JPH07260704A - Equipment for visual inspection of electrode part of board - Google Patents

Abstract

PURPOSE:To execute visual inspection of an electrode part of a board accurately and in a short time without relying on hands. CONSTITUTION:An image of an electrode part of a board 7 being conveyed is picked up by cameras 11. An image processing device 31 takes in data on the image of the electrode part picked up, determines vertical and horizontal projection data thereof and determines an electrode to be inspected from the image data, on the basis of the projection data determined. Next, data on the image of each determined electrode to be inspected are binary-coded. Then, three inspection windows for the opposite end parts of the electrode and the central part thereof are set for the binary-coded data and the number of pixels in each inspection window is computed. The electrode is determined as defective when a difference between the image data adjacent vertically in the inspection window is found to be a reference value or above, as the result of computation, in the opposite end parts, of the electrode while it is determined as defective when the value of the area measured in the inspection window is found to be a reference of determination or above in the central part of the electrode.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a substrate electrode portion visual inspection apparatus.

[0002]

2. Description of the Related Art A (print) substrate used for game software, an extended memory board, etc. is known in which a plurality of electrodes are arranged in alignment at the end of the substrate surface. I on this board
Even when the C element or the like is arranged and the circuit is assembled and one game software or the like is completed, the electrode portion thereof is exposed. When using game software or the like, the electrode portion is inserted from an insertion opening provided in a computer or the like to be attached in a fitted state with the computer.

Such a substrate is usually manufactured in a state where an even number of substrates (six in the figure) are connected as shown in FIG. 19 and finally separated for use.

By the way, the electrode portion is peeled off, scratched, dented,
Various defects such as defective plating and lumps (unevenness due to mixing of foreign matters) occur. These defects not only cause the quality of the finished product to be poor, but as described above, they are exposed to the public in an exposed state, which is not aesthetically pleasing.

Therefore, conventionally, at the substrate shipping stage,
The inspector looked at each of the substrates and inspected the electrodes for defects.

[0006]

However, the visual inspection by the inspector as described above is not only inefficient, but also an accurate inspection cannot be performed. In addition, there are some parts that depend on the experience of the inspector, and there may be variations in the inspection results by the inspector.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a substrate electrode portion visual inspection apparatus capable of performing a visual inspection of a substrate electrode portion accurately and in a short time without human intervention. To do.

[0008]

In order to achieve the above object, the invention according to claim 1 is a substrate electrode part appearance inspection device for detecting a defect of an electrode part arranged at an end part of a substrate, An image pickup means for picking up an image of the electrode portion of the substrate being conveyed, and image data of the imaged electrode portion are fetched, vertical and horizontal projection data thereof are respectively obtained, and the image data is inspected based on the obtained projection data. An inspection target electrode determining means for determining the target electrode, and binarizing means for binarizing the image data of the determined inspection target electrode,
Pixel number calculation means for setting three inspection windows for both ends of the electrode and the center of the electrode for the binarized data and calculating the number of pixels in each inspection window, and the inspection windows for both ends of the electrode. The difference between the image data adjacent to each other in the upper and lower sides of the inside is determined, and when each difference is equal to or larger than the reference value, it is determined as defective, while the area in the inspection window is measured in the central portion of the electrode, and the area value is equal to or larger than the determination reference. It is characterized in that it is provided with a judging means for judging that the item is defective.

According to a second aspect of the present invention, in the board electrode portion visual inspection apparatus according to the first aspect, a plurality of electrode portions of the substrate are provided on the front surface and the back surface of the substrate and on the left and right with respect to the transport direction. The imaging means is formed, and a set consisting of a left front surface electrode inspection and a left rear surface electrode inspection,
It is characterized by comprising two sets of cameras, one for the right side front surface electrode inspection and the other for the right side back surface electrode inspection.

Further, the invention according to claim 3 is the same as claim 1.
Alternatively, in the board electrode part visual inspection apparatus according to the item 2, the inspection target electrode determining means determines the horizontal projection threshold value from the number of electrodes obtained from the horizontal projection data, the electrode width, and the average density obtained from the vertical projection data. Characterize.

