection step S20, a second appearance inspection step S30, a third appearance inspection step S40, a reading step S50, and a selection step S60 .

The presetting step S10 is a step of controlling the inspection system according to the type of the ferrite model through the control unit 400 and inputting the setting data. That is, the presetting step (S10) is a step of preliminarily setting an inspection environment and an inspection standard suitable for the ferrite model before loading the ferrite into the inspection system in earnest.

In the presetting step S10, a control value for controlling the operation states of the photographing unit and the illumination unit of the first to third visual inspection units 100 to 300 is set. In the inspection system used in the present invention, the photographing unit and the illuminating unit are mounted at eight locations so that the eight areas of the ferrite are inspected as described above. In the presetting step (S10), the photographing condition of each photographing unit, Set the luminance control value.

In the presetting step S10, a model of ferrite as an inspected object is selected and a selection criterion value for a defect area and a dimension generated in the model is set. The physical properties of ferrites vary depending on the model, and the types of defects that occur during the manufacturing process also vary. Therefore, the sorting reference value is set according to the ferrite model through the presetting step S10, and is applied to the reading step S50.

The presetting step S10 configures the operator to manually input or automatically apply the setting through the interface of the processing module 410 and the control module 420 of the control unit 400, Display.

The first visual inspection step S20 is a step of loading the ferrite into the first visual inspection unit 100 and generating photographic data of the left area Z1 and the right area Z2.

4, the first external inspection unit 100 includes a left side inspection module 120 and a right side inspection module 130 on the left and right sides of a first conveyance module 110 for placing ferrite, Respectively. The left side inspection module 120 includes a first left illumination part 121 and a first left side imaging part 122 mounted on the first left side part 123 and disposed horizontally, The illumination unit 131 and the first right side photographing unit 132 are mounted on the first right side position part 133 and disposed horizontally.

Therefore, in the first appearance inspection step S20, the left and right sides of the ferrite are first conveyed to the first conveyance module 110 so as to face the left side inspection module 120 and the right side inspection module 130, respectively, The ferrite is positioned at a distance of 40 to 80 mm on the central portion between the illumination unit 121 and the first right illumination unit 131. The distance between the ferrite and both side illumination portions is set by the first adjusting means 116 controlled by the control unit 400. [ The first left photographing part 122 and the first right photographing part 132 located behind the first left illuminating part 121 and the first right illuminating part 131 have a sufficient viewing angle corresponding to an area three times the size of the ferrite FOV).

When the ferrite is positioned, light having a luminance of 50 to 150 cd / m 2 and a wavelength of 470 nm or less is incident on the left side zone Z 1 and the right side zone Z 2 of the ferrite in the first left illumination unit 121 and the first right illumination unit 131 Investigate. Since a long wavelength range light source has a high reflectivity at the time of irradiation of ferrite having a specific surface texture, it is very difficult to detect defects due to software image processing techniques such as machine vision from photographed data. Accordingly, the blue light having the brightness and the wavelength range is irradiated to separate the ferrite texture and the micro cracks.

When the illumination unit is set, the first left-side photographing unit 122 and the first right-side photographing unit 132 photograph the left zone Z1 and the right zone Z2 of the ferrite with a resolution of 20 M pixels or more and a frame rate per second of 6 fps or more . In the case of micro cracks generated during the manufacturing process of ferrite, those formed at a size of 100 μm or less can not be distinguished from images taken by a normal medium and low resolution camera. Therefore, in this step, the resolution is determined to be about 10.5 μm The left area Z1 and the right area Z2 of the ferrite are photographed to generate photographic data.

When the photographing is completed, the photographing data of the left area Z1 and the right area Z2 are transmitted to the control unit 400. [ The first visual inspection unit 100 and the control unit 400 are configured to perform real-time transmission through PLC communication.

The second visual inspection step S30 transfers ferrite to the second visual inspection unit 200 and generates photographic data of the upper central region Z3, the upper left region Z4, and the upper right region Z5 .

