JP2007040913A - Method, apparatus, and program for inspecting wood - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform wood inspection with both reflected light and transmitted light by one photographing means. <P>SOLUTION: A first illumination means 6 performs illumination for surface inspection of wood, and a second illumiantion means 5 illuminates the back surfaces of the wood to photograph the wood by the photographing means 8. Images of parts of transmitted light from the second illumination means 5 are separated from images of the photographed wood by an image processing means to measure the shape and size of the separated parts of transmitted light. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an inspection method and apparatus for cracks and holes in a wood material such as a veneer cut out from a log of wood or the like. For example, in order to manufacture plywood, a log is cut with a blade to obtain a single plate having a thickness of several millimeters, the single plate is made to a predetermined size and dried, and then a plurality of single plates are bonded. It is integrated by bonding with an agent. In these manufacturing processes, the surface layer of the plywood is composed according to the position, number, area, etc. of the defects, such as cracks, discoloration due to mold and spears, where the nodes on the veneer dropped out and became holes. It is necessary to select (for example, select 5 to 7 stages) what is to be done (aesthetically less defective) and what constitutes the inner layer of the plywood (there is no problem even if there are many defects).

  Conventionally, in order to sort into the one constituting the surface layer when it becomes a plywood and the one constituting the inner layer of the plywood, the single plate conveyed by the conveyor is determined by the operator's naked eye.

In addition, as a conventional automatic detection of a wood defect, a mechanism for irradiating a veneer plywood with a light beam, and a first CCD camera for detecting an internal defect of the veneer plywood by scanning the transmitted light of the light beam, There is a defect state inspection system provided with a second CCD camera that scans the reflected light of the light beam (see Patent Document 1).
JP-A-6-300713

The conventional device has the following problems.
In the determination with the naked eye, there are problems that the determination varies from person to person (inaccuracy) and the speed of the conveyor cannot be increased (productivity is poor).

  In addition, the conventional defect state inspection system requires two sets of CCD cameras of transmitted light (first CCD camera) and reflected light (second CCD camera), which increases the cost.

  An object of the present invention is to solve such a conventional problem and to photograph a transmitted light and a reflected light from wood with a single photographing means so that a wood defect can be accurately detected. .

  FIG. 1 is an explanatory view of a single plate sorting apparatus. In FIG. 1, 1 is an image processing apparatus (image processing means), 2 is a sorter control apparatus, 3 is an operation panel, 4 is a belt conveyor, 5 is illumination for transmitted light (second illumination means), and 6 is reflected light. Illumination (first illumination means), 7 is a graded distribution device, 8 is a line sensor camera (imaging means), and 9 is a single plate (wood).

The present invention is configured as follows to solve the above problems.
(1) Illumination for wood surface inspection is performed by the first illumination means 6, the back surface of the wood is illuminated by the second illumination means 5, the wood is photographed by the photographing means 8, and the image processing means An image of the transmitted light portion from the second illumination means 5 is separated from the photographed wood image, and the shape and size of the separated transmitted light portion are measured. For this reason, the inspection by the reflected light of wood and the inspection by the transmitted light can be performed with one photographing means.

  (2) In the wood inspection method or apparatus of (1), the illumination of the second illumination means 5 uses a color different from the color from the reflected light from the surface of the wood. For this reason, the image by the reflected light of wood and the image by the transmitted light can be easily separated.

  (3) In the wood inspection method or apparatus according to (1) or (2), the image processing means combines the surface inspection using the reflected light from the surface of the wood and the inspection using the transmitted light of the wood. Make a decision. For this reason, sorting according to the grade of wood can be performed accurately.

The present invention has the following effects.
(1) In order to separate the image of the transmitted light portion from the second illumination means from the image of the wood photographed by the image processing means, and to measure the shape and size of the separated transmitted light portion, Inspection by reflected light of wood and inspection by transmitted light can be performed.

  (2) Since the illumination of the second illumination means uses a color different from the color from the reflected light from the surface of the wood, the image by the reflected light of the wood and the image by the transmitted light can be easily separated.

