KR20060133271A - The method and an apparatus for inspecting harness by edge-detection - Google Patents

Abstract

A method and an apparatus for inspecting a harness are provided to improve accuracy of an inspection result by applying an edge-detection scheme instead of pixel-based inspection tools. An apparatus for inspecting a harness includes an inspection device(1), a switch(2), a controller(3), and a display(4). The inspection device includes a fixing frame(12), a transfer unit(13), a harness imaging unit(14), a rotation cylinder(15), and a connector box(16). A vertical transfer guiding rail, which is fixed at one upper surface of a lathe(11), is formed at the front surface of the fixing frame. The transfer unit is moved along with the fixing frame. The harness imaging unit includes plural lighting devices at one end of the transfer unit and a CCD(Charge Coupled Device) camera. The rotation cylinder is rotated at the lower portion of the fixing frame according to a control signal from the controller. The connector box rotates along the rotation cylinder and includes a harness insertion groove, to which a harness(5) is inserted, at the front surface thereof.

Description

The method and an apparatus for inspecting harness by edge-detection}

1 is a perspective view showing an example of a harness;

Figure 2 is a perspective view showing the inspection device of the harness using the contour detection method of the present invention.

Figure 3 is a partial perspective view of the transfer device and the rotating cylinder of the present invention.

Figure 4 is a schematic diagram showing the inspection step of the present invention.

5 is a schematic diagram illustrating contour detection.

6 is a schematic diagram illustrating an indexed algorithm of the present invention.

7 is a comparison chart showing the color distribution according to the H (chromaticity) value.

<Brief description of the main parts of the drawing>

1. Inspection device 2. Switching part

3. Control part 4. Display part

5. Harness

11. Shelf 12. Fixed Frame

13 Transfer Unit 14. Harness Camera Unit

15. Rotating Cylinder 16. Connect Box

131. Vertical Transport Frame 132. Vertical Transport Guide Rail

133. Vertical feed control unit 134. Horizontal feed frame

135. Horizontal feed guide rail 136. Horizontal feed control device

137. Before and after transfer frame 138. Before and after transfer guide rail

139. Front and rear feed control device

CCD camera 142. Lighting equipment

161. Harness Insertion Groove

501. Housing 502. Wires

The present invention relates to an apparatus for inspecting a defect of a harness, and more particularly, to an inspection chamber and an apparatus for inspecting a defect of a harness using a contour detection method.

Harnesses that are produced today are an essential part of automobiles and electronics. In particular, since the vehicle is used for various purposes such as an engine system harness, a control system harness, and a battery system harness, a problem in the wiring of the harness may cause a fatal problem that may cause malfunction of each connection device. Therefore, harness manufacturers carry out a complete inspection to prevent failure of the assembled product due to the harness.

The harness inspection products currently produced are inspected based on the quality control regulations specified by each company, so that the quality above a certain level is guaranteed, but most of the inspections performed by each company are energization inspections, which connect the wires and connectors connected to the harness. Focus on.

Since the wiring color inspection of the harness inspection equipment that is currently developed is developed with a focus on the pixel inspection method, the pixel inspection method performs inspection after fixing individual wires by placing a jig for fixing wires to recognize and process accurate color information. There was a hassle.

In addition, for the inspection of the harness having a two-stage structure, it was inconvenient to manually unplug the harness and insert it upside down.

The present invention has been made to solve the above-mentioned problems, Sobel mask in the contour detection method by inspecting the defect of the harness using the contour detection method that can be convenient and accurate inspection than the pixel inspection method, Roberts, Lapalcian, and Canny masks have been proposed, but here, noise, processing speed, and contour detection masks are detected. Canny masks suitable for application were used.