As shown in FIG. 5, the second external inspection unit 200 includes an upper central inspection module 220 on a vertical upper side of a second conveyance module 210 for positioning a ferrite, The upper left inspection module 230 and the upper right inspection module 240 are symmetrically disposed on the upper left side and the upper right side in the oblique direction of the upper left side inspection module 210, respectively. The second central illuminating section 221 and the second central illuminating section 222 are vertically disposed on the second center level section 223 and the upper left illuminating section 230 is provided with the second left central illuminating section 221, The illuminating unit 231 and the second left photographing unit 232 are mounted on the second left position unit 233 and disposed in the oblique direction and the upper right examining module 240 is provided with the second right illuminating unit 241 and the second right photographing unit 233, And the second right side portion 243 is disposed in the diagonal direction.

Therefore, in the second appearance inspection step S30, the upper left and right sides of the ferrite are first conveyed to the first conveying module 110 so as to be positioned in the direction of the upper left inspection module 230 and the upper right inspection module 240, respectively And the ferrite is positioned at a distance of 40 to 80 mm on the lower central portion of the second central illumination section 221. The distance between the ferrite and the upper illumination portion is set by the third adjusting means 224 controlled by the control unit 400. [ The second central photographing unit 222 located behind the second central illuminating unit 221 secures a sufficient viewing angle corresponding to three times the area of the ferrite size.

When the ferrite is located, the second central illuminating unit 221 irradiates the upper central region Z3 of the ferrite with light having a luminance of 50 to 150 cd / m 2 and a wavelength region of 470 nm or less. Since the curved surface formed in the upper central region Z3 of the ferrite has a high reflectivity when irradiated with a light source having a long wavelength region, it is possible to irradiate light of a blue tone having the above luminance and wavelength range to distinguish the surface texture of ferrite and the fine crack .

When the illumination unit is set, the second central photographing unit 222 photographs the upper central region Z3 of the ferrite with a resolution of 20 M pixels or more and a frame rate per second of 6 fps or more. The upper central region Z3 of the ferrite is photographed so as to derive a resolution of about 10.5 mu m in the resolution range to generate photographed data so that micro-cracks having a size of 100 mu m or less can be identified.

When the photographing of the upper central region Z3 is completed, the ferrite is moved to the second conveying module 210 at a distance of 40 to 80 mm on the lower center portion in the diagonal direction between the second left illumination portion 231 and the second right illumination portion 241 Place the ferrite. The diagonal distance between the ferrite and both side illumination portions is set by the second adjusting means 216 controlled by the control unit 400 and the fourth adjusting means 234 and the fifth adjusting means 244. The second left side illumination part 231 and the second left side imaging part 232 and the second right side imaging part 242 located behind the second right illumination ensure a sufficient viewing angle corresponding to three times the area of the ferrite size .

When the ferrite is located, the upper left region Z4 and the upper right region Z5 of the ferrite in the second left illumination portion 231 and the second right illumination portion 241 have a luminance of 50 to 150 cd / m 2 and a wavelength region of 470 nm or less Inspect the light. Since the curved surface and the corner portion formed in the upper left region Z4 and the upper right region Z5 of the ferrite have high reflectivity at the time of irradiating a light source having a long wavelength region, the light of the blue tone having the above- The surface texture of the ferrite, and the distinction between the edge and the micro crack.

When the illuminating unit is set, the upper left zone Z4 and the upper right zone Z5 of the ferrite are displayed in the second left-side photographing unit 232 and the second right-side photographing unit 242 with a resolution of 20M pixels or more and a frame rate per second I shoot. The upper left region Z4 and the upper right region Z5 of the ferrite are photographed so as to derive resolution of about 10.5 탆 in the resolution range to generate photographed data so as to be able to identify a microcrack crack having a size of 100 탆 or less.

The photographing data of the upper central zone Z3, the upper left zone Z4 and the upper right zone Z5 are transmitted to the control unit 400. [

The third visual inspection step S40 transfers the ferrite to the third visual inspection unit 300 and generates photographic data of the lower left zone Z6, the lower right zone Z7 and the lower central zone Z8 .

6, the third external inspection unit 300 includes a lower left inspection module 320 and a lower right inspection module 320 on the lower left side and the lower right side in the diagonal direction of the third feeding module 310 for positioning the ferrite, And the lower central inspection module 340 is provided vertically below the third conveying module 310. The lower central inspection module 340 is provided at the lower side of the third conveying module 310, The third left side illumination part 321 and the third left side imaging part 322 are mounted on the third left side part 323 and arranged in the oblique direction in the lower left side inspection module 320, The third right lighting part 331 and the third right lighting part 332 are mounted on the third right position part 333 and arranged in oblique directions. The lower central inspection module 340 is provided with the third central illumination part 341 and third And the central photographing unit 342 is mounted on the third central position unit 343 and disposed vertically.