  (3) Since the image processing means combines the surface inspection by the reflected light from the surface of the wood and the inspection by the transmitted light of the wood, and makes the overall determination, the sorting by the grade of the wood can be performed accurately.

(1): Description of Single Plate Sorting Device FIG. 1 is an explanatory diagram of a single plate sorting device. FIG. 1 shows an overall configuration of a single plate sorting apparatus, which includes an image processing device 1, a sorter control device 2, an operation panel 3, a belt conveyor 4, a transmitted light illumination 5, and reflected light. Illumination for lighting 6, distribution device 7 according to grade, line sensor camera 8, and single plate 9 are provided.

  The image processing apparatus 1 is an image processing unit that processes image data from the line sensor camera 8 and outputs a processing result such as a single plate quality grade to the sorter control apparatus 2. The sorter control device 2 outputs the output of the sorter conveyor control signal such as the operation and stop of the conveyor and the control signal of the graded distribution device 7 according to the output of the image processing device 1. The operation panel 3 is an operation panel for performing operations such as changing the set value of the image processing apparatus 1 and controlling the sorter control apparatus 2. The belt conveyor 4 is a conveying means for conveying the single plate 9. The transmitted light illumination 5 is an illumination means (light source) such as an LED for detecting a hole in the single plate 9 and uses illumination of a color different from that of the reflected light illumination 6 (for example, green illumination). This is for distinguishing from the reflected light from the reflected light illumination 6 (by color and intensity) to detect a single plate hole (node hole), crack or the like. The reflected light illumination 6 is an illumination means (light source) such as an LED for detecting the reflected light of the single plate 9, and normally uses white illumination. The line sensor camera 8 is a photographing unit that photographs a line image of the single plate 9.

  In the operation of this single plate sorting device, the single plate 9 sent by the belt conveyor 4 is photographed by the line sensor camera 8 and the image data is output to the image processing device 1. The image processing apparatus 1 processes this image data, and outputs the processing results such as the veneer quality grade to the sorter control apparatus 2. The sorter control device 2 outputs a control signal to the grade distribution device 7 to sort the single plate 9 by grade. This selection is performed according to the number of bug holes, number of holes / missing nodes, number of live nodes, number of dead nodes, number of cracks, number of cracks, number of spears / penetration, blue variables, etc. and their size (area). Is called.

(2): Description of Image Processing Device FIG. 2 is an explanatory diagram of the image processing device. In FIG. 2, the image processing apparatus includes three line sensor cameras 8a, 8b, 8c, camera image acquisition bases (substrates) 11a, 11b, 11c, laser markers 12a, 12b, laser drivers 13a, 13b, and a main computer. 14 is provided.

  The line sensor cameras 8a, 8b, and 8c are photographing means for photographing the single plate 9 by dividing it into three in a direction orthogonal to the transport direction by three cameras. The camera image acquisition bases (substrates) 11a, 11b, and 11c perform digitization processing and distribute image data to the main computer 14 each time one line image from each line sensor camera is captured. The laser markers 12a and 12b are marks for combining (combining) the images from the line sensor cameras 8a, 8b and 8c. The laser drivers 13a and 13b are connected to an AC power source and drive the laser markers 12a and 12b. The main computer 14 includes processing means, storage means, output means, and the like, and performs image processing (node search, defect search processing, etc.) of the single plate 9.

  The operation of the image processing apparatus is to illuminate the conveyed single plate 9 with light rays from the light source 5 for transmitted light and the light source 6 for reflected light, and the line sensor camera 8a, The data is distributed to the main computer 14 every time one line image is captured from 8b and 8c. The main computer 14 sequentially combines the received images. When the camera image acquisition boards 11a, 11b, and 11c finally finish capturing the image, the main computer 14 combines the color image and the black and white density. The image conversion is almost completed (images from the camera image acquisition boards 11a, 11b, and 11c in which the single plate image is divided into three are combined by the main computer 14).

