JP2022032933A - Defect inspection system, defect inspection method and defect inspection program of plate-like wood - Google Patents

Abstract

To enable a defect that is difficult to detect only by checking a color difference of a photographed image to be comparatively easily detected.SOLUTION: A defect inspection system of plate-like wood comprises: an illuminator 2 for reflection light which emits visible light for reflection light toward the front surface of a veneer 6; an illuminator 3 for transmission light which emits a near-infrared ray for transmission light toward the rear surface of the veneer 6; and an image processing device 1 which detects a defect of the veneer 6 by analyzing a photographed image generated by a line sensor camera 4. The system discriminates the defect of the veneer 6 with the combination of the concentration difference and shape of the infrared transmission light image based on the transmission light and the color of the visible light image based on the reflection light, and enables the defect that is difficult to detect only by checking the color difference of the visible light image to be comparatively easily detected by allowing the difference of the concentration difference between a defect portion and a normal portion to be reflected on the infrared transmission light image even for the defect with a small color difference with respect to the normal portion on the visible light image.SELECTED DRAWING: Figure 1

Description

The present invention relates to a defect inspection system for plate-shaped wood, a defect inspection method, and a defect inspection program, and is particularly used for a system for detecting defects existing in plate-shaped wood such as veneer or ground lumber cut from logs and the like. It is suitable.

Conventionally, a defect inspection device has been known in which a piece of wood such as a veneer is photographed by a photographing means and a defect portion due to discoloration of the wood surface is detected by utilizing the color distribution of the photographed image (for example, Patent Document 1). reference). In the defect inspection apparatus described in Patent Document 1, white visible light is emitted from the reflected light illumination installed on the front side of the single plate, and the reflected light is applied from the transmitted light illumination installed on the back side of the single plate. Irradiate visible light of a color different from the illumination (for example, green). Then, while detecting defects such as live nodes, dead nodes, and discolored parts due to mold from the photographed image based on the reflected light, defects such as bug holes and cracks in the single plate are detected from the photographed image based on the transmitted light. There is.

Although the defect is not inspected for the veneer, it is disclosed in Patent Documents 2 to 4, for example, that the defect is detected by using the transmitted light when the inspection object is irradiated with infrared rays.

Japanese Unexamined Patent Publication No. 2007-147442 Japanese Unexamined Patent Publication No. 2006-153633 Japanese Unexamined Patent Publication No. 2011-33449 Japanese Unexamined Patent Publication No. 2014-190797

However, like the defect inspection device described in Patent Document 1, there is a problem that there is a defect that is difficult to detect only by observing the color difference (color difference) of the image taken by irradiating visible light. was there. That is, it is difficult to detect defects such as a discolored portion that is lightly discolored due to a viable node, a dead node, a discoloring bacterium, etc., which has a small color difference from the plate color of the normal portion, from the captured image based on the reflected light. In addition, defects such as insect holes that do not penetrate along the irradiation direction of visible light from the front surface to the back surface of the veneer, insect holes that are blocked by feces, and cracks that are not open can be found in the captured image based on transmitted light. Difficult to detect.

The present invention has been made to solve such a problem, and an object of the present invention is to make it relatively easy to detect a defect that is difficult to detect only by looking at the color difference of a photographed image.

In order to solve the above-mentioned problems, the defect inspection system for plate-shaped wood according to the present invention includes visible light illumination that irradiates visible light for reflected light toward one surface of plate-shaped wood and one of plate-shaped wood. Invisible light illumination that irradiates invisible light for transmitted light toward the other surface opposite to the surface of the wood, an image pickup device that captures one surface of plate-shaped wood and generates an image, and an image pickup device. It is equipped with an image processing device that detects multiple types of defects in the plate-shaped wood by analyzing the captured image, and reflects the shade and shape of the image based on the invisible light transmitted through the plate-shaped wood and the plate-shaped wood. Multiple types of defects in plate wood are discriminated by combinations that include at least the color of the image based on visible light.