Therefore, in the third appearance inspection step (S40), ferrite is fed to the third conveying module 310 so that the lower left and right sides of the ferrite are positioned in the direction of the lower left inspection module 320 and the lower right inspection module 330, respectively And the ferrite is positioned at a distance of 40 to 80 mm on the upper center portion in the oblique direction between the third left illuminating portion 321 and the third right illuminating portion 331. The distance between the ferrite and the illuminating unit is set by the sixth adjusting means 324 and the seventh adjusting means 334 which are controlled by the control unit 400. [ The third left side photographing portion 322 and the third right side photographing portion 332 located behind the third left illuminating portion 321 and the third right illuminating portion 331 have a sufficient viewing angle corresponding to three times the area of the ferrite size .

When the ferrite is located, the luminance in the lower left region Z6 and the lower right region Z7 of the ferrite in the third left illuminating portion 321 and the third right illuminating portion 331 is 50 to 150 cd / m < 2 & Inspect the light. Since the curved surface and the corner formed in the lower left area Z6 and the lower right area Z7 of the ferrite have high reflectivity at the time of irradiating a light source having a long wavelength range, the light of the blue tone having the luminance and the wavelength range is irradiated The surface texture of the ferrite, and the distinction between the edge and the micro crack.

When the illuminating unit is set, the third left-side photographing unit 322 and the third right-side photographing unit 332 have a resolution of 20M pixels or more and a number of frames per second of 6 fps or more and a lower left zone Z6 and a lower right zone Z7 I shoot. The lower left region Z6 and the lower right region Z7 of the ferrite are photographed so as to derive the resolving power to the level of about 10.5 mu m in the resolution range to generate photographed data so as to be able to identify a microcrystalline crack having a size of 100 mu m or less.

When photographing of the lower left side zone Z6 and the lower right side zone Z7 is completed, the ferrite is positioned at a distance of 40 to 80 mm on the upper central portion of the third central illumination unit 341 by the third conveyance module 310 . The distance between the ferrite and the illuminating unit is set by the eighth adjusting means 344 controlled by the control unit 400. [ The third central photographing unit 342 located behind the third central illuminating unit 341 ensures a sufficient viewing angle corresponding to three times the area of the ferrite size.

When the ferrite is positioned, the third central illumination unit 341 irradiates light having a luminance of 50 to 150 cd / m 2 and a wavelength region of 470 nm or less in the lower central region Z 8 of the ferrite. Since the recessed curved surface formed in the upper central region (Z3) of the ferrite has a high reflectivity when irradiated with a light source having a long wavelength region, the light of the blue tone having the luminance and the wavelength range is irradiated, .

When the illumination unit is set, the third central photographing unit 342 photographs the lower central region Z8 of the ferrite with a resolution of 20 M pixels or more and a frame rate per second of 6 fps or more. The lower central region (Z8) of the ferrite is photographed so as to derive a resolution of about 10.5 mu m in the resolution range, and photographic data is generated so that micro-cracks having a size of 100 mu m or less can be identified.

The photographing data of the lower right zone Z7, the lower left zone Z6 and the lower central zone Z8 is transmitted to the control unit 400 when the photographing is completed.

On the other hand, the reading step S50 is a step of receiving photographed data for the areas in the control unit 400 and generating defect data by machine vision image processing to detect defects.

The control unit 400 includes a computer and a computer program stored in the medium, and controls the first to third exterior inspection units 100 to 300 and the later-described sorting unit 500 to be controlled.

Therefore, in the reading step (S50), the photographing data of the eight areas of the ferrite transferred from the first to third external inspection units (100 to 300) are subjected to image processing to emphasize the contrast, do.