  Here, the single plate 9 is irradiated with laser marks from the laser markers 12a and 12b and divided into three parts, and the line sensor cameras 8a, 8b and 8c can easily match the line images up to the respective laser marks. So that they can be combined. In order to improve the processing speed of the image, the node searching process can be performed with a black and white image with a large number of pixels, and the color image for searching for a dead node can be performed with a reduced image (with a small number of pixels).

Hereinafter, the operation of the image processing apparatus will be described separately for processing during shooting and processing after shooting.
<Description of processing during shooting>
Image data photographed by the line sensor cameras 8a, 8b, and 8c is distributed to the main computer 14 for each line, and is synthesized as one whole image.

Processing of camera image acquisition boards 11a, 11b and 11c One line color image is taken from line sensor cameras 8a, 8b and 8c, the position (bonding position) of the laser mark is detected, and the one line color image is detected together with the information. Transfer to the main computer 14.

Processing of the main computer 14 The arrived one-line color image is synthesized based on the position information. This is because the main computer 14 completes the synthesis of the entire color image at the stage where the photographing is completed with the camera image acquisition boards 11a, 11b, and 11c and the last one-line color image is received. Further, in order to effectively use the time during shooting, processing that can be performed for each line, such as black and white conversion and reduction processing, can be performed in parallel.

<Description of processing during image analysis after shooting>
Processing of camera image acquisition boards 11a, 11b, 11c Wait until the arrival of the next board (single board) is detected.

・ Processing of the main computer 14 Based on the predetermined information such as the size and type of the target board, the node search process and the defect detection process using the transmitted light are performed according to the area to be calculated and the set value, and finally the classification process is performed. Do. The result is displayed on a display device (not shown) and the result is output to the sorter control device 2.

  In the above description, the camera image acquisition boards 11a, 11b, 11c, the main computer 14 and the like in the image processing apparatus are described as using computers (PCs). However, the number of computers used is the amount of image data. Or the processing speed of the computer or the like (can be processed by one computer).

(3): Explanation of color image rotation correction An image photographed by a line sensor camera is obtained by sequentially accumulating line images perpendicular to the moving direction of the conveyor in the memory (in the main computer 14). . At that time, if the single plate to be photographed is turned, the image is turned as it is, and an accurate dimension cannot be measured. Therefore, a reference angle (corner) is determined, and pixels at the outermost edges (edges) in two perpendicular directions sandwiching the corner are linearly approximated by the least square method to obtain the rotation angle of the plate. By rotating and correcting the entire image around a reference corner (corner), an image always aligned in the same direction can be obtained. At the same time, an approximate quadrangle composed of an approximate straight line of edges can be obtained.

  FIG. 3 is an explanatory diagram of the rotation correction of the color original image. FIG. 3A shows a photographed image that is rotating. FIG. 3B shows a reference corner (upper right in the figure) of the binarized captured image. FIG. 3C shows a photographed image whose rotation has been corrected.

(4): Description of dimension measurement The maximum effective dimension in the horizontal and vertical directions is measured for the image binarized and rotationally corrected in the previous section. For example, in the case of the upper long side, the portion that is not transmitted light is scanned line by line from the slightly longer side of the upper side of the approximate quadrangle obtained by the outer edge straight line approximation in the previous section. The number of pixels is counted. A position where the ratio of the number of pixels to the long side length (number of pixels) of the approximate rectangle is equal to or greater than a certain value (for example, 90% or more) is defined as the outer edge frame of the upper side. Similarly, the outer edge frame position is obtained for the lower long side and the left and right short sides. This is a process necessary for determining the outer edge frame even when the edges are not generally straight or missing.

  Furthermore, since the outer edge (four rounds) of the plate is usually excised and shaped using a saw or the like in a later process, for safety, a certain width (for example, 1200 × 1800 mm plate) 20 mm) is subtracted as a sawing allowance, and the inside is defined as an effective single plate frame (effective plate region). By doing so, it is possible to prevent as much as possible an error that causes a region outside the plate to be a defect through which light is transmitted.

