JP2002296144A - Method and apparatus for detecting mark - Google Patents

Abstract

PROBLEM TO BE SOLVED: To relate to detection of a mark formed on a transparent body and to provide a method and an apparatus for detecting the mark which automatically carry out the detection of the mark traditionally relied on human eyes. SOLUTION: The transparent body is illuminated by an illuminating device that has a continuous bright and dark pattern at the projection plane. An image of the transparent body is obtained at the side opposite to the illuminating device where the transparent body is sandwiched. Then, the mark formed on the transparent body is detected from the image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for detecting a mark on a transparent body used for positioning a transparent body, and more particularly to a method for detecting a mark used for positioning a spectacle lens or a mold thereof. And an apparatus.

[0002]

2. Description of the Related Art A spectacle lens and a molding die thereof are provided with a mark used for identifying a product and a mark used as a standard for processing in a manufacturing process. For example, in the case of a plastic spectacle lens, a mark previously attached to a molding die is transferred to the spectacle lens surface in the molding process, or a method of engraving the spectacle lens surface in a process after molding, or A mark is formed by using a method of forming a mark by generating a minute crack by a laser in the inside. The mark thus added is used as a reference for processing and the like after the molding process. Also,
Even when the eyeglasses are framed into eyeglass frames to form eyeglasses, there are marks left on the surface and inside of the eyeglass lenses, and such marks are used for product identification in stores.

[0003] Conventionally, when a mark is used as a reference for processing or the like, the mark is detected by human eyes, and the eyeglass lens is aligned based on the mark. In addition, since the mark for product identification described above is preferably arranged at a predetermined position on the spectacle lens, various processes are performed by determining the position of the mark in consideration of the final finish.

[0004]

However, marks formed on a transparent body such as a spectacle lens, which are transferred in a molding process, stamped after molding, or formed inside by a laser, are colorless and transparent. It is also difficult to detect automatically because the detection is difficult. For this reason, it is necessary to detect the mark with human eyes before processing and perform manual alignment, or attach another easily detectable mark to human eyes so that the alignment can be easily performed. In some cases, the image captured by the camera is displayed on the monitor, and the position of the mark on the image displayed on the monitor is confirmed while manually aligning the image.However, it is only an auxiliary method that helps the human eyes. Automatic detection was difficult.

[0005] That is, prior to many processing and work,
Preliminary preparation work for positioning and assisting the positioning is required, which has taken a great deal of time and effort. In addition, since it relies on human eyes, the accuracy of alignment is influenced by the operator, and skill is required to keep the accuracy constant.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned circumstances, and automatically detects a mark that has been manually relied on, thereby eliminating the need for a skilled worker, and has a constant detection accuracy and a constant alignment. It is an object of the present invention to provide a mark detection method and device capable of ensuring accuracy.

[0007]

Means for Solving the Problems The present inventor has made intensive efforts to achieve the above-mentioned object, and as a result, irradiates the spectacle lens with an illuminating device having a continuous light and dark area on the irradiation surface. It has been found that the colorless and transparent mark attached to the mark can be visualized, and the mark can be detected by capturing an image of the spectacle lens at a position facing the eyeglass lens. This focuses on a process of detecting a mark formed by minute depressions, minute projections, and minute cracks with human eyes.
When detecting a mark attached to a spectacle lens by human eyes, the mark is easily detected when the spectacle lens is held in a bright direction and attention is paid to the vicinity of a boundary between a bright portion and a dark portion near the bright portion. Utilizing such a phenomenon enables mark detection.

Further, by inserting an adjusting lens between the illumination device and / or the imaging device and the spectacle lens according to the refractive power of the spectacle lens, the influence of the refraction of the spectacle lens is suppressed, and the image is taken in by the imaging device. In such images, a sufficient illumination area can be secured. When the adjusting lens is inserted between the imaging device and the spectacle lens, the image of the mark attached to the spectacle lens is affected by the refraction of the adjusting lens. When using the adjusting lens, it is preferable to insert the adjusting lens between the illumination device and the spectacle lens.

In an image captured by an image pickup device, a mark has an outline formed of pixels including light and dark,
Since the brightness difference is also close to the surrounding pixels, it may be difficult to stably separate the image from other elements (such as noise) in the image. Therefore, the captured image is subjected to a first differentiation process for enhancing the outline of the image and a second differentiation process for extracting the outline of the image, so that the outline of the mark is enhanced and the surrounding image is compared with the surrounding image. Makes it possible to extract a mark as an aggregate of different bright pixels.