As shown in Figs. 7 to 11, the left region Z1, the right region Z2, the upper central region Z3, the upper left region Z4, the upper right region Z5, Bottom left region Z6, lower right region Z7, and lower central region Z8 The image data is subjected to binarization processing by extracting blobs from the photographed data with emphasis on the photographed data, The read data is calculated for each type. The binarization image processing processes all the pixels on the photographic data with black and white based on the threshold value, thereby clearly displaying the contrast difference. The threshold value can be used by calculating an average value of brightness for all the pixels of the photographed data, and it is also possible to apply the variable according to the ferrite model. When the blob is extracted from the binarized photographic data, as shown in Fig. 12, read data is classified into types of defects such as a thread crack, a border crack, a vertical crack, a horizontal crack, a dark point defect, .

In addition, based on the preset data inputted in the presetting step S10, the read data is compared and analyzed to generate control signals corresponding to good, defective, and stored products according to the selection reference value and transmitted to the selection unit 500 .

In addition, the reading step (S50) includes a step of detecting an accuracy with respect to the machining dimension of the ferrite. That is, a boundary line is calculated by applying a differential function to two points having the largest slopes in the photographed data according to pixel positions, and the read data is calculated by calculating the boundary, and the defect is detected by comparing with the setting data.

Meanwhile, in the reading step (S50), the machine vision deep learning technique is applied to learn read data, so that it is configured to refer to the detection of a defect in the future. In other words, by learning the read data for micro-cracks and the like which are formed in arbitrary various forms in the course of image processing and reading of the photographed data, discrimination power is increased and rapid detection ability is ensured. As for the deep learning algorithm, a well-known technique will be referred to, and a detailed description thereof will be omitted.

The sorting step S60 is a step of feeding ferrite to the sorting unit 500 and selectively discharging ferrite according to the read data.

3, the sorting unit 500 includes first to third discharge belts 510 to 530 and selects the discharge direction of ferrite in accordance with a control signal transmitted from the control unit 400 And has a first sorting handle 540 and a second sorting handle 550.

Accordingly, in the selecting step S60, when the ferrite passing through the reading step S50 is read as a good product, a defective product, and a storage product, the first sorting handle 540 is operated according to each control signal, (510) or operates the second sorting table to discharge it to the second discharge stand (520) or the third discharge stand (530).

As described above, the ferrites selectively discharged through the first to third discharge bands 510 to 530 are unloaded to the loading means, and a predetermined autonomous traveling bogie is driven to be transported to carry out the inspection of the appearance of the ferrite, A fully automatic unattended operation can be realized.

As described above, the present invention provides an inspection method for effectively detecting defects such as micro-cracks generated during the manufacturing process of ferrite.

That is, according to the present invention, ferrites having curved surfaces, corners and the like and having a specific surface texture are divided into a left region Z1, a right region Z2, an upper central region Z3, an upper left region Z4, Z5), a lower left area Z6, a lower right area Z7, and a lower center area Z8, and a detailed visual inspection is performed by machine vision image processing based on the high resolution photographing data for each area .

Accordingly, the present invention significantly increases the productivity and efficiency compared to the conventional visual inspection method or image inspection method using a line scanner camera, and also allows setting and inspection standards of an inspection system according to different specifications and shapes for each model of ferrite And can be easily reflected in the inspection process. In particular, there is an advantage that the inspection accuracy and distinguishing power by deep running can be remarkably improved.

Therefore, it is expected that the present invention can be used industrially because the present invention has various advantages such as automation of the entire ferrite inspection work, minimization of detection error, and remarkable improvement in reliability of test results.

S10: Preset phase
S20: First appearance inspection step
S30: Second appearance inspection step
S40: Third appearance inspection step
S50:
S60: Selection step
100: First visual inspection unit
200: Second visual inspection unit
300: Third visual inspection unit
400: control unit
500: Selection unit

Claims (6)

