JP2889490B2 - Threshold setting method in image processing apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for setting a threshold value in an image processing apparatus used for inspecting the shape of a component using a CCD camera.

[0002]

2. Description of the Related Art Recently, various image processing apparatuses having an inspection function for recognizing the shape of various parts using a CCD camera and instantaneously determining the quality of the parts have been developed. As an example of this, there is a screw head shape inspection device for determining the quality of the screw head shape. This device is
It has two transport belts made of a transparent material arranged at a predetermined interval, and the heads of the screws are locked by the transport belts so that the screws are continuously supplied in a line. Further, a strobe and a CCD camera for receiving the light beam are arranged at positions sandwiching the movement path of the screw. When the screw is positioned in front of the CCD camera, the strobe operates and the head silhouette of the screw is moved to the CCD. Captured by the camera. At this time, the brightness value of the image obtained from the voltage value corresponding to the amount of light received by the pixels constituting the image of the CCD camera is compared with a predetermined threshold value, and the shape of the screw head is binarized. The binarized data is calculated as data, and the binarized data is sent to the pass / fail determination unit.
The pass / fail determination unit calculates a feature point of the head shape of the screw from the binarized data, compares the calculated feature point with a feature point of the head shape of the screw as a reference, and determines the pass / fail. It is configured.

[0003]

In the above apparatus, a CCD is used.
A brightness value is calculated for each pixel constituting the image of the camera, and the brightness value is compared with a predetermined threshold value to calculate binarized data. Is set constant over all pixels. However, when the CCD camera receives the illumination light from the strobe and captures the silhouette of the head of the screw, as shown in FIG. 5, a large amount of light goes around the periphery of the head of the screw 4 due to diffraction of the light. Because the shape of the head of is smaller than it actually is, it is necessary to increase the threshold as much as possible, but if this threshold is too large,
When the luminance value of the portion near the light amount is small, the binarized data of this portion is easily affected by noise. Even if a small amount of noise is added, the binarized data becomes part of the screw 4 regardless of the background portion. It is converted, and drawbacks such as the inability to accurately recognize the head shape of the screw 4 occur. Conversely, C
Even when the CD camera is arranged to receive the reflected light of the strobe light, when the head of the screw 4 has a curved surface, as shown in FIG. Since the amount of light reflected on the screw 4 decreases and the shape of the head of the screw 4 becomes smaller than the actual shape, it is necessary to reduce the threshold value as much as possible. For this reason, pixels in the bright portion of the background portion of the dark portion are easily affected by noise, and even if a small amount of noise is added, the pixel is converted into binary data as a part of the screw 4 regardless of the background portion, and the head of the screw is There are drawbacks such as the inability to accurately recognize the part shape.

SUMMARY OF THE INVENTION The present invention has been made for the purpose of eliminating the above-mentioned drawbacks, and provides a threshold value setting method for setting a threshold value for each pixel according to the brightness of a background image captured by a CCD camera. It is intended for.

Further, according to the present invention, when a CCD camera is arranged so as to receive irradiation light from a light source, the brightness value of a pixel in a peripheral area of each pixel is determined for each pixel based on the brightness value of each pixel constituting an image of the CCD camera. Calculates its own threshold value based on this value and sets a threshold value for each pixel according to the brightness of the background image, reducing the influence of shadows even in the background image. It is an object of the present invention to provide a threshold value setting method that can obtain binary data.

Further, even when the CCD camera is arranged so as to receive the reflected light of the light from the light source, when the head of the screw has a curved surface, the present invention is also applicable to each pixel constituting the image of the CCD camera. Calculate its own threshold based on the maximum luminance value of the pixels in the surrounding area for each pixel from the luminance value, and set the threshold value for each pixel according to the darkness of the background image in the dark area It is another object of the present invention to provide a threshold value setting method capable of obtaining binary data by reducing the influence of a difference in the amount of reflected light from a component.

In addition, the present invention provides a CRT displaying a grayscale image captured by a CCD camera on a CRT display unit.
It is an object of the present invention to provide a threshold value setting method for setting an arbitrary region on a display unit and arbitrarily setting a threshold value of a pixel in this region.