According to the present invention configured as described above, even if the defect has a small color difference from the plate color of the normal portion on the image taken by irradiating visible light, it is taken by irradiating invisible light. On the image, the difference in shade between the defective part and the normal part is projected on the captured image due to the difference between the amount of invisible light transmitted through the defective part and the amount of invisible light transmitted through the normal part. .. Therefore, it is possible to identify that there is a defect in the portion where there is a difference in shade. Further, it is possible to determine the type of possible defects from the combination of the shape of the defect portion identified by the captured image based on invisible light and the color of the corresponding portion on the captured image based on visible light. This makes it relatively easy to detect defects that are difficult to detect only by looking at the color difference of the captured image based on visible light.

It is a figure which shows the structural example of the defect inspection system of a plate-shaped wood by this embodiment. It is a block diagram which shows the functional structure example of the image processing apparatus by this Embodiment. It is a flowchart which shows the operation example of the defect inspection system of a plate-shaped wood by this embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a defect inspection system for plate-shaped wood according to the present embodiment. As shown in FIG. 1, the defect inspection system of the present embodiment is configured to include an image processing device 1, a reflected light illumination 2, a transmitted light illumination 3, and a line sensor camera 4, and is composed of belt conveyors 5a and 5b (hereinafter,). , Simply referred to as a belt conveyor 5), and detects a plurality of types of defects existing in the single plate 6 conveyed by the belt conveyor 5.

The reflected light illumination 2 includes visible light illumination within the scope of the claims, and irradiates the reflected light visible light toward one surface (surface) of the single plate 6. As this visible light illumination, for example, a white light source such as an LED is used. Hereinafter, the visible light emitted by the reflected light illumination 2 is referred to as "white visible light". The reflected light illumination 2 extends in a line shape in a direction orthogonal to the transport direction of the single plate 6 (width direction of the single plate 6), and irradiates the single plate 6 with white visible light in a band shape.

The transmitted light illumination 3 includes a second visible light illumination and a non-visible light illumination within the scope of the claims. The second visible light illumination is directed toward the other surface (back surface) opposite to one surface of the single plate 6, and is distinguished from the hue of the reflected light on the single plate 6 by the visible light illumination which is the reflected light illumination 2. Irradiate visible light for transmitted light with a color that is easy to use. As the second visible light illumination, for example, a blue light source such as an LED or a green light source is used. Hereinafter, the visible light emitted by the second visible light illumination of the transmitted light illumination 3 is referred to as "second visible light". The invisible light illumination irradiates the back surface of the veneer 6 with invisible light for transmitted light. As this invisible light illumination, for example, a near-infrared light source such as an LED is used. Hereinafter, the invisible light emitted by the invisible light illumination of the transmitted light illumination 3 is referred to as "near infrared rays". The transmitted light illumination 3 also extends in a line shape in a direction orthogonal to the transport direction of the single plate 6 (width direction of the single plate 6), and irradiates the single plate 6 with second visible light and near infrared rays in a band shape. do.

Here, as the wavelength band of the near infrared ray, an appropriate wavelength band capable of transmitting the veneer 6 is used in relation to the plate thickness of the veneer 6. Further, it is preferable to use an appropriate wavelength band in consideration of the sensitivity band of the line sensor camera 4. For example, when the thickness of the veneer 6 is about 6 mm, it is possible to use near infrared rays having a wavelength band of 750 to 1500 nm. However, it is not limited to this wavelength band.

The transmitted light illumination 3 is installed at a position facing the gap between the upstream belt conveyor 5a and the downstream belt conveyor 5b, and is not on the back surface of the single plate 6 from the back side of the belt conveyor 5 through this gap. Irradiate visible light (near infrared rays) and second visible light. On the other hand, the reflected light illumination 2 is installed at a position slightly upstream (or downstream) of the gap between the upstream belt conveyor 5a and the downstream belt conveyor 5b, and is near infrared rays and second visible. White visible light is obliquely irradiated from the front side of the belt conveyor 5 to the region of the front surface opposite to the region of the back surface of the single plate 6 to be irradiated with light.