Further, the invention according to claim 4 is the substrate electrode part appearance inspection device according to claim 1, 2 or 3, wherein the binarizing means is means for simply binarizing each electrode image data, It is characterized by comprising means for performing binarization after differentiating.

Further, the invention according to claim 5 is the substrate electrode part visual inspection apparatus according to claim 1, 2, 3 or 4, wherein the first electrode and the second electrode have different widths. And the inspection window is created by extracting from a plurality of different windows in which the first electrode and the second electrode are combined.

[0013]

In the above structure, the electrode portion of the substrate being conveyed is imaged, the image data of the imaged electrode portion is captured, the vertical and horizontal projection data thereof are respectively obtained, and based on the obtained projection data. An electrode to be inspected of the image data is determined. Next, the determined electrode image data to be inspected is binarized. Then, three inspection windows for both end portions of the electrode and the central portion of the electrode are set for this binarized data, and the number of pixels in each inspection window is calculated. As a result of the calculation, at both ends of the electrode, it is determined that the difference between vertically adjacent image data in the inspection window is equal to or larger than a reference value, while at the center of the electrode, the area in the inspection window is measured. If the area value is equal to or larger than the determination standard, it is determined as defective.

Since the cameras constituting the image pickup means are provided in pairs on the left and right sides with the substrate electrode portion interposed therebetween, all electrodes provided on the left and right sides and front and back sides of one substrate can be inspected by a single conveyance. , The inspection speed is increased.

Further, when determining the horizontal projection threshold for different electrode patterns when determining the electrodes to be inspected, the number of electrodes obtained from the horizontal projection data, the electrode width, and the average obtained from the vertical projection data. The calculation is performed based on the density and the accurate horizontal projection threshold value can be obtained, and the inspection accuracy is improved.

Further, in the binarization, both the binarization of the image data and the binarization of the image data after the binarization are carried out, and the state of the electrode defect. Since the defect determination is appropriately performed using either data, the defect determination is accurate and the reliability of the apparatus is improved.

Furthermore, even when there are two types of electrodes having different widths, it is possible to create an inspection window and perform inspection without being affected by the positions of the electrodes in the input image. it can.

[0018]

1 is a system block diagram showing the configuration of an embodiment of a substrate electrode section visual inspection apparatus according to the present invention, and FIG. 2 is an image processing apparatus which is a main part of the substrate electrode section visual inspection apparatus. 3 is a block diagram showing the functional configuration of FIG. The configuration of the substrate electrode portion to be inspected in the present embodiment is similar to that shown in the figure, and is intended for a pair provided on the left and right with respect to the traveling direction (inspection direction).

As shown in FIG. 1, the apparatus of this embodiment is roughly divided into a conveying / sorting apparatus 1, an image processing case 3, and an operation panel 5, and an operator operates the operation panel 5. Then, while the substrate to be inspected is transported, the image of the electrode part is captured, image processing is performed, and defects in the electrode part are determined,
Sort defective products.

The transfer / sorting apparatus 1 has a transfer line 9 on which a substrate 7 to be inspected is placed and transferred, and four cameras 11 _{-1 to} 11 arranged along the transfer line 9.
_-4 , a strobe illumination device 13 for providing strobe illumination light to each of the cameras 11 _{-1 to} 11 _-4 for carrying out image input while transporting the substrate 7, and a transport mechanism 15 for driving the transport line 9.
A sequencer 17 for supplying a control command to the transfer mechanism, a position detection unit 19 for supplying an image input trigger signal to the image processing apparatus in accordance with the transfer position of the substrate, a sorted good product substrate, and a defective product substrate. And a printer 21 for printing out the information of.

FIG. 3 shows the arrangement position of the camera.
(A) is a plan view and (b) is a side view. As shown in FIG. 3, the four cameras 11 _{-1 to} 11 _-4 described above are divided into groups of two, and the group of the cameras 11 _-1 and 11 _-3 sandwiches the substrate 7 between them. The electrodes arranged on the left and right sides of the substrate 7 are imaged respectively.
The set of cameras 11 _-2 and 11 _-4 is arranged to sandwich the substrate 7 and images the electrodes formed on the front and back sides of the right side of the substrate 7, respectively. Further, the cameras of each group are arranged apart from each other (350 mm, which is longer than the substrate length in this embodiment). In addition, the upper and lower cameras 11
The distance between _-1 and the camera 11 _-3 and the distance between the camera 11 _-2 and the camera 11 _-4 are about 100 mm apart in the front and rear. This is because if they are arranged on the same line, the strobe lights interfere with each other and good image data cannot be obtained. The image data captured by the upper and lower cameras are processed at high speed by different image processing devices, respectively, as described later.