A method for inspecting a ferrite outer appearance vision using an inspection system comprising a first outer appearance inspection unit to a third outer appearance inspection unit (100-300), a control unit (400), and a selection unit (500)
A presetting step S10 for controlling the inspection system and inputting the setting data in accordance with the type of the ferrite model through the control unit 400,
A first appearance inspection step (S20) of loading ferrite in the first appearance inspection unit (100) and generating photographic data of the left area (Z1) and the right area (Z2)
A second visual inspection step S30 for transferring ferrite to the second visual inspection unit 200 and generating photographic data of the upper central region Z3, the upper left region Z4 and the upper right region Z5 ,
A third visual inspection step S40 for transferring ferrite to the third visual inspection unit 300 and generating photographic data of the lower left area Z6, the lower right area Z7 and the lower central area Z8 ,
A reading step (S50) for receiving photographic data for the areas in the control unit (400) and generating read data by machine vision image processing to detect defects and
And a selection step (S60) of transferring ferrite to the sorting unit (500) and selectively discharging ferrite according to the read data,
In the first appearance inspection step S20,
Placing the ferrite at a distance of 40 to 80 mm on the central part between the first left illumination part 121 and the first right illumination part 131,
A step of irradiating light in a left region Z1 and a right region Z2 of the ferrite in the first left illumination unit 121 and the first right illumination unit 131 with a luminance of 50 to 150 cd / m 2 and a wavelength region of 470 nm or less; ,
Photographing the left zone Z1 and the right zone Z2 of the ferrite with the resolution of 20 M pixels or more and the number of frames per second at 6 fps or more in the first left photographing unit 122 and the first right photographing unit 132,
And transferring the photographed data of the left area (Z1) and the right area (Z2) to the control unit (400). The method according to claim 1,
In the presetting step S10,
A control value for controlling the operation states of the illumination unit and the photographing unit of the first to third visual inspection units 100 to 300 is set,
Selecting a model of ferrite as an object to be inspected, and setting a read reference value for a defect area and dimensions generated in the model. delete The method according to claim 1,
The second visual inspection step (S30)
Placing the ferrite at a distance of 40 to 80 mm on the lower central portion of the second central illumination portion 221,
Irradiating the upper center region Z3 of the ferrite with light having a luminance of 50 to 150 cd / m < 2 > and a wavelength region of 470 nm or less in the second central illumination unit 221,
Photographing the upper central region Z3 of the ferrite with the resolution of 20 M pixels or more and the number of frames per second at 6 fps or more in the second central photographing unit 222,
Positioning the ferrite at a distance of 40 to 80 mm on the lower center portion in the diagonal direction between the second left illumination portion 231 and the second right illumination portion 241,
The upper left side zone Z4 and the upper right side zone Z5 of the ferrite in the second left illumination part 231 and the second right illumination part 241 are irradiated with light having a luminance of 50 to 150 cd / m 2 and a wavelength area of 470 nm or less Step,
Photographing the upper left zone Z4 and the upper right zone Z5 of the ferrite with the resolution of 20M pixels or more and the number of frames per second of 6 fps or more in the second left photographing part 232 and the second right photographing part 242; ,
And transferring the photographed data of the upper central zone (Z3), the upper left zone (Z4) and the upper right zone (Z5) to the control unit (400). The method according to claim 1,
The third visual inspection step (S40)
Placing the ferrite at a distance of 40 to 80 mm on the upper center portion in the oblique direction between the third left illumination portion 321 and the third right illumination portion 331,
The lower left side zone Z6 and the lower right side zone Z7 of the ferrite in the third left illumination section 321 and the third right illumination section 331 are irradiated with light having a luminance of 50 to 150 cd / m 2 and a wavelength range of 470 nm or less Step,
Photographing the lower left zone Z6 and the lower right zone Z7 of the ferrite with the resolution of 20 M pixels or more and the number of frames per second at 6 fps or more in the third left photographing part 322 and the third right photographing part 332; ,
Positioning the ferrite at a distance of 40-80 mm on the upper central portion of the third central illumination portion 341,
Irradiating light in a lower central region (Z8) of ferrite in the third central illumination unit (341) with a luminance of 50 to 150 cd / m < 2 > and a wavelength region of 470 nm or less;
Photographing the lower central region Z8 of the ferrite with a resolution of 20M pixels or more and a frame rate per second of 6 fps or more in the third central photographing unit 342,
And transferring the photographed data of the lower right zone (Z7), the lower left zone (Z6) and the lower central zone (Z8) to the control unit (400). The method according to claim 1,
The reading step (S50)
A step of image-processing the photographed data to emphasize the difference in light and darkness to remove the afterimage,
Extracting a blob and calculating a blob area to calculate read data for each kind of defect,
Comparing and analyzing the read data based on the preset data input in the presetting step S10 to generate a control signal for selection and transmitting the control signal to the selection unit 500,
And a step of learning the read data by applying a machine vision deep running technique to reference the defect detection.

Images