The line sensor camera 4 corresponds to an image pickup device within the scope of claims, and the surface of the veneer 6 is photographed in color to generate an image. The line sensor camera 4 is installed at a position facing the transmitted light illumination 3 with a gap between the upstream belt conveyor 5a and the downstream belt conveyor 5b, and is in a direction orthogonal to the transport direction of the single plate 6 (single). It extends in the width direction of the plate 6), and the single plate 6 is photographed in a line shape. When the veneer 6 is being conveyed on the belt conveyor 5, the line sensor camera 4 is 1 at a predetermined sampling time from one end side to the other end side of the veneer 6 in the conveying direction. Images for lines are repeatedly generated, and they are sequentially output to the image processing device 1.

The line sensor camera 4 includes a light receiving element having sensitivity to the white visible light emitted by the reflected light illumination 2 and the second visible light emitted by the transmitted light illumination 3, and a near light emitting element irradiated by the transmitted light illumination 3. It is equipped with a light receiving element having sensitivity to infrared rays. The line sensor camera 4 transmits the reflected light of the white visible light single plate 6 illuminated by the reflected light illumination 2 and the near infrared rays and the second visible light single plate 6 irradiated by the transmitted light illumination 3. By receiving light and photoelectric conversion of these, a captured image for one line of the single plate 6 is generated.

Here, the line sensor camera 4 has an image based on the reflected light of white visible light and the transmitted light of the second visible light (hereinafter referred to as a visible light image) and an image based on the transmitted light of near infrared rays (hereinafter referred to as infrared transmission). (Called an optical image) and. When the veneer 6 has a hole penetrating substantially vertically from the back surface to the front surface, the second visible light passes through the through hole, and that portion is imaged as a part of the visible light image. On the other hand, near-infrared rays pass through the entire veneer 6 including the portion having and not having the through hole as described above, and are imaged as an infrared transmitted light image. At this time, a portion of the veneer 6 whose thickness, fiber density, fiber direction, etc. are different from those around the veneer 6 appears to be whitened or blackened.

In the following, among the visible light images, the image portion based on the transmitted light of the second visible light is referred to as a second visible light image, and the image portion based on the reflected light of white visible light is reflected in white in the other image portions. It is called an optical image. Strictly speaking, there is a possibility that white visible light is reflected in the direction of the line sensor camera 4 even in the portion through which the second visible light is transmitted, but for convenience of explanation, the portion other than the second visible light image (second visible). The image of the part where light is not transmitted) is called a white reflected light image.

The image processing device 1 combines (combines) captured images for a plurality of lines sequentially output from the line sensor camera 4 at predetermined sampling times, so that each of the visible light image and the infrared transmitted light image is a single plate. Generate the whole image of 6. Then, by analyzing the generated visible light image and infrared transmitted light image, a plurality of types of defects existing in the veneer 6 are detected.

FIG. 2 is a block diagram showing a functional configuration example of the image processing device 1. As shown in FIG. 2, the image processing apparatus 1 includes a line image acquisition unit 11, an overall image generation unit 12, and an image analysis unit 13 as functional configurations. Each of these functional blocks 11 to 13 can be configured by any of hardware, DSP (Digital Signal Processor), and software. For example, when configured by software, each of the above functional blocks 11 to 13 is configured to include a CPU, RAM, ROM, etc. of a computer, and a program stored in a recording medium such as RAM, ROM, hard disk, or semiconductor memory operates. It is realized by that.

The line image acquisition unit 11 sequentially acquires one line of captured images generated by the line sensor camera 4 at predetermined sampling times. The whole image generation unit 12 generates the whole image of the veneer 6 by synthesizing (combining) the captured images of a plurality of lines acquired by the line image acquisition unit 11. At this time, the overall image generation unit 12 generates a visible light image including a white reflected light image and a second visible light image, and an infrared transmitted light image, respectively.