It consists of an inspection device section for inspecting the failure of the harness, a switching section for operating the device of the inspection device section, a microcomputer for analyzing and calculating the input image data from the inspection device section and a power supply device for supplying power to the electronic device. In the failure detector of the harness consisting of a control unit and a display unit for recognizing the results from the control unit, photographing the upper surface of the harness inserted into the connect box with a CCD camera, and transmitting the photographed primary inspection results to the control unit Step, the connect box rotates 180 ° forward by the rotation of the rotating cylinder, photographing the lower surface of the harness with the CCD camera rotated, 180 ° reverse rotation of the rotated connect box and the secondary test results The step of transmitting to the control unit, comparing and calculating the test results by the image processing and analysis algorithm preset in the control unit System, to a method of the harness to transmit the comparison and calculation results from the control section to the display by using the edge detection method which comprises a step of recognizing the user and checking device.

Hereinafter, described in detail by the accompanying drawings of the present invention.

1 is a perspective view showing an example of a harness, the harness is composed of three or more wires or wires in one housing, and the harness inspection device according to the present invention shows a harness having more complicated six-wire wires composed of two stages. It was.

In general, the harness 5 is composed of a wire 502 for transmitting power or a signal, and a housing 501 serving as a connector, and the wiring sequence of the wire 502 in the housing 501 is a component of the device due to incorrect wiring. This is necessary because it prevents malfunction and can be easily explained to workers elsewhere through wire or network.

Fig. 2 is a trial showing the inspection apparatus of the harness using the contour detection method of the present invention, wherein the inspection apparatus unit 1 for detecting a failure of the harness 5 and the switching device for operating the apparatus of the inspection apparatus unit 1 are shown. A control unit (3) comprising a unit (2), a microcomputer for analyzing and calculating image data input from the inspection device unit (1), and a power supply unit for supplying power to the electronic device; and the result of the control unit (3). It is composed of a display unit (4) for recognizing the user.

And the inspection device unit 1 is fixed to the upper surface portion of the shelf 11 for fixing the device device is fixed frame 12 is composed of a vertical transfer guide rail 132 for guiding vertical movement up and down is configured in the front portion And, it is composed of a conveying device portion 13 for conveying movement along the fixed frame 12.

Meanwhile, as illustrated in FIG. 3, the transfer device 132 vertically moves along the vertical transfer guide rail 132 which is formed to guide the vertical transfer to the front portion of the fixed frame 12 which is vertically fixed. The vertical transport frame 131 is formed to exercise the movement of the vertical transport frame 131 and the fixed can be adjusted by the vertical transport control device 133 formed on the rear portion of the fixed frame 12. .

A horizontal transfer guide rail 135 is formed on the front portion of the vertical transfer frame 131 so as to be transported horizontally so that the horizontal transfer frame moves horizontally to the left and right along the horizontal transfer guide rail 135. 134 is formed and the adjustment of the horizontal transfer frame 134 can be adjusted by the horizontal transfer control device 136 is configured on one side of the upper surface of the horizontal transfer frame 134.

In addition, the front and rear transfer guide rail 138 is formed on the side of the horizontal transfer frame 134 to guide the transfer movement in the front and rear directions, and the front and back along the transfer rail (138) It is configured to adjust the front and rear feed frame 137, which performs the feed movement in the rear direction, by the front and back feed adjusting device 139 which is configured on one side of the side part.

The front and rear transfer frame 137, the front of the CCD camera 141 for photographing the harness (5) and the side of the CCD camera 141 a plurality of lights that can adjust the brightness or the amount of reflection when shooting The harness imaging device section 14 composed of the device 142 is configured.

In addition, a lower portion of the fixed frame 12 includes a rotating cylinder 15 that rotates in response to a signal from the controller 3 and a harness insertion groove 161 into which a harness 5 is inserted into a front portion of the rotating cylinder 15. Exposed connect box 16 is configured.

The harness 5 is inserted into the harness insertion groove 161 as shown by the arrow in FIG. 2, and then, when the start button configured in the switching unit 2 is pressed, the upper surface of the harness 5 inserted first is photographed and the harness 5 is inserted. After the shooting of the upper surface of the rotation cylinder (15) is rotated 180 ° forward, after 180 ° forward rotation the lower surface of the harness (5) is photographed and after the end of the rotation cylinder 15 is rotated 180 ° reverse It is configured to return to the original state.

By mounting the rotary cylinder 15 as described above can be inspected by automating the previous work that was inspecting the upper and lower surfaces by hand during the defect inspection of the harness (5) consisting of two stages.