  FIG. 4 is an explanatory diagram of dimension measurement. FIG. 4 (a) is an explanation of moving search for an image whose rotation is corrected until an edge portion is detected in the vertical and horizontal directions. FIG. 4B illustrates the detected outer edge frame as an effective single plate frame with an allowance for sawing.

(5): Explanation of node exploration The following (1) to (5) can be considered as the requirements for a clause to detect a clause.

(1) Overall dark (low brightness).
(2) It is darker than the surrounding area in the partial area (lightness is low).
(3) Concentration rises sharply at the node boundary.
(4) Many round shapes.
(5) There is a concentric grain around the periphery.

  A section candidate that has many items that fall under the requirements of these sections can be determined as a section candidate. For this reason, paying attention to the fact that the dark part (low brightness), which is one of the features of the node, has a circular shape, the probability distribution at that location is obtained, and the node candidate is specified. That is, in the method described in the section of (b) shape integration method described later, the black and white grayscale image is binarized while changing the threshold level, and a larger value is obtained when the shape of each binarized block is close to a circle. The node candidate can be determined by adding and integrating. In the node exploration, it is also possible to change which requirements are to be viewed larger or less depending on the wood material.

  FIG. 5 is a flowchart of the node search process. Hereinafter, a description will be given according to the processes S1 to S9 in FIG.

  S1: The image processing apparatus 1 obtains a minimum threshold value Tmin, a maximum threshold value Tmax, a preset threshold level division number N, a change value Td = (Tmax−Tmin) / N, and a repetition variable I of the received gray image. It is initialized to 0, and the process proceeds to process S2.

  S2: The image processing apparatus 1 changes the density threshold (T = Tmin + (Td × I)) and proceeds to processing S3.

  S3: The image processing apparatus 1 binarizes the image with the light and dark threshold value T, and proceeds to processing S4.

  S4: The image processing apparatus 1 calculates the circularity of each binarized figure, and adds individual circularity integration data (for each pixel for which the circularity is calculated by adding a weight based on the circularity to another storage area. Are accumulated), and the process proceeds to step S5.

  S5: The image processing apparatus 1 adds 1 to the repetition variable I (I = I + 1). If the repetition variable I is equal to or smaller than N (I ≦ N), the process returns to step S2, and the repetition variable I is If it is larger than N (I> N), the process proceeds to step S6.

  S6: The image processing apparatus 1 normalizes the circularity accumulated data, and proceeds to processing S7.

  S7: The image processing apparatus 1 creates an integrated image of the shape from the normalized integration data, and proceeds to processing S8.

  S8: The image processing apparatus 1 performs binarization processing on the integral image of the shape, and proceeds to processing S9.

  S9: The image processing apparatus 1 determines a node candidate.

(A) Description of the method of calculating the circularity of each binarized figure The method of calculating the circularity is performed as follows. FIG. 6 is an explanatory diagram of how to obtain the circularity, FIG. 6 (a) is an explanation of a circle, and FIG. 6 (b) is an explanation of an ellipse. In FIG. 6A, the radius of the circle is r. In FIG. 6B, the longer radius of the ellipse is a, and the shorter radius is b.

In the elliptic diagram of FIG. 6B, the length-to-short ratio is p = a / b.
<Circle> is area A = πr ²
Moment of inertia I = (π / 4) r ⁴
<Ellipse> is area A = πab
Moment of inertia I = (π / 4) a ³・ b
It has been known.

Where the moment of inertia of the ellipse is I = (π / 4) a ³・ b = (1 / 4π) (π ²・ a ²・ b ² ) (a / b)
= (1 / 4π) A ²・ p
(P = a / b: long / short ratio) When this equation is transformed, the following equation 1 is obtained.

p = 4π (I / A ² ) ... Equation 1
As an actual measurement, the moment of inertia is the center of the block as the origin and the image is g (x, y).
I ′ (actual measurement value) = Σ (x ² + y ² ) · g (x, y) (sum of squares of pixel positions of all pixels)
A '(actual value) = Σg (x, y) (all pixels)
Therefore, the long / short ratio p can be obtained by the following equation by substituting these into equation 1.