The image analysis unit 13 detects a plurality of types of defects in the veneer 6 by analyzing the entire image of the veneer 6 generated by the whole image generation unit 12. Here, the image analysis unit 13 uses a combination of the color, shape and pattern of the white reflected light image, the color and shape of the second visible light image, and the shading and shape of the infrared transmitted light image to obtain the single plate 6. Identify multiple types of defects. The defects of the veneer 6 detected by the image analysis unit 13 are, for example, discolored parts caused by insect holes, live nodes, dead nodes, open cracks, non-open cracks, skins, tartars, lumbers, discoloring bacteria, and the like. , Blade clogging, etc. The discolored part caused by the discolored bacterium is a concept that broadly includes the part of the wood discolored by the discolored bacterium, the part to which the discolored bacterium adheres, and the part discolored due to the discolored bacterium. The discolored portion caused by the discoloring bacterium or the like is, for example, a portion of wood that has been discolored blue or black by blue mold or black mold caused by the blue discoloring bacterium. Hereinafter, the discolored part caused by the discolored bacteria or the like is simply referred to as a “discolored part”.

Among these various defects, the color and shape of visible light images (white reflected light image and second visible light image) for open cracks, skins, tarts, and blade clogging, even without an infrared transmitted light image. , It is possible to distinguish to some extent by the pattern.

The opening crack is a type of crack that penetrates from the back surface to the front surface of the veneer 6 in the thickness direction along the optical axis of the second visible light and is open. Since the second visible light irradiated by the transmitted light illumination 3 is transmitted through the aperture crack, the defective portion of the aperture crack (second visible light image) and the normal portion of the peripheral material thereof (white reflected light image). The color difference between the light and the light becomes large, and it is possible to detect the presence of a defect by the color difference. Further, it is possible to determine that the defect is a "crack" from the shape of the defective portion in which the color difference occurs.

Invagination is a defect that tends to have a dark brown or black tint. Therefore, in the white reflected light image based on the white visible light emitted by the reflected light illumination 2, if the color difference between the defective portion of the skin and the normal portion of the peripheral material becomes large, the defect is caused by the color difference. It is possible to detect its presence. Further, when the pattern pattern such as the grain of wood of the defective portion where the color difference occurs is characteristic of the skin, it is possible to determine that the defect is the “bark”.

The acupuncture points are defects in the holes where the acupuncture points are stored, and are often discolored to black. Therefore, in the white reflected light image, when the color difference between the defective portion of the tartar and the normal portion of the peripheral material is large, it is possible to detect the presence of the defect by the color difference. Further, if it penetrates in the thickness direction, the second visible light irradiated by the transmitted light illumination 3 is transmitted. At this time, if the color difference between the defective part (second visible light image) and the normal part of the peripheral material (white reflected light image) becomes large (when the periphery of the hole is not blackened), the color difference It is possible to detect the presence of defects. In addition, it is possible to determine from the color of the defective part where the color difference is occurring that the defect may be either a "yani pot", a "skin" or a blackish "knot". be.

The blade clogging is a defect in which the plate thickness varies as compared with the peripheral portion and the continuity of the wood grain pattern is interrupted. Therefore, in the white reflected light image based on the white visible light emitted by the reflected light illumination 2, if the continuity of the wood grain pattern is clearly interrupted, it can be determined that there is a defect of "clogging". It is possible.

For the above defects, the image analysis unit 13 is red based on the transmitted light of the near infrared rays emitted from the transmitted light illumination 3 for the insect holes, the raw nodes, the dead nodes, the non-opening cracks, the cover material, and the discolored part. A plurality of types of defects in the single plate 6 are discriminated by the combination of the shade and shape of the external transmitted light image and the color of the white reflected light image based on the reflected light of the white visible light emitted from the reflected light illumination 2.

A bug hole is a hole formed by insect damage. The insect hole may be a through hole penetrating in the thickness direction of the veneer 6 or a non-penetrating hole not penetrating in the thickness direction. In the case of a non-penetrating hole, since the second visible light from the transmitted light illumination 3 is not transmitted, the insect hole cannot be detected based on the second visible light image. Further, it is difficult to detect a bug hole only with a white reflected light image based on white visible light from the reflected light illumination 2 unless there is a remarkable difference in color from the peripheral material due to discoloration of the worm-eaten portion or the like. be. It is even more difficult to detect if the insect holes are blocked by insect droppings.