FIG. 4 is a flowchart illustrating a working step of the present invention, in which the harness 5 inserted into the connect box 16 is photographed by the CCD camera 141, and the photographed primary inspection result is transferred to the control unit 3. Transmitting, the connect box 16 is rotated forward by 180 ° by the rotation of the rotation cylinder 15, photographing the lower surface of the harness 5 by the CCD camera 141 with the rotated connect box 16, 180 ° reverse rotation of the rotated connect box 16 and transmitting the photographed secondary inspection result to the control unit 3, comparing and calculating the inspection result by the image processing and analysis algorithm preset in the control unit 3 And transmitting the result of the comparison / operation in the control unit 3 to the display unit and recognizing the result to the user.

FIG. 5 is a schematic diagram illustrating contour detection. As shown in FIG. 5, a contour image is detected using a Canny mask, which is a contour detection algorithm, for an image separated into three primary colors of red, green, and blue in the original image. do.

The contours of the three colors detected are synthesized to find the wiring area of the harness.

The black area surrounded by white shown in FIG. 5 may be regarded as an area having the same color. Finally, color information of the black area is extracted and color recognition is performed to obtain connection information of the harness wiring.

FIG. 6 is a flowchart illustrating an indexed algorithm of the present invention, in which the extracted data is stored in the microcomputer of the controller.

The overall flow receives pixel information distributed in a specific color gamut obtained from the CCD camera 141 and stores RGB information and HLS values through conversion of the information.

Among the stored information, templates of each of 12 colors are composed of five pieces of information of R (red), G (green), B (blue), L (brightness), and S (saturation).

Here, the H (chromatic) information has a different property from the other five pieces of information, and the H (chromatic) value is displayed in a phase of 0 to 360 degrees.

0 and 360 degrees represent the same crimson, with yellow, green, cyan, blue and purple in the order of 60 degrees.

The chromaticity of any color allows us to know which of the six basic chromaticity members is closer to the source color of that color.

However, chromaticity may be different in the case of chromatic color, and in the case of pure achromatic color, H (chromatic) value may not be obtained. In this case, it is replaced with the value having the highest frequency in the program.

For example, if you zoom in on a point in the dark areas of an arbitrary pink and sky blue area, both have similar gray values. However, if we calculate the H value through the GLS conversion formula, the dark part of pink comes out 26 and the dark part of sky blue comes out 264.

Here, 26 is a red component close to gray because it is close to red between 0 degrees of red and 60 degrees of yellow.

In addition, the dark portion of the sky blue region has a chromaticity value of 264, which is present between 240 of blue and 300 of purple, and is closer to blue, and thus, a blue component close to gray.

These characteristics have color recognition characteristics that are robust to lighting conditions in index expressions in a natural color environment.

The remaining values except H (chromatic) are obtained from the CCD camera 141 by obtaining 100 pixel color values for each color among the areas representing a specific color, and then converting each pixel information into R (red color). Split into G (green) and B (blue) and save them as variables.

Next, convert it to GLS using RGB information and store that information as a variable. Using the information stored in this process, the information of R (red), G (green), B (blue), L (brightness), and S (saturation) exists in the range of values from 0 to 255.

Except for the H (chroma) value, the remaining data of R (red), G (green), B (blue), L (brightness), and S (saturation) for each color gamut are used. Create membership information by calculating membership.

for R(붉은색), G(녹색), B(파란색), L(명도), S(채도)

for R (red), G (green), B (blue), L (brightness), S (saturation)

Where μ is the fuzzy membership, n is the number of data belonging to that part, and N is the total number of samples.

On the other hand, in order to apply the concept of compensating for lost information while being digitized, it is difficult to simply regard fuzzy membership as a distribution of sample data. The outline of this distribution was then transformed into a fuzzy membership. The algorithm for this modification is as follows.

i = 2

for i <256

if μ _i > μ _{i -1} then

temp1 _i = μ _i

    else

temp1 _i = μ _{i -1}

    end if

_{if μ 256 - (i- 1)} > μ 256 - (i- 1) then

    else

temp2 _{256- ( i -1)} = μ _{256- ( i -1)}

    else

temp2 _{256- ( i -1)} = μ _{256- ( i -1)}

    end if

end for

μ = temp1 ∩ temp 2

Here, temp is a sample, and the final output generates membership information μ having a convex value.