As shown in FIG. 4, the board 7 is composed of the same small boards 8 connected to an even number of boards such as two, four, six, and eight boards. The small substrate 8 is used by being separated. Therefore, the substrate 7 is configured to be bilaterally symmetrical in the transport direction, and electrodes having the same shape pattern are aligned and arranged at both ends of the substrate 7. The electrode portion 10 is formed by gold plating. The total length A of the board is about 100 to 300 mm, and the total width B
Is about 70 to 200 mm and its thickness is 0.5 to 2.4 mm
It is a degree. Further, the size of the electrode portion 10 is the total electrode length a.
Is about 3 to 12 mm, the electrode spacing b is about 1 to 5 mm, and the electrode width c is about 0.5 to 3 mm.

In the present embodiment, the size and type of electrodes on one substrate 7 can be inspected up to two types, that is, a normal electrode and a ground electrode having an electrode width wider than the normal electrode.
Further, in the present embodiment, the field of view of the electrode having the total electrode length a of 7 mm or more is divided, that is, the inspection is performed twice. Defects to be inspected are scratches, spots,
There are dents and defective plating (including peeling).

On the other hand, the image processing housing 3 includes two image processing devices 31 and 33, and image monitor devices 35 and 37 connected to the image processing devices 31 and 33 to display respective image data and the like. And a host computer 39 that controls the image processing devices 31 and 33 and controls the printer.

FIG. 2 shows the configuration of the image processing apparatus. Since the image processing devices 31 and 33 have the same configuration, only the image processing device 31 is shown in FIG.

As shown in the figure, the image processing device 31 is
The CPU 41 is the control center, and the hard disk driver 4
3, the communication interface 45, the digital I / O 47, and the display control unit 49 are connected via the multibus 51.

Further, the image processing apparatus 31 performs, as an image processing system, an input unit 53, a correction unit 55 for correcting image data such as shading correction, and averaging of image data. An averaging unit 57 for removing noise components, a differential binarizing unit 59, a simple binarizing unit 61, and a sector density adding unit 63 are provided.

The input section 53 inputs the image data of the camera 11 _-1 or the camera 11 _-2 and executes a calculation of projection data as will be described later to specify the electrode section data to be inspected. Run.

Further, the sector density addition unit 63 inputs the output data of the differential binarization unit 59 and the simple binarization unit 61 separately, and the input binarized data is looked up in each look-up table ( Based on (inspection window), arithmetic processing is performed for each sector and the result is written in the result memory.

The host computer 39, as its main menu processing, as shown in FIG. 5, a board information creating function for setting and storing information such as dimensions of various boards and the number of electrodes and the position of image input timing, and image processing. Image processing adjustment function that creates and saves various windows and various inspection parameters required for image processing, image input adjustment function that performs image input adjustment such as focus, squeeze, and field of view range, various maintenance of image processing equipment, and output of product list It is a mode in which each processing is performed according to instructions from the operation panel, such as maintenance functions to be performed, product type and automatic inspection, and various processing functions such as sequencer transmission processing, image processing transmission processing, automatic inspection function for print output processing, etc. There is.

Next, the operation of this embodiment will be systematically described.

First, the outline of the processing in the host computer performed prior to the execution of the appearance inspection will be described with reference to the flowchart of FIG.

First, board information is set in the host computer 39.

Next, the input timing of the board 7 (see FIG. 7) is created by the host computer 39. Then, various parameters required for the inspection are set from the host computer 39. Then, each camera 11 is adjusted to focus and squeeze suitable for the substrate 7. When changing the magnification, the visual field range is set. From the operation panel, type No. And press the setting button. The setting of inspection information for each product is completed by the series of processes described above.