On the other hand, the near-infrared rays emitted by the transmitted light illumination 3 transmit from the back surface to the front surface of the veneer 6 including both the normal portion and the defective portion. Here, the wormholes that do not penetrate and the wormholes that are blocked by feces are white on the infrared transmitted light image due to the difference in the amount of near-infrared transmission from the normal part of the peripheral material. Easy to change. Therefore, the image analysis unit 13 can detect that there is a defect in the whitened portion in the infrared transmitted light image. Further, when the shape of the whitened portion is an irregular slotted hole shape or a slotted hole shape in the direction orthogonal to the fiber, it can be determined that the defect is a "worm hole". It is possible. When the shape of the whitened portion is close to a circle, the image analysis unit 13 determines whether it is a bug hole, a missing node, or a pinhole in consideration of the color tone and shape of the white reflected light image of the corresponding portion. Is possible. In addition, some insects have the property of causing mold around the damaged part. In the insect hole in this case, mold is generated around the hole, and the infrared transmitted light image tends to be blackened in the mold portion. Therefore, by detecting such a state, it is possible to determine whether or not it is a bug hole.

Raw knots are a type of knot in which the peripheral material and fibers are not torn. Since the veneer 6 has a color close to the material color, the boundary is often unclear, and it is often difficult to detect it only with a visible light image. On the other hand, in the infrared transmitted light image, the portion where the raw node exists is whitened. Therefore, the image analysis unit 13 can detect that there is a defect in the whitened portion in the infrared transmitted light image. In the case of a raw node, the shape of this whitened portion is close to a circle, so the image analysis unit 13 considers the color and shape of the white reflected light image of the corresponding portion and determines whether it is a bug hole or a raw node. Determine.

The dead node is a type of node in which peripheral materials and fibers are torn, and more of them are black than the raw node. Therefore, in many cases, the boundary with the surrounding material is relatively clearer than that of the raw node, and it may be detectable based on the visible light image. It is difficult to detect when the difference in color from the surrounding material is small. On the other hand, in the infrared transmitted light image, the portion where the dead node exists is whitened. Therefore, the image analysis unit 13 can detect that there is a defect in the whitened portion in the infrared transmitted light image. The shape of the dead node is close to a circle, but it is often blackened in the white reflected light image, so it is possible to distinguish the dead node by the hue.

Non-opening cracks are cracks that penetrate from the back surface to the front surface of the veneer 6 but do not open along the optical axis of the second visible light, or cracks that do not penetrate from the back surface to the front surface of the veneer 6. .. Since the second visible light from the transmitted light illumination 3 is not transmitted, the non-aperture crack cannot be detected based on the second visible light image. Further, since the difference between the color and the color pattern (pattern such as wood grain) is small from the normal portion of the peripheral material, it is difficult to detect only by the white reflected light image. On the other hand, in the infrared transmitted light image, the portion where the non-opening crack is present is whitened due to the influence of the insufficient plate thickness due to the chipping of the plate material and the light leaking from the space where the near-infrared light is likely to be scattered due to the crack. .. Therefore, the image analysis unit 13 can detect that there is a defect in the whitened portion in the infrared transmitted light image. Further, it is possible to determine that the defect is a "crack" from the shape of the whitened portion.

The backing material is a place where wood fibers are dense and tough. Fine cracks may occur due to the difference in density from the surrounding normal wood, but the color may not be characteristic, and it is often difficult to detect with only a visible light image. On the other hand, in the infrared transmitted light image, the part of the material is blackened if the wood fiber is simply high density. Therefore, the image analysis unit 13 can detect that there is a defect in the blackened portion in the infrared transmitted light image. In addition, from the shape of the blackened part and the color of the white reflected light image of the corresponding part, the discolored part (for example, the part blackened by blue mold or black mold) that is blackened in the infrared transmitted light image. It is possible to distinguish and determine that the defect is a "reliable material". The portion of the infrared transmitted light image with a clogged blade may be blackened, but it can be distinguished from the shape and the color of the white reflected light image.