[0008]

Means for Solving the Problems A CCD camera is arranged on a moving path of a component receiving light from a light source, and
When positioned in front of the D camera, the voltage value of each pixel constituting the image of the CCD camera is compared with a preset threshold value, and the shape of the part is recognized as binary data. In the image processing apparatus for displaying the shape of the component on the CRT display unit based on the above, the voltage value of each pixel of the CCD camera is converted into the luminance value of the image and temporarily stored in the state where there is no component when the light source is activated, and each pixel is temporarily stored. Each time, a luminance value of a pixel in the surrounding area is obtained, and the threshold value is changed based on the luminance value.

When a CCD camera is arranged at a position for receiving light emitted from a light source, the minimum value of the luminance value of a pixel in a certain peripheral area is determined for each pixel, and the minimum value is determined based on the minimum value. May be set.

Further, when the CCD camera is disposed at a position where the reflected light of the light emitted from the light source is received, the maximum value of the pixels within a certain area is obtained for each pixel, and the maximum value of the pixel is determined based on the maximum value. You may comprise so that a threshold value may be set.

In addition, a CCD camera is arranged on the moving path of the component receiving the light from the light source, and when the component is positioned in front of the CCD camera, the voltage value of each pixel constituting the image of the CCD camera is set in advance. In the image processing apparatus which recognizes the shape of the component as binarized data by comparing the threshold value with the set threshold value and displays the shape of the component on the CRT display unit based on the binarized data, The voltage value of each pixel of the CCD camera in a state where there is no image is converted into a luminance value of an image, a grayscale image based on the luminance value is displayed on a CRT display unit, and an arbitrary area setting signal on the grayscale image is displayed. When receiving the threshold value, the threshold value of the pixel in the area may be set after the input of the desired threshold value.

[0012]

[Operation] In the image processing apparatus of the present invention, before starting work,
When the area or automatic threshold setting mode is selected,
When the light source is activated, the voltage value of each pixel constituting the image of the CCD camera in a state where there are no components is converted into a luminance value of the image and temporarily stored. Further, for each pixel, an appropriate value is selected from the luminance values of the peripheral pixels, and its own threshold value is calculated and stored based on the selected value.

When determining the threshold value of each pixel, C
When the CD camera receives the reflected light from the light source, it reads out the brightness value of the pixels in the fixed peripheral area for each pixel, and the CCD camera receives the light from the light source based on the maximum value. In this case, the threshold value of each pixel is calculated and stored by increasing or decreasing the offset value in consideration of the noise component based on the minimum value of the luminance value. This threshold value is set for all pixels, and all pixels can have threshold values according to the brightness or darkness of the surrounding pixels. In addition, even if the background image has a shadow or a reflective portion such as a belt, the threshold value of the pixel located in the shadow or the reflective portion can be smoothly changed to reduce the influence of the shadow or the reflective portion. .

Further, when determining the threshold value of the pixel, a grayscale image is displayed on a CRT display unit from the luminance value of each pixel, and an operator sets an arbitrary area, and sets the pixels within this area. The threshold value may be set arbitrarily.
In this case, an extra portion in the background image can be deleted.

After the threshold value setting for each pixel is completed, CC
The component is positioned in front of the D camera, and its shape is captured by the CCD camera, and the shape of the component is displayed on the CRT display unit as a binary image. At this time, since the threshold value for generating binarized data from the luminance value of the image of each pixel is set according to the brightness or darkness around the pixel, the peripheral edge of the component due to light diffraction Thinning or thinning of the part due to reflection of light when the head of the part has a curved surface is minimized. In addition, even if illumination unevenness occurs in all pixels, it can have almost constant noise resistance over all pixels, and not only does the deformation of the peripheral portion of the component due to light diffraction or light reflection amount decrease, but also the background portion Even if a noise component is added to the voltage value of the pixel constituting the pixel, a binary image closest to the actual shape of the component can be obtained without making the background part a part of the component.

[0016]

Embodiments of the present invention will be described below with reference to the drawings. In FIG. 1, reference numeral 1 denotes a screw head shape inspection device (hereinafter, referred to as an inspection device) as an example of an image processing device. The inspection device 1 includes two transparent materials arranged at predetermined intervals. It has a transport belt 2. The conveyor belt 2 is wound around a rotatably mounted pulley (not shown), and moves in the same direction at the same speed in response to the rotation of a rotary drive source (not shown). The screw 4 is continuously supplied in a line in a correct posture to the inspection position below the camera 3.