[0010] Thereafter, the image is binarized by a predetermined threshold value, and a blob having a characteristic closest to the mark to be detected in the image is extracted so that the mark can be specified and detected. Hereinafter, an aggregate composed of pixels having a predetermined brightness is referred to as a blob.

Further, in order to stably detect the mark,
After the first differentiation process and / or the second differentiation process, a smoothing process is performed to remove noise from the image and to avoid blob division or omission.

[0012] Before or after the binarization processing, a pixel having a predetermined brightness in an image is expanded to avoid loss or division of the blob.

Then, by masking a predetermined area of the image, unnecessary noise is removed, and the detection of the mark is stabilized.

The present invention is preferably applied to a spectacle lens or a mold thereof. However, a mark formed by minute dents, minute projections, and minute cracks is a transparent body having a surface or an interior. If there is, it can be widely used.

Therefore, in the mark detecting method and the mark detecting apparatus according to the present invention, the invention of claim 1 is a method for detecting a mark attached to a transparent body, wherein the illumination surface has a continuous light and dark pattern. A mark detection method characterized by irradiating a transparent body with a device, capturing an image of the transparent body on the side facing the lighting device with the transparent body interposed, and detecting a mark attached to the transparent body from the image. provide.

According to a second aspect of the present invention, there is provided the mark detecting method according to the first aspect, wherein the transparent body is a spectacle lens or a mold thereof.

According to a third aspect of the present invention, in the mark detection method according to the second aspect, an adjusting lens is provided between the illumination device and / or the imaging device and the transparent body according to the refractive power of the spectacle lens or its mold. And adjusting the illumination area on the image captured by the imaging device.

According to a fourth aspect of the present invention, in the mark detecting method according to any one of the first to third aspects, in the step of detecting the mark from the captured image, the first method for enhancing the outline of the image is provided. Differentiation processing, second differentiation processing for extracting the outline of the image, binarization processing for binarizing the image with a predetermined threshold, and blob processing for extracting blob having predetermined characteristics from the image. And a mark detection method for performing a mark detection process for specifying a blob having characteristics closest to a mark to be detected as a mark.

According to a fifth aspect of the present invention, in the mark detection method according to the fourth aspect, a smoothing process for smoothing the image is performed after the first differentiation process and / or the second differentiation process. And a mark detection method characterized by the following.

According to a sixth aspect of the present invention, in the mark detection method according to the fourth or fifth aspect, an expansion process for expanding a pixel having a predetermined brightness in the image before or after the binarization process. And a mark detection method characterized by performing the following.

According to a seventh aspect of the present invention, in the mark detection method according to any one of the fourth to sixth aspects, a masking process for excluding a predetermined area of the image from detection is performed. The present invention provides a method for detecting a mark.

The invention according to claim 8 is an apparatus for detecting a mark on a transparent body, comprising: a holding means for holding the transparent body;
An illumination device that has a continuous light-dark pattern on the irradiation surface and illuminates the transparent body, an imaging device that captures an image of the transparent body on the side facing the illumination device with the transparent body interposed, and an imaging device that is connected to the imaging device. An image processing apparatus for detecting a mark from an image is provided.

According to a ninth aspect of the present invention, there is provided the mark detecting device according to the eighth aspect, wherein the transparent body is a spectacle lens or a mold thereof.

According to a tenth aspect of the present invention, in the mark detecting device according to the ninth aspect, the mark is inserted between the illuminating device and / or the imaging device and the transparent body in accordance with the refractive power of the spectacle lens or its mold. The present invention also provides a mark detection device including an adjusting lens for adjusting an illumination area on an image captured by an imaging device.

According to an eleventh aspect of the present invention, in the mark detecting device according to any one of the eighth to tenth aspects, the image processing device includes an image captured from the imaging device and an image obtained by performing image processing on the image. A storage device for storing, first differential processing means for enhancing the outline of the image, second differential processing means for extracting the outline of the image, and binary processing means for binarizing the image with a predetermined threshold value A blob processing unit for extracting a blob having a predetermined feature from an image, and a mark detection unit for identifying a blob having a feature closest to a mark to be detected from the extracted blob as a mark. The present invention provides an apparatus for detecting a mark.