FIG. 7 is a comparison chart showing color distribution according to H (chromatic) value, and based on the above-described theory, the distribution of chromaticity values is analyzed and the color series of the color information of each wiring is summarized.

The theorem calculates the chromaticity values of the color elements of the corresponding color gamut and presupposes only one chromaticity value having a value of 90% or more when one value exists over 90%.

The reason for this is to remove unnecessary information due to the influence of white noise that may occur in the information obtained from the imaging equipment and the information of the part number description part of the wiring itself.

As shown in FIG. 7, various H (chromatic) components appear in the achromatic color, but are classified according to the series of the corresponding colors in the achromatic color.

Using this property, you can get the advantage of not having to calculate the unnecessary comparison between each membership except achromatic color.

Harness defect inspection device according to the present invention is a tester for inspecting the defects of the harness using the contour detection method is superior to the conventional pixel method, the data accuracy is excellent and because the equipment installed to realize the technology is also simple We also expect the cost of equipment to be lowered.

Claims (5)

The inspection device unit 1 for detecting a defect of the harness, the switching unit 2 for operating the device of the inspection device unit 1, and the microcomputer and electronics for analyzing and calculating the input image data from the inspection device unit 1. In the inspection device for inspecting the failure of the harness consisting of a control unit 3 consisting of a power supply device for supplying power to the device, and a display unit 4 for recognizing the results from the control unit 3, The inspection device unit 1 is a fixed frame 12 is fixed to one side of the upper surface of the shelf 11 for fixing the device device is configured with a vertical transfer guide rail 132 vertically moved up and down in the front portion, and A conveying device part 13 for conveying movement along the fixed frame 12, a harness photographing device part 14 including a plurality of lighting devices and a photographing CCD camera at an end of the conveying device part 13, and Connect box having a harness insertion groove 161 in which the harness 5 is inserted at the front of the fixed frame 12 and integrally rotates with the rotary cylinder 15 that is rotated by the control signal of the controller 3. Harness inspection apparatus using the contour detection method characterized in that it consists of (16). In the method of inspecting the failure of the harness, Photographing the upper surface of the harness 5 inserted into the connect box 16 by the CCD camera 141, transmitting the photographed primary inspection result to the controller 3, by rotating the rotation cylinder 15. The connecting box 16 rotates forward by 180 °, photographs the lower surface of the harness 5 with the rotated connect box 16 by the CCD camera 141, rotates 180 ° reversely of the rotated connect box 16, and Transmitting the photographed secondary inspection result to the control unit 3, comparing and calculating the inspection result by the image processing and analysis algorithm preset in the control unit 3, and comparing and calculating the result by the control unit 3 Method of inspecting the harness using the contour detection method characterized in that the step of transmitting to the display unit (4) to recognize the user. The method of claim 1, The transfer device unit 13 includes a vertical transfer frame 131 for moving along a vertical transfer guide rail 132 configured at the front portion of the fixed frame 12, and a front portion of the vertical transfer frame 131. A horizontal transfer frame 134 horizontally moving left and right along a horizontal transfer guide rail 135 configured in a horizontal direction, and front and rear transfers formed in the front and rear directions at the side portions of the horizontal transfer frame 134. Harness inspection device using the contour detection method characterized in that consisting of the front and rear transfer frame 137 to move forward and backward along the guide rail (138). The method of claim 1, The rotation cylinder 15 is rotated by 180 ° forward rotation and reverse rotation inspection device of the harness using a contour detection method characterized in that it is possible to inspect both the upper and lower surfaces of the harness (5). The method of claim 2, In the step of comparing and calculating the inspection result by the image processing and analysis algorithm set in advance in the control unit (3), the analysis algorithm is a method of inspection of the harness using the contour detection method characterized in that the contour method algorithm is adopted.

Images