On the other hand, the inspection apparatus 1 has a strobe 5 as an example of a light source disposed at the inspection position and at a position sandwiching the moving path of the screw 4 and a CCD camera 3 for receiving irradiation light. Further, the inspection device 1 has a control device 6 that receives a voltage value corresponding to the amount of received light of each pixel constituting an image of the CCD camera 3 as a video signal. The control device 6 includes a control unit 7 composed of a microprocessor. The control unit 7 includes various kinds of features including a feature point of the head shape of the screw 4 which is a reference necessary for determining whether the head shape of the screw 4 is good. A storage unit 8 for storing information, cursor coordinate data, and information necessary for setting a threshold value such as a new threshold value is connected. These pieces of information are configured to be input from an input unit 9 described later, calculated by the control unit 7, and sent to the storage unit 8. Also, the control unit 7 detects, when the screw 4 has passed the inspection position, the component detection unit 10 that outputs the screw detection signal when the screw 4 has passed the inspection position, and receives light from the strobe 5a in response to the strobe drive flag via the interface bus. A strobe drive unit 5 to be irradiated, a mode command signal such as an area threshold setting mode, an automatic threshold setting mode or an inspection mode, a display mode command signal for commanding display of a grayscale image or a binary image, and other necessary information. An input unit 9 for inputting a pixel value, a CCD camera interface 11 for receiving the voltage value of each pixel of the CCD camera 3 to calculate binarized data, and specifying the luminance value and the binarized data of each pixel by the control unit 7. The frame RAM 12, the display control unit 13, and the CRT display unit 14 stored in any one of a luminance value area, a threshold area, and a binary area. It is.

The CCD camera interface 11 includes:
A timing pulse generator 11a that outputs a timing pulse when the image capture flag is set by the controller 7;
A / D for converting the voltage value of each pixel constituting the image of the CCD camera 3 into a luminance value each time a timing pulse is received
Conversion section 11b, threshold value set in fixed threshold value setting section 11c by selection signal from control section 7, frame R
A selection unit that sends one of the threshold values for each pixel stored in the AM 12 to a signal binarization unit 11d described later, compares the luminance value with the threshold value, calculates binary data, and calculates A signal binarizing unit 11d for outputting to a frame RAM 12 to be described later, an address generating unit 11 for generating an address each time a timing pulse is received and sending the address to the frame RAM 12
f. This CCD camera interface 1
When the control unit 7 receives an inspection mode command signal and sets an image capture flag, the control unit 7 converts a voltage value for each pixel into a luminance value in synchronization with a timing pulse, and a threshold value from the control unit 7 A threshold value corresponding to the value selection command signal is compared to calculate binarized data for each pixel, and output this to the image address created for the pixel in the binary area of the frame RAM 12 It is configured to be. At this time, since a video signal including the voltage value of each pixel is always transmitted from the CCD camera 3, the A / D conversion is started after waiting for the voltage value corresponding to the first pixel of the image, and the luminance value and the second It is configured to sequentially calculate the quantified data.

The CCD camera interface 1
When the control unit 7 receives the automatic threshold setting mode command signal and sets the image capture flag, similarly, the control unit 7 converts the brightness value for each pixel in synchronization with the timing pulse, and outputs the brightness value to the frame RAM 12. It is configured as follows.

Further, when the control unit 7 receives the area threshold setting mode command signal and sets the image capture flag, the CCD camera interface 11 similarly converts the image into a luminance value for each pixel in synchronization with the timing pulse. It is configured to output the luminance value of the lever to the frame RAM 12.

The frame RAM 12 receives a binary area set command signal from the control unit 7 in the inspection mode, and receives a luminance value area set command signal from the control unit 7 in the automatic threshold setting mode and the area threshold setting mode. hand,
The luminance value and the binarized data given from the CCD camera interface 11 and the threshold given from the control unit 7 are written in each area. The frame RAM 12 is configured to output an image capture completion signal to the control unit 7 upon completion of converting the voltage values of all pixels into luminance values or binary data, and to be directly accessible from the control unit 7. , Any new threshold value can be written to any threshold address in the threshold area. Further, the frame R
When the AM 12 receives the grayscale image display command signal or the binary image display command signal from the display control unit 13,
According to the command signal, the luminance value of the luminance value area or the binarized data of the binary area is output to the CRT display section 14 through the display control section 13. Moreover, the frame RAM 12 is a text RAM.
(Not shown), and receives a cursor pattern display command signal and stores the cursor coordinates at the CRT display unit 1
4 is configured to display a cursor pattern at a point above.