According to a twelfth aspect of the present invention, there is provided the mark detecting apparatus according to the eleventh aspect, wherein the image processing apparatus includes a smoothing processing means for smoothing an image. I do.

The invention of claim 13 is the invention of claim 11 or 1
3. The mark detection device according to 2, wherein the image processing device includes an expansion processing unit that expands a pixel having a predetermined brightness in an image.

The invention of claim 14 is the invention of claims 11 to 13
The mark detection apparatus according to any one of the above, wherein the image processing apparatus includes a mask processing unit that excludes a predetermined area of an image from a detection target.

[0029]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings, taking the detection of hidden marks attached to spectacle lenses as an example. It is not limited to the form.

FIG. 1 is a diagram showing an example of a mark detection device according to the present invention. The lighting device 2 is arranged below the spectacle lens 1, and the imaging device 3 is arranged above the spectacle lens 1. In this example, the object-side surface of the spectacle lens 1 faces upward and faces the imaging device 3, but the object-side surface may face the illumination device 2. Of course, there is no problem even if the positional relationship between the illumination device 2 and the imaging device 3 is opposite. In FIG. 1, the spectacle lens 1 is placed on the holding unit 4, and the holding unit 4 is made of a material that transmits light emitted from the illumination device 2. Of course, even if it is not the form shown in FIG. 1, the irradiation light of the illumination device 2 can irradiate the spectacle lens 1 and the imaging device 4 can capture an image of a predetermined area of the spectacle lens 1. I just need.

The surface of the lighting device 2 has a continuous light and dark pattern as shown in FIG. When detecting the mark 7 attached to the spectacle lens 1 by human eyes, the spectacle lens 1 is held in a bright direction, and the mark 7 is detected when attention is paid to the vicinity of a boundary between a bright part and a dark part near the bright part. It's easy to do. Utilizing such a phenomenon, the illumination device 2 provided with a continuous pattern of light and dark illuminates the spectacle lens 1 so that the mark 7 attached to the spectacle lens 1 can be detected in the imaging device 3. . FIG. 2 shows a pattern in which white circles are continuously arranged on a black background, but the pattern is not limited to this. Further, the term “continuous” does not necessarily mean that predetermined patterns are arranged regularly, and there is no problem even if light and dark portions are irregular or the patterns have various shapes.

In FIG. 1, a spectacle lens 1 and a lighting device 2
An adjustment lens 5 for adjusting the illumination area is disposed between the two. When the spectacle lens 1 is a divergent lens, the illumination device 2 appears to be reduced in the image captured by the imaging device 3. In other words, at the same time as the light and dark patterns are reduced,
Since the surface of the spectacle lens 1 cannot be sufficiently irradiated, it is difficult to detect the mark 7. To avoid this, the adjusting lens 5 is inserted according to the refractive power of the spectacle lens 1. The adjustment lens 5 can be switched in and out depending on the refractive power of the spectacle lens 1, and can be automatically operated by obtaining refractive power information from a host computer or the like (not shown).

Image processing device 6 connected to imaging device 3
Has a function of receiving an image captured by the imaging device 3 and processing the image. FIG. 3 is a block diagram illustrating a hardware configuration of the image processing device 6. Image processing device 6
Is a CPU (Central Processing Unit) 61 that performs overall control
In addition, a display device 62 such as a CRT display or a liquid crystal display, a storage device 63 including a RAM, a ROM, and a magnetic disk, and an input / output device 64 connected to the imaging device 3 are provided. ing. Although not shown, a keyboard for inputting characters and numerals, a mouse for performing various operations with a cursor displayed on the screen of the display device 62, and an input / output device for connecting to a host computer or the like via a network are connected. Is possible.

FIG. 4 is a block diagram showing the internal structure of the storage device 63. The image storage unit 101 stores an image (hereinafter, referred to as an original image) fetched from the imaging device 3 and also stores an image after various image processing has been performed as necessary.

The first differential processing section 102 has a function of enhancing the outline of the image, and can enhance the unclear mark 7 in the original image. In this contour enhancement process,
A technique called a convolution operation (weighted average) is used. The convolution operation means that, for example, luminance data of each 3 × 3 pixel is set to A to I as shown in FIG. The weighting factors a to i are set as shown in (1) and luminance data of the pixel of interest E is obtained. Such calculation is performed for each pixel of the image to obtain a new image.