Next, in the upper-level computer 39, the product type No. set in advance as described above is set. Is read and the board information and parameters suitable for the product type are transmitted to the conveying / sorting apparatus 1 and the image processing apparatuses 31 and 33 to be automatically set. Host computer 39, image processing devices 31, 33, transport / sorting device 1
When all are prepared for inspection, the substrate 7 to be inspected is set in the supply stocker (not shown).

When the operator presses the operation button on the operation panel 5,
The automatic inspection is started, and good / defective determination, substrate distribution, and determination result transmission are executed. If there is a defective board, the defective position is printed out from the printer 21. The automatic inspection process is repeatedly executed until the stop button is pressed.

In the automatic inspection, the transport / sorting apparatus 1 sets the substrate 7 to the camera 11 based on the substrate information and the image input timing inputted by the operator in the host computer 39.
The sheet is conveyed downward and a control signal is output to the image processing apparatus 31.33. In the image processing device 31.33, the timing signal supplied from the position detection unit 19 (see FIG. 7)
Based on the image input, image processing and determination processing, the quality of the substrate 7 is determined and the sorting unit of the transport / sorting apparatus 1 sorts the substrate 7 into good products and defective products. Data on the number of defective electrodes for each of the small boards 8 to be formed is printed out.

Next, the process from the image input process to the determination process will be described in detail.

<Image Input Processing> The substrate 7 placed on the substrate inspection stage is conveyed at a predetermined speed, and as shown in FIG. 7, the camera 11 _-1 and the camera 11 _-1 according to the image input timing during the conveyance. The image data of the front surface and the back surface of the left electrode part is obtained by _-3 and supplied to the image processing devices 31 and 33, respectively. Since the image is input while the substrate 7 is being transported, strobe light is used for illumination.

That is, as shown in the time chart of FIG. 7, the image input triggers 1 and 3 have the same timing and 37
It is supplied every ms, and the image is input at that timing.
This corresponds to receiving a timing signal every time the substrate 7 advances by 6.48 mm from the substrate start. In addition,
In the uninspected range, it is possible to recognize only the timing position, do not perform image input, and prevent strobe light emission.

Next, the input unit 53 of the image processing apparatus 31 obtains the horizontal projection and vertical projection of the input image data. Then, as shown in FIG. 8, the edges of the horizontal projection data are examined to recognize the electrodes that are completely input within one visual field. In particular, it is important for the direction of travel that the input is complete.

When there are both edges, the edge to edge is identified, the electrode position, the electrode width, the electrode type, the number of electrodes, etc. are read, the LUT (look-up table) described below is selected, and the shift for each LUT is made. Set the quantity and the transfer information to the judgment routine. If there is no edge, it is determined that the input is for the electrodeless portion, and no defect determination is made.

Next, the coordinates of the next input at which the frontmost position of the electrode inspected this time will be calculated in consideration of an error (for example, an input error or a strobe light emission prohibited section).

○○○ _pix ± ○○ _{pix As} shown in FIG. 9, the nearest electrode front part within the allowable range is the last final input.

FIG. 10 collectively shows the above-described image input timing processing. Trigger timing signal (1)
Is input, one field of view of the inspection screen of the electrode unit 10 is 9.
Since it is 73 mm, the electrodes can be inspected,
Part of the edge of the electrode is out of the visual field. When the substrate 7 is conveyed by 6.48 mm and the next trigger signal is supplied, the electrode image data of one visual field corresponding thereto is fetched. At this time, the electrodes that have already been inspected on the previous screen (which can be found by calculating the amount of movement) are also present in the field of view, so this electrode is excluded from the image of the electrode. The image of the electrode is partially out of the visual field.

In this way, the timing is adjusted so that the electrode image data to be inspected will not be inspected in duplicate in each visual field.

[Method of Determining Projection Threshold] Next, the method of determining the projection threshold will be described separately for the case of vertical projection and the case of horizontal projection.

1. Vertical projection The threshold for vertical projection is determined for each shot by the following method.

The electrode position (upper electrode, lower electrode) is recognized.

As shown in FIG. 11, the vertical projection MAX value is read (excluding the uninspected range).

From the information of what percentage of the projected MAX value set in the memory of the image processing device is used as the threshold value,
Determine the projection threshold.

It is checked whether the calculated projection threshold value is equal to or larger than the lower limit value of the projection set in the memory of the image processing apparatus.