p = 4π (I '/ A' ² )
This long / short ratio p becomes larger as a flat ellipse with a perfect circle being 1.0. Therefore, if this reciprocal 1 / p is circularity, the circularity is in the range of 0.0 to 1.0, and the closer to a perfect circle, the larger the value closer to 1.0. A veneer is obtained by cutting a log with a cutter parallel to the longitudinal direction of the log, for example, and inside the log there are branches inclined with respect to the longitudinal direction, and this appears as a node. Therefore, since the shape of the node is an ellipse rather than a perfect circle, for example, a node having a circularity of 1/8 or more may be regarded as a node.

  FIG. 7 is an explanatory diagram of a binarized shape of a node. In FIG. 7, the binarized shape of the nodes is not the ideal shape as described above, and there are many shapes on the left side of FIG. Therefore, after performing processing for filling the inner side from the outermost periphery and making it the right side of FIG. 7, the elliptical approximate length-to-short ratio of the shape is obtained, and the reciprocal number is set as the circularity.

  The circularity obtained in this way is added to a memory using the coordinates of each pixel of the black block in FIG. 7 as an index, and the circularity integration of pixels binarized with a certain threshold is performed. In this way, the closer the block is to a perfect circle, the greater the degree of circularity is integrated and the larger the value.

  Furthermore, by performing the above integration while changing the threshold value from the minimum to the maximum, the density and shape can be examined simultaneously.

(B) Description of shape integration method A density contour line is obtained by searching the outer periphery of each block in a black and white image, and binarized with a specific threshold value. The threshold is changed at regular intervals, and the binarized images at each threshold are individually integrated. What is important here is that a dark portion has a value with respect to a larger number of threshold values, so that more contour lines (binarized images) can be obtained (a large integration effect).

(C) Explanation of smoothing of uneven color using local average value The surface of a single plate is not always a single color, and uneven color often exists. If there is a node on the veneer with such color unevenness (in a portion with a difference in light and shade), it is necessary to remove the color unevenness on the surface and emphasize only the lightness and darkness of the node. Therefore, the image processing apparatus 1 obtains the neighborhood average value for each pixel, and corrects the contrast of the original image based on the result (pre-node candidate detection process).

(6): Explanation of determination of node shape The process of determining the node shape is to obtain an optimum frame from the density change around the node position in order to obtain a more accurate size of the node frame. Specifically, it is an operation for obtaining an optimum threshold value for each node and binarizing it. Thus, the optimum shape and size are determined for each node candidate. This processing is performed for each connected pixel component (hereinafter referred to as a block) obtained by the previous binarization (see FIG. 7) in a region larger than each node candidate block, for example, in an extended region that is four times the shape integration. A threshold value for obtaining an optimum shape is obtained from the image and the differential image. Since this is performed in each subspace, the individual node shape can be accurately determined. Hereinafter, the number of pixels of the block when binarized is referred to as a block size (abbreviated size). The size stability (S) is the amount of change in size when the threshold value is changed (even if the node portion changes the threshold value, the amount of change in size is small up to a certain threshold value. Large, but when the threshold value is changed and the grain appears, the amount of change in size increases, which can remove blurred patterns such as stains.)

(7): Explanation of determination of life and death knots A knot is a knot that has a skin part, a knot that is easy to form a knot hole by losing the knot, and is a bad knot. In addition, since the skin portion of the dead node is carbonized in the dryer drying process, the color deviation value becomes large. If the ratio of the color deviation value is large in the area surrounding the node block obtained earlier (in this case, a color image is used), it can be determined as a dead node.

  FIG. 8 is an explanatory diagram of a three-dimensional color distribution. FIG. 8 shows a three-dimensional color distribution (RGB) in a region surrounding the node block, and a blackened portion and a blue changed portion (such as a burnt or mold that has entered outside) are surrounded by an ellipse. This region has a distribution that deviates from the standard color distribution of the original veneer. These parts are likely to be some sort of defect. FIG. 9 shows only the blackened portion in FIG. 8 as an image.