In the contour emphasizing process, for example, the luminance data E of the target pixel E is obtained by applying the weight coefficient shown in FIG.

[0037]

(Equation 1) The second differentiation processing unit 103 performs a differentiation process on the image processed by the first differentiation processing unit 102 to extract a contour. Original image and first differential processing unit 102
In the image processed by, the outline of the mark 7 is formed of a light and dark outline, and even if the binarization processing is performed in this state,
Part of the outline of the mark 7 is missing. Therefore, for example, a convolution operation is performed using two types of weighting coefficients shown in FIGS. 8B and 8C to extract the mark 7 in the image as an aggregate of bright pixels. The weighting coefficients of FIGS. 8B and 8C are applied to Equation 1 to obtain the absolute values of the two luminance data obtained respectively, and to calculate the average value of the absolute values of the final target pixel E
Luminance data E.

The binarization processing section 104 binarizes a gray image composed of several gradations of light and shade into black and white at a predetermined threshold value. Since the threshold value varies depending on conditions such as the illuminance of the illumination device 2 and the opening and closing of a diaphragm included in the imaging device 3, it cannot be said unconditionally. In this example, spectacle lenses 1 having various refractive powers are set in advance, and a threshold value suitable for a detection target is set.

The blob processing unit 105 has a function of extracting the blob, and can examine features such as the area, the position of the center of gravity, and the shape (whether round or square) of the blob. This makes it possible to separate unnecessary elements such as noise from the mark 7 to be detected.

Further, several processing units are provided to make mark detection more stable. Smoothing processing unit 10
Reference numeral 6 denotes smoothing to make the difference in brightness between adjacent images gentle, and is used when it is desired to suppress noise or to avoid blob division or omission before binarization processing. In the smoothing process, for example, a weighting coefficient as shown in FIG. 8D is used, a convolution operation is performed by applying the weighting coefficient to Expression 2, and luminance data E of the target pixel E is obtained. Note that there are various methods for the smoothing process, and the present invention is not limited to this. Further, in the present invention, the image processed by the first differential processing unit 102 and / or the second differential processing unit 103 is particularly subjected to a smoothing process, but may be used as needed.

[0041]

(Equation 2) The expansion processing unit 107 expands a pixel having a predetermined brightness, and is used to avoid division or omission of a blob.
The expansion process is a process mainly used for a binary image. When at least one pixel having luminance data of 1 exists near a target pixel, the luminance data of the target pixel is set to 1. The expansion processing also has a concept of 4 neighborhoods, 8 neighborhoods, and the like, and it is different how many pixels are to be considered centering on the target pixel. In this case, the expansion processing in the vicinity of 8 is preferable. Note that the expansion processing is not limited to the binary image.

The mask processing section 108 has a function of masking a predetermined area in an image so as to exclude the area from various image processing. Specifically, the area on the image to be masked is set to the luminance data 0, the luminance data for the other areas is set to the maximum luminance (for example, 255), and the image to be masked is ANDed. This makes it possible to set an area where the mark 7 does not exist clearly to the luminance level 0 and remove noise existing there.

The mark detecting unit 109 specifies a blob having a feature closest to the mark 7 to be detected from one or more blobs obtained by the blob processing. In this example, as the easiest identification method, a method of comparing a predetermined area of the detection target with an area of one or a plurality of blobs is adopted.

Next, the operation of the mark detecting device according to the present invention will be specifically described. FIG. 5 is an example of an operation flowchart of the mark detection device according to the present invention. First, in step S1, the spectacle lens 1 is set on the holding means 4. Next, in step S2, the refractive power of the set spectacle lens 1 is changed to a preset refractive power D
It is determined whether the distance is smaller than the predetermined value. If it is smaller, the adjusting lens 5 is inserted between the spectacle lens 1 and the illumination device 2 in step S3. The refractive power D is positive for a condenser lens, negative for a divergent lens, and zero for no refraction. The refractive power D may be set in consideration of the stability of detection, and the condition varies depending on the distance relationship between the imaging device 3, the spectacle lens 1, and the illuminating device 2.
In this example, it is -2.0D.