If the calculated threshold value is equal to or higher than the lower limit of projection, the projection data is searched and the electrode width, position and number are analyzed. At the same time, the (temporary) average concentration is calculated from the resolution and the electrode length. This (temporary) average density is used when determining the horizontal projection threshold.

Inspection and counting are started from the electrode next to the last electrode which was effective in the previous shot.

The inspection window is selected while paying attention to the combination (presence or absence of the ground electrode and the position of the normal electrode), pitch deviation such as groove and cut in the electrode portion, and the displacement amount of each look-up table (LUT) is calculated. And output the value.

2. Horizontal projection Horizontal projection processing is performed only when the electrodes are normally recognized by vertical projection.

(1) Determination of Horizontal Projection Threshold Two types of electrodes, the normal electrode and the ground electrode, are to be inspected. Therefore, as shown in FIG. 12A, in the horizontal projection of the image in which the two types of electrodes are aligned, the threshold value can be easily set. However, as shown in FIG. 12B, when there is an electrode other than these two types (this electrode is not an inspection target) and a gold-plated portion (for example, an electronic component such as a resistance element) other than the electrode portion is present. The threshold value is difficult to determine.

Therefore, from the information of the horizontal projection, the threshold value is obtained by the following equation from the number of electrodes, the electrode width, and the temporary average density obtained by the vertical projection.

Horizontal projection threshold = temporary average density × (electrode width 1 (pix) + ... + electrode width n (pix)) / X where X is a parameter stored in the memory of the image processing apparatus. , Is variable.

Then, by searching from both ends of the projection data with the horizontal projection threshold value thus obtained, and obtaining the edge, an accurate result can be obtained.

The method of determining the horizontal projection threshold differs depending on the number of inspections. In the case of one-time inspection, both ends of the edge (upper end and lower end) are used, and if impossible, only the upper end is used.

Further, since the height of the electrode portion is high (7 mm or more in the present embodiment), it cannot be inspected once.
Inspect in divided times. In this case, the first time, as shown in FIG. 13, the upper end and the 480 line are used to align the inspection window with its center. Then, in the second time, as shown in the figure, the lower end and one line are used to align the inspection window with the center thereof.

<Determination Processing> When the inspection electrodes are specified as described above, the image data is subjected to shading correction and averaging processing by the correction section 55 and the averaging section 57 to remove noise components. After that, it is supplied to the differential binarization unit 59 and the simple binarization unit 61.

The differential binarization unit 59 binarizes the input image of the electrode portion and then binarizes it (the threshold value can be freely set), and the simple binarization unit 61 inputs the electrode portion. Binarize image data. Then, the sector density addition unit 6
In No. 3, the densities in the inspection window created by dividing each into a rectangle as shown in FIG. 14 are added. An inspection is carried out based on the result of the addition of the density, and if there are defects such as scratches, pits, dents, and plating defects taken in black from the electrode, it is determined as defective.

<Setting of Inspection Window> In the defect judgment, as shown in FIG. 14, the electrode ends have an area difference between adjacent upper and lower inspection windows, and if the difference is equal to or larger than the judgment reference value, it is judged as defective. To determine. In addition, the central portion of the electrode measures the area (number of pixels) in the window, and if the area value is equal to or larger than the criterion, it is determined to be defective.

Further, the electrode is composed of a ground electrode and a normal electrode.
As shown in FIG. 15, the structure of the inspection window, that is, the look-up table, is also 3 types depending on whether the ground electrode is on the right side, the normal electrode only, or the ground electrode is on the left side. The type is set.

Then, a look-up table of two trees is selected from the look-up tables of these three types, and an inspection window corresponding to the image data is created.