  FIG. 9 is an explanatory diagram of a color deviation image. In FIG. 9, the color deviation image is an image of the spatial distance from the center color (average value of rgb) in the standardized color space of each pixel. In this case, the blackened portion in FIG. 8 is imaged. It is a thing. In actual processing, the image processing apparatus 1 sets the overall average hue value to 0 (black) and multiplies the deviation value of each pixel by an appropriate coefficient (emphasizes a blacker portion) to form an image. Thereby, the discoloration part by factors other than natural woody, such as a burnt-out, can be detected. The dead part is charcoalized during the drying process of the dryer, so that the color deviation value becomes large and can be detected, resulting in an image as shown in FIG.

(8): Explanation of measurement of single-plate defect by transmitted light Defects such as cracks and holes in single plate face the line sensor camera (color sensor camera) and shine light from the back side (hereinafter referred to as backlight). This can be detected by examining the shape of the portion where the transmitted light is detected on the captured image.

  However, if a single plate surface is to be photographed at the same time, it is necessary to shine white light from the camera side. At that time, if there is strong reflection due to gloss or the like on the surface, it is very difficult to distinguish it from backlight light.

  Therefore, this problem was solved by using a light source of a color different from the white light source applied to the surface of the single plate as a light source for the backlight and a color that is not possible as the surface color of the single plate.

  For example, as the backlight illumination for inspection of wooden boards, the main component of the surface color of many wooden boards is a red color, so green or blue high-intensity LEDs that are separable from them or fluorescent lights of the same color Can be used.

  In general, in order to separate the pixels of the transmissive part and the reflective part in an image obtained by photographing an inspection object that is irradiated with transmitted light and reflected light by a line sensor camera, the surface color of the distribution in the color space of the image is separated. What is necessary is just to have the separation which a reflection part and a transmission part can isolate | separate. That is, if the linear distance between two colors positioned in the color space is a color difference, a light source that has a large color difference between the captured images of the transmission part and the reflection part may be used.

  For example, if the color space is represented by brightness, saturation, and hue, and the brightness of the surface color is low, even if the transmitted light is the same system color (same hue), the transmitted light pixel Can be separated.

  Even when the surface color includes white (the lightness of the surface color is high), the transmitted light pixel can be similarly separated if a light source having a different hue is used for the transmitted light.

(A) Explanation of transmitted light and reflected light FIG. 10 is an explanatory diagram of transmitted light and reflected light. In FIG. 10, transmitted light enters the line sensor camera 8 from the transmitted light source 5 through a hole or the like of the single plate 9, and reflected light is reflected from the reflected light source 6 by the single plate 9 to the line sensor camera 8. Incident.

  FIG. 11 is an explanatory view of measurement of a single plate defect portion by transmitted light. FIG. 11A illustrates the case where it is difficult to distinguish the surface light from the glossy reflection portion when the transmitted light is white. FIG. 11 (b) is an explanation that makes it easy to distinguish the glossy reflection portion by using different colors for transmitted light.

(B) Explanation of separation of transmitted light portion and reflected light portion An image taken in color by applying white light to the surface of a single plate to be inspected and illuminating with a specific color from the rear is reflected and reflected by the reflected surface color. It consists of pixels of a specific color. In order to separate only the transmitted light pixels of a specific color, if the line sensor camera output is an RGB output, it is only necessary to select pixels each having a specific value for R (red), G (green), and B (blue). It will be.

The RGB value ranges of the transmitted light portion are known in advance, and R (r1 to r2), G (g1 to g2), and B (b1 to b2)
(r1 <R <r2) AND (g1 <G <g2) AND (b1 <B <b2)
Pixels satisfying the conditions become transmitted light, and the others can be reflected light portions.

Alternatively, regarding the RGB value of the pixel of the transmitted light portion,
Luminance: S = R + G + B
Difference from red: Crg = (R−G) / S: Crb = (R−B) / S
Seeking
A pixel satisfying S> brightness threshold AND Crg> Crg threshold AND Crb> Crb threshold may be used as the transmitted light portion.