The discolored part has no shape feature and was conventionally detected as a discolored part of the white reflected light image. However, the color is from black to blue-black, and it is difficult to distinguish it from stains. When the color of the material of the veneer 6 is black, it is difficult to detect its presence. It is also difficult to detect light-colored blue mold from the white reflected light image. On the other hand, in the infrared transmitted light image, the discolored portion is blackened. Therefore, the image analysis unit 13 can detect that there is a defect in the blackened portion in the infrared transmitted light image. In addition, the shape of the blackened part is not a shape peculiar to the contact material or blade clogging, and the defect is a discolored part based on the color of the white reflected light image of the corresponding part. It is possible to determine.

As described above, it is possible to discriminate the opening crack, the skin, the tar pot, and the blade clogging to some extent by the color, shape, pattern, etc. of the visible light image even if there is no infrared transmitted light image. However, in the infrared transmitted light image, these defective parts appear to be whitened or blackened. Therefore, after analyzing the infrared transmitted light image to extract the whitened or blackened part, the type of defect is determined by analyzing the color, shape, pattern, etc. of the corresponding part on the visible light image. You may do so.

In particular, when the acupuncture points have a large amount of accumulated acupuncture points and are only discolored and do not penetrate, it may be difficult to detect them only by analyzing a visible light image. If there is no significant change in the color of the blade clogging compared to the surrounding area, it may be difficult to detect it only by analyzing the visible light image. On the other hand, in the infrared transmitted light image, the portion where the acupuncture points are present is whitened, and the portion where the blade clogging is present is whitened or blackened, so that the presence of these defects can be detected. Further, it is possible to determine the type of defect from the shape of the whitened or blackened portion and the color of the corresponding portion of the visible light image.

FIG. 3 is a flowchart showing an operation example of a single plate defect inspection system according to the present embodiment configured as described above. The flowchart shown in FIG. 3 starts when an operation for instructing the defect inspection system to start the operation is performed. When the defect inspection system is instructed to start the operation, a plurality of veneers 6 are sequentially conveyed on the belt conveyor 5, and defects are sequentially detected for each veneer 6. An operation example when processing a single veneer 6 is shown.

First, in the line sensor camera 4, white visible light is emitted from the reflected light illumination 2 toward the front surface of the single plate 6, and the second visible light and near infrared rays are emitted from the transmitted light illumination 3 toward the back surface of the single plate 6. An image for one line is taken in the width direction of the single plate 6 in an environment where the light is irradiated, and the obtained taken image (visible light image and infrared transmitted light image) is output to the image processing device 1 ( Step S1).

The line image acquisition unit 11 of the image processing device 1 acquires a captured image for one line output from the line sensor camera 4 (step S2). Further, the whole image generation unit 12 further combines the captured images for one line acquired by the line image acquisition unit 11 with the captured images generated by synthesizing (combining) them (step S3). Then, the whole image generation unit 12 determines whether or not the whole image of the veneer 6 is generated (step S4).

For example, when a line-shaped edge due to a clear color difference in the width direction of the veneer 6 is detected in the combined photographed image, it can be determined that the entire image of the veneer 6 is generated. If the entire image of the veneer 6 has not yet been generated, the process returns to step S1. Then, the processes of steps S1 to S4 are repeated at predetermined sampling times.

On the other hand, when the whole image generation unit 12 determines that the whole image of the veneer 6 has been generated, the image analysis unit 13 analyzes the whole image of the veneer 6 generated by the whole image generation unit 12 (step S5). ). Here, the image analysis unit 13 analyzes the color, shape, and pattern of the white reflected light image, the color and shape of the second visible light image, and the shading and shape of the infrared transmitted light image. As a result, a plurality of types of defects existing in the veneer 6 are detected (step S6), and the results are displayed on the display of the defect inspection system.