The inspection window is the look-up tables LUT0, LUT1 of the sector addition boards SWI ^# 0, SWI ^# 1, SWI ^# 2, SWI ^# 3 provided in the sector density addition unit 63 as shown in FIG. Is created by using. Each sector addition board SWI ^# 0, SWI ^# 1, SWI ^# 2
, SWI ^# 3 for differential binarization, SWI ^# 0, SWI ^# 1
, SWI ^# 2 and SWI ^# 3 are for simple binarization. In order to perform inspection using the same inspection window for differential binarization and simple binarization, differential binarization sector addition board SWI ^# 0 and simple binarization sector addition board SWI ^# 2
, The differential binarization sector addition board SWI ^# 1 and the simple binarization sector addition board SWI ^# 3 have the same lookup table. And sector addition board
SWI ^# 0 and SWI ^# 2 look-up table LUT0 has a window to the right of the ground electrode, and look-up table LUT1 has a normal electrode window.
The lookup table LUT0 of ^# 1 and SWI ^# 3 has a window on the left of the ground electrode, and the lookup table LUT1 has a window of the normal electrode. By combining these, an inspection window corresponding to the image data is created. .

For example, as shown in FIG.
When is the normal electrode and is the image data of the ground electrode, the lookup table LUT0 is selected from the differential binarization sector addition board SWI ^# 0, and the two right windows corresponding to the normal electrode and the ground electrode are selected. Is extracted. Next, the lookup table LUT1 is selected from the differential binarization sector addition board SWI ^# 1, and the two left windows corresponding to the normal electrodes of are extracted. Then, the windows extracted from the lookup table LUT0 and the lookup table LUT1 are combined to create an inspection window. Then, also in the case of simple binarization, an inspection window is created by the same method as differential binarization.

In this way, the inspection window is created and the inspection is performed. In the case of simple binarization, the inspection window is created by the same method as the differential binarization, but it is also possible to create the inspection window based on the extraction information of the differential binarization. In addition, when the polishing of the electrode is rough, in order to prevent it from being recognized as a flaw and being judged as a defect, differential binarization is performed when the electrode polishing direction is the electrode height direction and the electrode width direction. Either in the horizontal direction or the vertical direction depending on the case.

FIG. 16 shows the determination result of the defective electrode. As shown in FIG. 16A, the number of defective electrodes is detected in electrode units. Further, as shown in FIG. 16A, the number of defective electrodes is detected for each electrode even in the case of the double inspection. The detection result is output by the printer 21.

[0072]

As described above, according to the first aspect of the present invention, the appearance inspection of the substrate electrode portion can be performed accurately and at high speed without human intervention, and the inspection work efficiency is dramatically improved. To do.

Further, according to the second aspect of the invention, since the pair of cameras forming the image pickup means are provided on the left and right sides with the substrate electrode portion interposed therebetween, the cameras are provided on the left and right sides and front and back sides of one substrate. Moreover, all the electrodes can be inspected by one transportation, and the inspection speed can be increased.

Further, according to the third aspect of the invention, when determining the horizontal projection threshold of the electrodes having different electrode patterns when determining the electrodes to be inspected, the number of electrodes obtained from the horizontal projection data and Since the calculation is performed from the electrode width and the average density obtained from the vertical projection data, an accurate horizontal projection threshold value can be obtained, and the inspection accuracy is improved.

Further, according to the invention described in claim 4, the binarization includes both binarization of image data and binarization of image data after differentiation. The defect determination is performed by using either data as appropriate according to the defect state of the electrode, so that the defect determination can be performed more accurately and the reliability of the device is improved.

Further, according to the invention of claim 5, even when there are two kinds of electrodes having different widths, the inspection is performed without being influenced by the positions of the electrodes of the input image. You can create windows for inspection.

[Brief description of drawings]

FIG. 1 is a block diagram showing a system configuration of a substrate electrode part visual inspection apparatus according to the present invention.

FIG. 2 is a block diagram showing a configuration of an image processing apparatus which constitutes a main part of the substrate electrode section visual inspection apparatus shown in FIG.

FIG. 3 is an explanatory view showing the arrangement position of a camera.

FIG. 4 is an explanatory diagram showing a configuration of a substrate to be inspected.

FIG. 5 is an explanatory diagram showing main menu processing of a host computer.

FIG. 6 is a flowchart showing a procedure of processing executed by a host computer.

FIG. 7 is a time chart illustrating the operation of the embodiment.

FIG. 8 is an explanatory view showing an inspectable electrode within one visual field.

FIG. 9 is an explanatory diagram showing an inspectable electrode within one visual field.

FIG. 10 is an explanatory diagram showing electrodes that can be inspected within one visual field based on a timing signal.