The display control unit 13 includes an input unit 9.
When a display command signal for a grayscale image or a binarized image output from the control unit 7 is received based on a display mode command signal from the CPU, any one of the luminance value and the binary data in the frame RAM 12 according to the content of the command is received. Is sent to the frame RAM 12. Also, CR
When the T display unit 14 receives any one of the luminance value and the binarized data from the frame RAM 12, the T display unit 14
It is configured to display a binarized image. Although not shown, the display control unit 13 displays a grayscale image having a brightness corresponding to the brightness value of the image captured by the CCD camera 3 and a binarized image obtained from the brightness value on a CRT display unit 14 in real time. Can also be displayed.

As shown in FIGS. 2A and 2B, the control section 7 1) waits for a mode command signal. 2) In the case of the inspection mode command signal, jump to 27). 3) Jump to 11) at the time of the area threshold setting mode command signal. 4) At the time of the automatic threshold setting mode command signal, the frame R
A luminance value area set command signal is output to the AM. Set the flash drive flag and image capture flag. 5) Wait for an image capture completion signal from the frame RAM. 6) Designate a pixel for which a threshold is to be set (pixels adjacent in the X coordinate direction and Y coordinate direction are sequentially called from the pixel at the end of the image of the CCD camera. Excluded.) 7) For a specified pixel, a luminance value of a pixel within a preset area width (number of pixels) is called from a luminance value area, and a value obtained by subtracting a predetermined offset value for noise from the minimum value is obtained. It is calculated as a new threshold value of the pixel. 8) The new threshold value is stored in a threshold address of a threshold area corresponding to the pixel. 9) It is determined whether the designated pixel is the last pixel. If this is not the last pixel, the process returns to 6). 10) When the designated pixel is the last pixel, return to 1). 11) Output a luminance value area setting command signal to the frame RAM, and set a flash drive flag and an image capture flag. 12) Wait for an image capture completion signal from the frame RAM. 13) Output a display command signal of a grayscale image based on the luminance value of the luminance value area to the display control unit. 14) Write the cursor coordinate data to the first cursor coordinate address in the memory section (the initial value is an arbitrary point). Write the cursor pattern into the text RAM. 15) Wait for input of cursor coordinate data from the input unit. 16) Rewrite the cursor coordinate data in the first cursor coordinate address with the new cursor coordinate data. Rewrite the cursor pattern into the text RAM. 17) Wait for an area setting start command signal from the input unit. If there is no such signal, return to 15). 18) Write the current cursor coordinate data to the second cursor coordinate address in the storage unit, and change the cursor pattern to text R
Write to AM. 19) Wait for input of cursor coordinate data. 20) Rewrite the cursor coordinate data in the second cursor coordinate address with the new cursor coordinate data. Rewrite the cursor pattern into the text RAM. 21) Wait for an area setting completion command signal from the input unit. If there is no such signal, return to 19). 22) Wait for the input of a new threshold from the input unit (initialize the threshold stored in the storage unit). 23) The new threshold is temporarily stored in the storage unit. 24) A threshold address corresponding to a pixel in a rectangular area having two points in the cursor coordinate address as diagonals is sequentially calculated, and the temporarily stored new threshold value is written into this. 25) It is determined whether or not there is an area threshold completion command signal. 26) Return to 1) when there is an area threshold completion command signal. 27) Wait for a screw detection signal from the component detector. 28) Output the binary area set command signal to the frame RAM, and set the image capture flag and the flash drive flag. 29) Wait for an image capture completion signal from the frame RAM. 30) Calculate the characteristic point of the silhouette of the head of the screw from the binarized data in the binary area, compare it with the reference value stored in the storage unit, and determine the quality of the head shape of the screw. , And outputs the determination result, and returns to 1). It is configured as follows.

In the above inspection apparatus, when the automatic threshold setting mode command signal or the area threshold setting mode command signal is output from the input unit 9 to the control unit 7 without the screw 4 prior to the start of the inspection, A luminance value area set command signal is output to the frame RAM 12. At the same time, a strobe drive flag and an image capture flag are set, and the strobe drive unit 5 activates the strobe 5 a to emit light toward the CCD camera 3. At this time, CCD
A luminance value of the image is calculated from an analog voltage value of each pixel constituting the image of the camera 3 in synchronization with the timing pulse,
This is output to the frame RAM 12 for each pixel. Further, a threshold address for each pixel is output from the address generation unit 11f to the frame RAM 12, and the luminance value is written in a luminance value address corresponding to each pixel.