  Also, for example, when high-brightness pure green is used for transmitted light, there is usually no high-brightness pure green like a light source on a wooden board, but in rare cases it is a board with a color close to white, and it emits light with luster. May be white (all RGB components show a large value) due to total reflection, so if only the G (green) component of each pixel stored in RGB has a large value selected, Good.

Therefore,
Depending on whether or not the pixel satisfies the condition of G> G threshold AND R <R threshold AND B <B threshold, the captured image can be separated into a transmitted light portion and a reflected light portion and binarized.

(C) Description of defect type discrimination (method) by transmitted light Detection of a defect by transmitted light is performed within the frame (effective plate area) determined by the dimension measurement up to the previous item (see FIG. 4). In that case, it is assumed that the effective image area is an image area of the board completely, there is no area outside the outer edge frame of the board, and all transmitted light parts are treated as defective parts.

  The types of defects determined by transmitted light are roughly divided into holes, cracks and defects. When the holes are subdivided, they become pin holes (eg bug holes), node holes, and other missing holes. A defect is a hole in contact with the outer periphery and having a certain width or more. It is assumed that the length of the crack is a certain ratio or more (for example, 6 or more) with respect to the width.

  The image processing apparatus searches for a continuous block (block) of transmitted light within the effective image area of the binarized image. For one detected block, for example, the type of defect is determined as follows (see the explanatory diagram of the method for determining the type of defect using transmitted light in FIG. 12).

(Example of defect type discrimination method by transmitted light)
The type of defect due to transmitted light is determined by the circularity, whether or not it touches the edge, the ratio L (length) / W (width) of the block width to the length, and the maximum block width.

  ・ "Hole" has a circularity of less than 6 (close to a circle), does not touch the edge, the block width to length ratio L (length) / W (width) is less than 4, and the maximum block width is irrelevant .

  “Defect” has a circularity of less than 6 (close to a circle), a block width to length ratio L / W that is in contact with the edge is less than 4, and a maximum block width of 5 mm or more (less than 5 mm is cracked).

  ・ “Clearance” has a circularity of 6 or more (elongate), regardless of whether it touches the edge, the block width to length ratio L / W is 4 or more, and the maximum block width is less than 30 mm (30 mm or more is a hole) It is.

(Explanation of block width and length)
FIG. 13 is an explanatory diagram of the block width and length. In FIG. 13, the block length L is the length of the ridge line of the block, and the block width W is the maximum block width.

(D) Explanation of wood inspection processing by transmitted light and reflected light FIG. 14 is a wood inspection processing flowchart. Hereinafter, the wood inspection process will be described according to the processes S11 to S18 of FIG.

  S11: The image processing apparatus 1 sequentially stores the image of the line sensor camera (one line image) on the memory of the image processing apparatus to create one color photographed image, and proceeds to processing S12 and processing S16.

  S12: The image processing apparatus 1 separates the transmitted light color and the reflected light portion, binarizes them, and proceeds to processing S13.

  S13: The image processing apparatus 1 inspects the outermost periphery from the binarized image, measures the shape and dimensions, and proceeds to processing S14.

  S14: The image processing apparatus 1 determines the type and size of the defect from the shape of the transmitted light portion inside the rectangle determined by the dimension measurement, and proceeds to processing S15.

  S15: The image processing apparatus 1 outputs the transmitted light determination result, and proceeds to processing S18.

  S16: The image processing apparatus 1 performs surface image analysis (such as node exploration) using reflected light, and proceeds to processing S17.

  S17: The image processing apparatus 1 outputs the surface determination result, and proceeds to processing S18.

  S18: The image processing apparatus 1 compares the transmitted light determination result and the surface determination result with the selection reference value (specified value), and outputs a comprehensive determination result.

(9): Combined determination with surface determination result With respect to what is determined to be a hole in transmitted light determination, whether or not the hole is related to a node cannot be determined only by the transmitted light information. Therefore, detailed classification is performed by combining with the surface determination result.