FIG. 11 is an explanatory diagram showing how to determine a vertical projection threshold.

FIG. 12 is an explanatory diagram showing how to determine a horizontal projection threshold.

FIG. 13 is an explanatory diagram showing how to align the inspection windows during the double inspection.

FIG. 14 is an explanatory diagram showing setting of an inspection window.

FIG. 15 is an explanatory diagram showing the structure of a lookup table.

FIG. 16 is an explanatory diagram showing an electrode inspection result.

FIG. 17 is an explanatory diagram showing a main configuration of a sector density addition unit.

FIG. 18 is an explanatory diagram showing a process of extracting a window from a lookup table according to an electrode.

FIG. 19 is an explanatory diagram of a board to be inspected.

[Explanation of symbols]

1 Conveying / sorting device 3 Image processing case 5 Operation panel 7 Substrate 8 Small substrate 9 Conveying line 10 Electrode part 11 _-1 , 11 _-2 , 11 _-3 , 11 _-4 Camera 13 Strobe lighting device 15 Conveying mechanism 17 Sequencer 19 Position detection unit 21 Printer 31, 33 Image processing device 39 Host computer 53 Input unit 55 Correction unit 57 Averaging unit 59 Differential binarization unit 61 Simple binarization unit 63 Sector density addition unit

Continuation of the front page (51) Int.Cl. ⁶ Identification number Internal reference number for FI Technical location H05K 1/11 Z 7511-4E 3/00 Q (72) Inventor Kozo 66 Horikawa-cho, Saiwai-ku, Kawasaki-shi, Kanagawa No. 2 In Toshiba Engineering Co., Ltd. (72) Inventor Shinjiro Iwama 66 Horikawa-cho, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture No. 2 In Toshiba Engineering Co., Ltd. (72) Tomio Suzuki Horikawa-cho, Saiwai-ku, Kawasaki-shi, Kanagawa 66-2 Toshiba Engineering Co., Ltd.

Claims (5)

[Claims] 1. A substrate electrode part visual inspection apparatus for detecting a defect of an electrode part arranged on an end part of a substrate, comprising: an image pickup means for picking up an image of the electrode part of the substrate to be conveyed; and an imaged electrode part. Of the image data of the image data, the vertical and horizontal projection data thereof are respectively obtained, and the inspection target electrode determining means for determining the inspection target electrode of the image data based on the obtained projection data, and the electrode to be the determined inspection target Binarizing means for binarizing the image data of the above, and three inspection windows for both ends of the electrode and the central portion of the electrode are set for the binary data, and the number of pixels in each inspection window is calculated. The pixel number calculation means and the both ends of the electrode determine the difference between vertically adjacent image data in the inspection window, and if each difference is greater than or equal to a reference value, the difference is determined to be defective, while at the center of the electrode. The area in the inspection window the measured Te, the area value is the substrate electrode section appearance inspection apparatus characterized by comprising: a determining means for determining a failure of not more criteria more. 2. The board electrode part visual inspection apparatus according to claim 1, wherein a plurality of electrode parts of the board are formed on a front surface and a back surface of the board and on the left and right with respect to a transport direction. Is a pair of cameras for left surface electrode inspection and left rear surface electrode inspection, and a set for right surface electrode inspection and right rear surface electrode inspection, and a board electrode portion visual inspection apparatus. . 3. The substrate electrode part visual inspection apparatus according to claim 1, wherein the inspection target electrode determining means determines the number of electrodes obtained from the horizontal projection data, the electrode width, and the average density obtained from the vertical projection data. A substrate electrode part visual inspection apparatus characterized by obtaining a projection threshold value. 4. The substrate electrode part visual inspection apparatus according to claim 1, 2 or 3, wherein the binarizing means is a means for simply binarizing each electrode image data and a means for binarizing after being differentiated. A board electrode part visual inspection device comprising: 5. The substrate electrode part visual inspection device according to claim 1, 2, 3 or 4, wherein the electrode part includes a first electrode and a second electrode having different widths, and the inspection window is A substrate electrode part appearance inspection device, characterized in that the substrate electrode part appearance inspection device is created by extracting from a plurality of different windows in which the first electrode and the second electrode are combined.