When the automatic threshold setting command signal is output from the input unit 9, the luminance value of the pixels in the certain peripheral area is sequentially read out from the frame RAM 12 for each pixel, and a predetermined value is determined from the minimum value. Is subtracted from the frame RAM as a new threshold value of the pixel.
12 is stored in the threshold address. Each time the setting of the new threshold is completed, it is determined whether the pixel for which the new threshold is set is final or not. When this is the final pixel,
Complete the setting of the new threshold and wait for the next mode command signal. This new threshold value is set for all pixels of the image, and a threshold value is set for each pixel based on the minimum value of the surrounding brightness and having a fixed offset value for noise over all pixels. You. Therefore, as shown in FIG. 3, a minimum value indicated by a dashed line with respect to the luminance value of the background image captured by the CCD camera 3 is obtained,
Further, a new threshold value indicated by a thin line is set.

When the area threshold setting mode command signal is output from the input unit 9, the frame R
The luminance value in the luminance value area of the AM 12 is output to the display control unit 13, and a grayscale image based on the luminance value is output to the CR.
It is displayed on the T display unit 14. At the same time, the control unit 7 waits for the input of the cursor coordinate data, rewrites the cursor coordinate data of the first cursor coordinate address in the storage unit 8 with the new cursor coordinate data, and writes the cursor pattern in the text RAM in the frame RAM 12; The cursor is superimposed on the grayscale image on the CRT display unit 14 and displayed. In this state, an area setting start command signal from the input unit 9 is waited, the current cursor coordinate data is written to the second cursor coordinate address by this command signal, the cursor pattern is written to the text RAM in the frame RAM 12, and the area setting completion command is issued. Wait for the signal. When this region setting completion command signal is output, the input unit 9 waits for the input of a new threshold value, stores the input new threshold value in the storage unit 8, and
The threshold addresses of the pixels in the rectangular area having the point as the diagonal are sequentially calculated and rewritten with the new threshold. Therefore, CRT
An actual image is displayed on the display unit 14, an area is set on this image, and a threshold value of a pixel in the area can be set according to the brightness on the image. In addition to the above, it is possible to easily eliminate an image of a portion unnecessary for the head inspection of the screw 4. At this time, a binarized image based on the new threshold value may be displayed on the CRT display unit 14.

On the other hand, when the inspection mode command signal and the motor drive command signal are output from the input unit 9, the screws 4 pass by the conveyor belt 2 in front of the CCD camera 3.
When the screw 4 reaches the position in front of the CCD camera 3 and the sensor detects this, the strobe drive flag and the image capture flag are set, and light is emitted from the strobe 5a as in the threshold setting mode. At the same time, the silhouette of the head of the screw 4 is captured by the CCD camera 3,
The analog voltage is output to the CCD camera interface 11 for each pixel. A luminance value for each pixel is calculated from the analog voltage value, and the luminance value is calculated based on a fixed threshold value selected by the selection command signal from the control unit 7 to the selection unit 11e or a threshold value for each pixel. It is converted into binary data by comparison. Further, the binarized data is stored in an image address in the frame RAM 12 for each pixel, and based on the binarized data, the binarized image is stored in a CR.
It is displayed on the T display unit 14. At this time, as shown in FIG. 3, since the threshold value is set in accordance with the illumination unevenness of the background image, the threshold value is set to a value that is substantially constant, that is, offset from the luminance value of the background image. Binarized data can be obtained, and a binarized image obtained from the binarized data is an image with less thinning of the screw 4 caused by diffraction of light. Also, even if noise is added to the brightness value of the peripheral pixels of the image of the CCD camera 3 where the amount of light from the strobe 5a is insufficient, the background portion is converted to binary data as a part of the screw 4. There is no such thing as being displayed on the CRT display unit 14, and the binarized image can be displayed with constant noise resistance over the entire image.

The binarized image is displayed on a CRT display unit 14.
At the same time, the feature point of the silhouette of the head of the screw 4 is calculated from the binarized data in the frame RAM 12 and compared with a preset reference value to determine whether the head shape of the screw 4 is good or bad. Is done.

When the threshold value of each pixel is determined, as shown in FIG. 4, the maximum luminance value of the pixels in the surrounding area is selected, and an offset value for noise is added thereto. The new value may be used as the new threshold. In this case, it is suitable when the CCD camera captures the shape of the component by receiving the reflected light from the strobe, and if the component has a curved surface where the difference in the amount of reflected light is large, the shape is actually Can be displayed as a binarized image close to the shape of.