Next, in step S4, an image is fetched from the imaging device 3 and stored in the image storage unit 101 of the storage device 63 of the image processing device 6. Then, in step S5, the mask processing is performed on the fetched original image by the mask processing unit 108, so that a predetermined area in the image is excluded from processing in subsequent steps. For example, if there is a mark 7 having a size of 2 mm square on the circumference of 17 mm from the geometric center of the spectacle lens 1, the area within a radius of 14 mm from the geometric center and the radius 2
Since the mark 7 does not exist in an area outside the area within 0 mm, those areas can be masked.
Of course, there is no problem if this processing is omitted.

Subsequently, in step S6, a first differentiation process is performed by the first differentiation processing section 102 to emphasize the brightness of the entire image as shown in FIG. Thereafter, in step S7, a second differentiation process is performed by the second differentiation processing unit 103, and as shown in FIG. 6B, a portion where adjacent pixels change from bright to dark and dark to bright is brightened. The emphasis is made and the mark 7 in the image is extracted as an aggregate of bright pixels. At this time, in a portion other than the mark 7, only the mark 7 is extracted brightly because the amount of change in shading of the image is small. Further, in step S8, a smoothing process is performed by the smoothing processing unit 106, so that noise is prevented from being generated in the image after the binarization process in step S9, which is performed next, and at the same time, the division or omission of the blob is also suppressed. This smoothing process may be performed before or after step S7, only before step S7, or may be omitted.

Next, in step S9, as shown in FIG. 6C, the image is binarized to black and white by a predetermined threshold value by the binarization processing unit 104. In the present invention, a gray image of 256 gradations from black to white is assumed,
The present invention is not limited to this, and there is no problem with gray images of other gray scales, binary images, or color images. Thereafter, in step S10, expansion processing for expanding pixels of a predetermined brightness is performed by the expansion processing unit 107 to avoid division or loss of the mark 7 as shown in FIG. The expansion processing is preferably performed before or after the binarization processing in step S9, but may be omitted without any problem. Subsequently, in step S11,
The blob processing unit 105 performs a blob process for extracting a blob having a predetermined brightness. The blob processing can be performed on a gray image, but in the present invention, in order to improve the processing speed required for detecting the mark 7, the blob processing is performed on the image that has been subjected to the binarization processing in step S9. ing. If a sufficient processing speed can be secured, the binarization processing in step S9 may be omitted. In the blob processing, for example, a blob having a size close to the size of the mark 7 to be detected is extracted.

When the blob processing is completed, step S1
In 2, the mark 7 is detected by the mark detection unit 109. If there are multiple blobs that can be extracted,
The blob closest to the size and shape of the mark 7 to be detected is specified as the mark 7. If the position of the center of gravity of the specified mark 7 is obtained, it can be used for positioning the spectacle lens 1.

As described above, with the use of the mark detecting device of the present invention, the colorless and transparent mark 7 attached to the spectacle lens 1 can be automatically and accurately detected. In addition,
INDUSTRIAL APPLICABILITY The present invention can be widely used not only for spectacle lenses but also for molds and transparent bodies.

[0050]

As described above, according to the mark detection method and apparatus of the present invention, since the mark detection work conventionally relying on humans can be automated, the alignment of the transparent body can be performed accurately and automatically. As a result, it is possible to realize not only labor saving in the manufacturing process but also improvement in accuracy of positioning.

[0051]

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of a mark detection device according to the present invention.

FIG. 2 is a diagram illustrating an example of a pattern attached to a lighting device.

FIG. 3 is a block diagram illustrating a hardware configuration of the image processing apparatus.

FIG. 4 is a block diagram illustrating an internal structure of a storage device of the image processing apparatus.

FIG. 5 is a diagram showing an example of an operation flowchart of the mark detection device according to the present invention.

FIG. 6 is a diagram showing an example of each image in the image processing process;
(A) shows the image after the first differentiation processing, (b) shows the image after the second differentiation processing, (c) shows the image after the binarization processing, and (d)
FIG. 3 is a diagram showing images after dilation processing.

7A is a diagram illustrating luminance data of each pixel, and FIG. 7B is a diagram illustrating a weight coefficient corresponding to each pixel, for a 3 × 3 matrix centered on a target pixel E;

8A and 8B are diagrams illustrating an example of weighting factors used in a convolution operation according to the present invention, wherein FIG. 8A illustrates a weighting factor used in an image enhancement process, and FIG. 8B illustrates a first weighting factor used in an image extraction process. FIG. 3C is a diagram illustrating a weight coefficient, FIG. 3C is a diagram illustrating a second weight coefficient used in an image extraction process, and FIG. 3D is a diagram illustrating a weight coefficient used in a smoothing process.