As described in detail above, in the present embodiment, the visible light illumination (reflected light illumination 2) that irradiates the white visible light for reflected light toward the front surface of the single plate 6 and the back surface of the single plate 6 are directed. Generated by invisible light illumination (a part of transmitted light illumination 3) that irradiates near infrared rays for transmitted light, a line sensor camera 4 that captures the surface of a single plate 6 to generate an image, and a line sensor camera 4. It is equipped with an image processing device 1 (image analysis unit 13) that detects a plurality of types of defects in the single plate 6 by analyzing the captured image, and the shade and shape of the infrared transmitted light image based on the transmitted light, and By the color combination of the white reflected light image based on the reflected light, defects such as insect holes, live nodes, dead nodes, non-open cracks, pads, and discolored parts are discriminated.

According to the present embodiment configured in this way, even if the defect has a small color difference from the plate color of the normal portion on the white reflected light image taken by irradiating the white visible light, the defect portion is transmitted. The difference between the amount of near-infrared rays and the amount of near-infrared rays transmitted through the normal portion causes the difference in shade between the defective portion and the normal portion to be projected on the infrared transmitted light image. Therefore, it is possible to identify that there is a defect in the portion where there is a difference in shade. Furthermore, the type of possible defect can be determined from the combination of the shape of the defect portion identified on the infrared transmitted light image and the color of the corresponding portion on the white reflected light image. This makes it relatively easy to detect defects that are difficult to detect only by looking at the color difference of the white reflected light image. In particular, it is possible to easily detect defects such as insect holes, live nodes, dead nodes, non-open cracks, lumbers, and discolored parts.

Further, in the present embodiment, a second visible light illumination (a part of the transmitted light illumination 3) that irradiates the back surface of the single plate 6 with the second visible light for transmitted light is also provided, and the white reflected light image is displayed. The combination of the color, shape and pattern, the color and shape of the second visible light image, and the shading and shape of the infrared transmitted light image is used to discriminate a plurality of types of defects of the single plate 6. As a result, there are multiple types such as insect holes, live nodes, dead nodes, open cracks, non-open cracks, skins, tartars, lumbers, discolored parts, and blade clogging, including defects that can be detected only from visible light images. Defects can be easily detected. Even for defects that can be detected without using the second visible light image, if the veneer 6 penetrates almost vertically from the back surface to the front surface, it is easier to detect by adding analysis of the second visible light image. It is possible to do.

In the above embodiment, a line sensor camera 4 provided with a light receiving element having sensitivity to visible light and a light receiving element having sensitivity to near infrared light is used as an image pickup device to obtain a visible light image and an infrared transmitted light image. Although an example of generating the above separately has been described, the present invention is not limited to this. For example, a photographed image may be generated by using an image pickup device provided with a light receiving element having sensitivity from the visible region to the near infrared region, and the photographed image may be separated into a visible light image and an infrared transmitted light image.

Further, in the above embodiment, an example of analyzing a visible light image including a white reflected light image and a second visible light image has been described, but the present invention is not limited thereto. For example, the white reflected light image and the second visible light image may be separated from the visible light image, and each image may be analyzed separately.

Further, in the above embodiment, an example in which the line sensor camera 4 is used as the image pickup apparatus has been described, but the present invention is not limited thereto. For example, an area sensor camera capable of capturing the entire single plate 6 as a single image may be used. In this case, the reflected light illumination 2 and the transmitted light illumination 3 irradiate the entire single plate 6 with white visible light, second visible light, and near infrared rays in an area shape.

Further, in the above embodiment, an example in which a white light source is used as the visible light illumination and a blue light source or a green light source is used as the second visible light illumination has been described, but the present invention is not limited thereto. Other color combinations may be used as long as the color of the reflected light and the color of the transmitted light can be distinguished.

Further, in the above embodiment, an example including a second visible light illumination that irradiates the second visible light for transmitted light has been described, but this may be omitted. Even if this is omitted, defects such as insect holes, live nodes, dead nodes, non-open cracks, pads, discolored parts, etc., which were difficult to detect only by looking at the color difference of the visible light image, are relatively detected. It can be made easier.