[0030]

As described above, according to the present invention, transmitted light,
When capturing the shape of a component with a CCD camera irrespective of reflected light, the threshold value of each pixel constituting the image of the CCD camera is changed and set according to the brightness of the background image when the component passes. Therefore, even if there is uneven illumination in the background image through which the parts pass, a threshold value can be set according to the brightness, and the noise resistance can be made substantially constant for all pixels. Further, the present invention is applicable to a case where a CCD camera receives transmitted light, and the shape of the component is recognized as being smaller than the actual shape due to diffraction of light, or when the CCD camera receives reflected light reflected on a curved surface portion of the component. Is recognized as smaller than the actual part due to the large difference in the amount of reflected light,
This can be made as close as possible to the actual shape without reducing the noise resistance. Further, the present invention is configured such that an arbitrary area is set on the CRT display unit and the threshold value of the pixel in the area is changed.
It is possible to easily remove an image of a part unnecessary for the inspection of the shape of the part.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a screw head shape inspection apparatus as an example of an image processing apparatus according to the present invention.

FIG. 2A is a flowchart illustrating an operation of a control unit according to the present invention.

FIG. 2B is a flowchart illustrating the operation of the control unit in the area threshold setting mode according to the present invention.

FIG. 3 is an explanatory diagram showing luminance values of one column of pixels constituting an image of the CCD camera according to the present invention.

FIG. 4 is an explanatory diagram showing luminance values of one column of pixels constituting an image of a CCD camera according to another embodiment of the present invention.

FIG. 5 is an explanatory diagram showing luminance values for one column of pixels forming an image of a CCD camera according to a conventional example.

FIG. 6 is an explanatory diagram showing luminance values for one column of pixels forming an image of a CCD camera according to another conventional example.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Inspection device of screw head shape 2 Conveyor belt 3 CCD camera 4 Screw 5 Strobe drive unit 5a Strobe 6 Control unit 7 Control unit 8 Storage unit 9 Input unit 10 Component detection unit 11 CCD camera interface 11a Timing pulse generation unit 11b A / D conversion section 11c fixed threshold value setting section 11d signal binarization section 11e selection section 11f address generation section 12 frame RAM 13 display control section 14 CRT display section

Continuation of front page (56) References JP-A-5-278205 (JP, A) JP-A-6-58722 (JP, A) JP-A-5-115005 (JP, A) JP-A-6-201609 (JP) , A) JP-A-7-30753 (JP, A) JP-A-7-43129 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01B 11/00-11/30 G06T 5/00

Claims (4)

(57) [Claims] 1. A CCD camera is arranged on a moving path of a component receiving light from a light source, and when the component is positioned in front of the CCD camera, a voltage value of each pixel constituting an image of the CCD camera is set in advance. In the image processing apparatus which compares the shape of the component as binarized data by comparing the threshold value with the threshold value and displays the shape of the component on the CRT display based on the binarized data, Convert the voltage value of each pixel of the CCD camera to an image brightness value and temporarily store it in the absence state, and change its own threshold for each pixel according to the brightness value of the pixels in the surrounding area A threshold setting method characterized by the following. 2. C is located at a position for receiving light emitted from a light source.
2. The threshold value setting according to claim 1, wherein the CD camera is arranged, and the threshold value is set for each pixel based on the minimum value of the luminance value of the pixel in a certain peripheral area. Method. 3. A method in which a CCD camera is arranged at a position where reflected light of light emitted from a light source is received, and its own threshold value is set for each pixel based on the maximum value of pixels within a certain area. The threshold setting method according to claim 1, wherein 4. A CCD camera is arranged with respect to a moving path of a component receiving light from a light source, and when the component is positioned in front of the CCD camera, a voltage value of each pixel constituting an image of the CCD camera is set in advance. In the image processing apparatus which compares the shape of the component as binarized data by comparing the threshold value with the threshold value and displays the shape of the component on the CRT display based on the binarized data, The voltage value of each pixel of the CCD camera in the absence state is converted into a brightness value of an image, a grayscale image based on the brightness value is displayed on a CRT display unit, and an arbitrary area setting signal on the grayscale image is received. And setting a threshold value of a pixel in the area after inputting a desired value.