  In the surface determination, all the holes are painted black, and the node positions and sizes are obtained in detail. Therefore, if a hole for determining transmitted light exists at that position and has the same size, it can be determined that the hole has a complete missing joint. In addition, if the size of the hole for transmitted light determination is less than one third of the size of the node for surface determination, it is considered that a part of the node has fallen out, so the hole is considered to be part of the dead node. Conceivable.

(10): Explanation of program installation Image processing device (image processing means) 1, sorter control device (sorter control means) 2, camera image acquisition boards (substrates) 11a, 11b, 11c, main computer 14, etc. It can be configured by a program, is executed by the main control unit (CPU), and is stored in the main memory. This program is processed by a computer. The computer includes hardware such as a main control unit, main memory, a file device, an output device such as a display device, and an input device.

  The program of the present invention is installed on this computer. In this installation, these programs are stored in a portable recording (storage) medium such as a floppy disk or a magneto-optical disk, and a drive device for accessing the recording medium provided in the computer is used. Alternatively, it is installed in a file device provided in the computer via a network such as a LAN. Accordingly, it is possible to easily provide a wood inspection apparatus capable of performing inspection using reflected light and inspection using transmitted light on a single photographing unit.

It is explanatory drawing of the single-plate sorting apparatus of this invention. It is explanatory drawing of the image processing apparatus of this invention. It is explanatory drawing of the rotation correction | amendment of the color original image of this invention. It is explanatory drawing of the dimension measurement of this invention. It is a node search process flowchart of this invention. It is explanatory drawing of how to obtain | require circularity of this invention. It is explanatory drawing of the binarization shape of the clause of this invention. It is explanatory drawing of the three-dimensional color distribution of this invention. It is explanatory drawing of the color deviation image of this invention. It is explanatory drawing of the transmitted light and reflected light of this invention. It is explanatory drawing of the measurement of the single-plate defect part by the transmitted light of this invention. It is explanatory drawing of the discrimination method of the kind of defect by the transmitted light of this invention. It is explanatory drawing of the block width and length of this invention. It is an inspection processing flowchart of the wood of the present invention.

Explanation of symbols

1 Image processing device (image processing means)
2 sorter control device 3 operation panel 4 belt conveyor 5 transmitted light illumination (second illumination means)
6 Illumination for reflected light (first illumination means)
7 Distributor according to grade 8 Line sensor camera (photographing means)
9 Single board (wood)

Claims (7)

Illumination for surface inspection of the wood is performed by the first illumination means, the back surface of the wood is illuminated by the second illumination means, the wood is photographed by the photographing means, and the second image is taken from the photographed wood image. A method for inspecting wood, comprising separating an image of a transmitted light portion from the illumination means and measuring a shape and a dimension of the separated transmitted light portion. 2. The wood inspection method according to claim 1, wherein the illumination of the second illumination means uses a color different from the color from the reflected light from the surface of the wood. 3. The method for inspecting wood according to claim 1, wherein a comprehensive determination is performed by combining a surface inspection using reflected light from the surface of the wood and an inspection using transmitted light from the wood. First illumination means for performing illumination for surface inspection of wood;
Second illumination means for illuminating the back surface of the wood;
Photographing means for photographing the wood;
Wood having an image processing means for separating an image of a transmitted light portion from the second illumination means from the photographed wood image and measuring a shape and a dimension of the separated transmitted light portion Inspection equipment. 5. The wood inspection apparatus according to claim 4, wherein the second illumination means for illuminating the back surface of the wood uses a color different from the color from the reflected light from the surface of the wood. 6. The wood inspection apparatus according to claim 4, wherein the image processing unit performs a comprehensive determination by combining a surface inspection using reflected light from the surface of the wood and an inspection using transmitted light of the wood. . The image of the reflected light from the first illumination means reflected on the surface of the wood and the image of the transmitted light from the second illumination means on the back of the wood are separated from the photographed wood image, and the separated transmission A program for causing a computer to function as image processing means for measuring the shape and dimensions of the light portion.

Images