[Explanation of symbols]

 1. Eyeglass lens 2. Lighting device 3. Imaging device 4. Holding means 5. Adjustable lens 6. Image processing device 7. Mark 61. CPU (central processing unit) 62. Display device 63. Storage device (RAM, ROM, magnetic disk, etc.) 64. Input / output device 101. Image storage unit 102. First differential processing section 103. Second differential processing unit 104.2 Binarization processing unit 105. Blob processing unit 106. Smoothing processing section 107. Expansion processing section 108. Mask processing section 109. Mark detector

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2G086 FF06 5B047 AA11 AB02 BA02 BC05 BC07 5C054 AA04 CA04 CC00 CD03 CE01 EA00 ED12 ED17 EJ04 FA04 FC05 FC14 FC15 GA01 GA04 GB11 HA01 5L096 AA03 AA06 BA08 BA02 FA02 EA02 EA12 FA69 GA03 GA12

Claims (14)

[Claims] 1. A method for detecting a mark on a transparent body, comprising irradiating the transparent body with an illuminating device having a continuous light-dark pattern on an irradiation surface, and illuminating the transparent body with the illuminating device facing the illuminating device across the transparent body. A method for detecting a mark, comprising taking an image of a transparent body and detecting a mark attached to the transparent body from the image. 2. The mark detecting method according to claim 1, wherein said transparent body is a spectacle lens or a mold thereof. 3. A method for detecting a mark according to claim 2, wherein an adjusting lens is inserted between the illuminating device and / or the image pickup device and the transparent body in accordance with the refractive power of the spectacle lens or its molding die, and the image is picked up. A mark detection method comprising adjusting an illumination area on an image captured by a device. 4. The mark detecting method according to claim 1, wherein in the process of detecting the mark from the captured image, a first differentiation process for enhancing a contour of the image is performed. A second differentiation process for extracting a contour of the image;
A binarization process for binarizing the image with a predetermined threshold, a blob process for extracting a blob having a predetermined feature from the image, and a blob having a feature closest to a mark to be detected as a mark And a mark detection process for detecting a mark. 5. The mark detection method according to claim 4, wherein a smoothing process for smoothing the image is performed after the first differentiation process and / or the second differentiation process. Detection method. 6. A method for detecting a mark according to claim 4, wherein expansion processing for expanding pixels having a predetermined brightness in the image is performed before or after the binarization processing. Detection method of the mark to be used. 7. The mark detection method according to claim 4, wherein a mask process is performed to exclude a predetermined area of the image from detection. . 8. An apparatus for detecting a mark on a transparent body, comprising: holding means for holding the transparent body; an illuminating device having a continuous light-dark pattern on an irradiation surface and irradiating the transparent body; A mark detection device comprising: an imaging device that captures an image of a transparent body on a side facing a lighting device with a body therebetween; and an image processing device that is connected to the imaging device and detects a mark from the image. 9. The mark detecting device according to claim 8, wherein the transparent body is a spectacle lens or a mold thereof. 10. The mark detecting device according to claim 9, wherein the spectacle lens or a molding die thereof has a refractive power according to:
A mark detection device comprising an illumination device and / or an adjustment lens inserted between an imaging device and a transparent body, the adjustment lens adjusting an illumination area on an image captured by the imaging device. 11. The mark detecting device according to claim 8, wherein the image processing device includes a storage device that stores an image captured from an imaging device and an image obtained by performing image processing on the image. First differentiation processing means for enhancing the contour of the image, second differentiation processing means for extracting the contour of the image, binarization processing means for binarizing the image with a predetermined threshold value, and A mark detection device comprising: a blob processing unit for extracting a blob having a feature; and a mark detection unit for specifying, as a mark, a blob having a feature closest to a mark to be detected from the extracted blob. . 12. A mark detecting device according to claim 11, wherein said image processing device comprises a smoothing processing means for smoothing an image. 13. A mark detection apparatus according to claim 11, wherein said image processing apparatus includes expansion processing means for expanding pixels having a predetermined brightness in an image. apparatus. 14. The mark detection device according to claim 11, wherein the image processing device comprises:
A mark detection apparatus comprising: a mask processing unit that excludes a predetermined area of an image from a detection target.