Further, in the above embodiment, an example in which a near-infrared light source is used as invisible light illumination has been described, but the present invention is not limited thereto. Any wavelength band that can transmit through the single plate 6 and is compatible with the sensitivity band of the light receiving element of the line sensor camera 4 may be used. For example, mid-infrared rays, far-infrared rays, terahertz waves, ultraviolet rays, X-rays, etc. can be used. Is.

Further, in the above embodiment, an example of detecting a defect existing in the veneer 6 has been described, but the present invention is not limited thereto. It is also possible to detect defects in other plate-shaped wood such as sawn timber.

In addition, the above embodiments are merely examples of the embodiment of the present invention, and the technical scope of the present invention should not be construed in a limited manner. That is, the present invention can be implemented in various forms without departing from its gist or its main features.

1 Image processing device 2 Reflected light lighting (visible light lighting)
3 Lighting for transmitted light (second visible light lighting, non-visible light lighting)
4-line sensor camera (imaging device)
11 Line image acquisition unit 12 Overall image generation unit 13 Image analysis unit

Claims (8)

Visible light illumination that irradiates visible light for reflected light toward one surface of plate-shaped wood,
Invisible light illumination that irradiates invisible light for transmitted light toward the other surface opposite to the one surface of the plate-shaped wood, and
An image pickup device that captures one surface of the plate-shaped wood and generates an image,
It is equipped with an image processing device that detects a plurality of types of defects in the plate-shaped wood by analyzing the captured image generated by the image pickup device.
The image processing apparatus is a combination including at least the shade and shape of the image based on the invisible light transmitted through the plate-shaped wood and the color of the image based on the visible light reflected from the plate-shaped wood. A defect inspection system for plate-shaped wood, which is characterized by discriminating between multiple types of defects. The defect inspection system for plate-shaped wood according to claim 1, wherein the visible light illumination is a white light source, and the non-visible light illumination is a near infrared light source. A second visible light illumination that irradiates the other surface of the plate-shaped wood with visible light for transmitted light having a color that is easily distinguishable from the shade of the reflected light in the plate-shaped wood by the visible light illumination. The defect inspection system for plate-shaped wood according to claim 1, further comprising. The plate shape according to claim 3, wherein the visible light illumination is a white light source, the non-visible light illumination is a near infrared light source, and the second visible light illumination is a light source having a color different from that of the white light source. Wood defect inspection system. Visible light illumination irradiates visible light for reflected light toward one surface of the plate-shaped wood, and transmitted light is transmitted from the non-visible light illumination toward the other surface opposite to the one surface of the plate-shaped wood. The step of generating an image by photographing the one surface of the plate-shaped wood with an image pickup device after irradiating with invisible light for
The image processing device has a step of detecting a plurality of types of defects in the plate-shaped wood by analyzing the captured image generated by the image pickup device.
The image processing apparatus is a combination including at least the shade and shape of the image based on the invisible light transmitted through the plate-shaped wood and the color of the image based on the visible light reflected from the plate-shaped wood. A defect inspection method for plate-shaped wood, which comprises identifying a plurality of types of defects. The defect inspection method for plate-shaped wood according to claim 5, wherein the invisible light illumination is a near-infrared light source. A defect inspection program installed in an image processing device.
Visible light illuminates one surface of the plate-shaped wood with visible light for reflected light, and invisible light illumination emits transmitted light from the non-visible light illumination toward the other surface opposite to the one surface of the plate-shaped wood. A means for acquiring a photographed image generated by photographing the one surface of the plate-shaped wood in an environment irradiated with invisible light, and analyzing the photographed image to transmit the plate-shaped wood. By discriminating a plurality of types of defects of the plate-shaped wood by a combination including at least the shade and shape of the image based on the invisible light and the color of the image based on the visible light reflected from the plate-shaped wood, the plate-shaped wood is used. A program for inspecting defects in plate-shaped wood for operating the computer of the image processing apparatus as a means for detecting multiple types of defects in wood. The program for defect inspection of plate-shaped wood according to claim 7, wherein the invisible light illumination is a near-infrared light source.

Images