An automatic identification system and method for wood chip surface defects

Abstract

The invention discloses an automatic recognition system and method for wood chip surface defects, wherein the system comprises a high-definition camera, a linear laser light source, a reflective lens and a motor; the linear laser light source is arranged in front of the reflective lens, the driving motor shaft of the motor is connected to the reflective lens, and the projection light emitted by the linear laser light source and formed on the wood chip surface and the plane formed by the linear laser light source form an angle of about 30-70 degrees with the wood chip surface; the automatic recognition system and method for wood chip surface defects provided by the invention are an efficient and reliable automatic grading system for wood chip surface defects, which can significantly improve the product quality of plywood and reduce production costs, and can quickly identify the position and outline of defects when used, and determine the grade of wood chips.

Description

Technical Field

[0001] The invention discloses a wood processing research and application development technology field, and in particular, relates to an automatic recognition system and method for wood chip surface defects. Background Art

[0002] Plywood is one of the important wood products. Its production process is to peel, peel, dry, glue, and mold the logs to make plywood. However, during the growth cycle of wood, defects such as knots and wormholes are easily formed during the growth process. In addition, during the drying process of the peeled wood chips, due to the growth stress of the wood and the evaporation rate of water, the surface of the veneer shrinks and cracks. These factors not only affect the appearance of the plywood, but also seriously affect the strength of the plywood, and thus affect the quality of the plywood.

[0003] Therefore, in the process of wood processing, dividing wood chips into different grades is conducive to improving the utilization rate of wood chips and improving the quality of products. For example, veneers with fewer surface defects are used for plywood panels, while wood chips with more defects are used for board cores or low-end plywood products. With the development of the wood processing industry, the analysis and judgment of the surface defects of wood chips and the automatic sorting of wood chips are particularly important. At present, many small processing companies still use manual grading in the wood chip grading process. This method uses the vision and work experience of the staff to judge the surface defect category of wood chips and evaluate the grade of veneers. This method has certain defects. On the one hand, manual visual judgment is greatly affected by subjective factors, resulting in different judgment results of different staff; on the other hand, wood chip sorting is labor-intensive, and manual sorting is not only inefficient, but also easy to produce visual fatigue for sorting personnel, resulting in wrong judgment of wood chip grading. All of the above factors affect the quality of plywood and the utilization rate of wood chips.

[0004] Therefore, establishing an efficient and reliable automatic grading system for wood chip surface defects is particularly important for improving the product quality of plywood and reducing production costs. Summary of the invention

[0005] In view of this, the present invention discloses an automatic identification system and method for wood chip surface defects, so as to realize intelligent and efficient sorting of wood chips.

[0006] In a first aspect, the present invention provides an automated identification system for wood chip surface defects, comprising a high-definition camera, a linear laser light source, a reflective lens, and a motor;

[0007] The linear laser light source is arranged in front of the reflective lens, the driving motor shaft of the motor is connected to the reflective lens, and the projection light formed on the surface of the wood chip emitted by the linear laser light source and the plane formed by the linear laser light source form an angle of about 30-70 degrees with the surface of the wood chip;

[0008] The high-definition camera is arranged vertically above the wood chip;

[0009] The motor shaft of the motor can rotate at an angle of 0-360 degrees;

[0010] The motor drives the motor shaft to drive the reflective lens to rotate at an angle, and the angle of each rotation is 0.001-0.2 degrees;

[0011] Furthermore, the plane exceeds the boundaries of the wood chips.

[0012] In a second aspect, the present invention provides a method for automatically identifying surface defects of wood chips using the system, comprising the following steps:

[0013] S1: The motor drives the motor shaft to drive the reflective lens to rotate at an angle, and the angle of each rotation is 0.001-0.2 degrees. The plane formed by the projection light and the linear laser light source is adjusted to form an angle of about 30-70 degrees with the surface of the wood chip, and the surface of the wood chip is scanned line by line. The high-definition camera located above the wood chip collects the image of the surface area of ​​the wood chip;

[0014] S2: When identifying defects on the wood chip surface, the motor drives the motor shaft to drive the reflective lens to rotate, projecting the projection light to the lower left corner of the wood chip, and the high-definition camera located above the wood chip collects the image of the wood chip surface area; if there is no defect on the wood chip surface, the projection of the light on the wood chip surface is a straight line; if there is a crack or hole defect on the wood chip surface, the straight line projected by the light on the wood chip surface will become a broken line;

[0015] S3: For wood chips with cracks or holes, a rapid detection method for wood chip surface defects is used to quickly identify the location and outline of the defects;

[0016] S4: After completing the above steps, the motor rotation angle is set to 0.01 degrees, driving the projection light to move parallel to the upper right corner, and the above scanning is repeated until the projection light leaves the wood chip from the upper right corner of the wood chip; through the above scanning, the number and distribution pattern of defects on the surface of the wood chip are obtained, and the grade of the wood chip is determined.

[0017] Furthermore, the method for rapid detection of wood chip surface defects comprises:

[0018] S31: Obtain the laser projection pattern on the wood chip surface; S32: Obtain the outer contour points of the defects; S33: Obtain the defect points; S34: Obtain the defect area; S35: Calculate the defect area; S36: Determine the grade of the wood chips.

[0019] Further, the S31 is specifically as follows: the image data obtained by shooting with a camera is pixel data containing red, green and blue colors, and the number of pixel points in the image row is N, and the number of pixel points in the column is M; the image data is described using a five-dimensional array Image[i,j,r,g,b], where i represents the row number of the current pixel, j represents the column number of the current pixel, r represents the red value of the pixel, g represents the green value of the pixel, and b represents the blue value of the pixel;

[0020] Perform color screening on the collected wood chip surface image Image[i,j,r,g,b]: set the critical values ​​Cr, Cg, and Cb of the red, green, and blue colors of the image, and judge the red, green, and blue colors of all pixels in the image. If the red, green, and blue color values ​​of the pixel point are all higher than the critical values ​​Cr, Cg, and Cb of the red, green, and blue colors, then set the r, g, and b pixel values ​​of the point to 0. Otherwise, set the r, g, and b pixel values ​​of the point to 1. After completing the setting of all pixels, a black and white image is obtained.

[0021] Further, the S32 is specifically:

[0022] Create a curve array Outline[k,i k ,j k ], k is the serial number of the point on the curve, and its initial value is 1; i k Indicates the row number of the current pixel in the Image array, j k Indicates the column number of the current pixel in the Image array;

[0023] Create a counter Num to record the number of curve array elements, with an initial value of 0;

[0024] Take the first row of data of Image[i,j,r,g,b], that is, i=1, and analyze the information of the pixels column by column from the first column to the last column, that is, from j=1 (that is, Image[1,1,r,g,b]) to j=M (that is, Image[1,M,r,g,b]), and judge whether the rgb values ​​of the point Image[i,j,r,g,b] are all equal to 0. If not, then j=j+1, and continue to judge Are the rgb values ​​all equal to 0 until j=M; if the rgb values ​​of Image[i,j,r,g,b] are all equal to 0, then record the i and j values ​​of the current Image[i,j,r,g,b] pixel in the Outline[k,i,j] array, and set the ordinal k=k+1, and increment the counter Num value, that is, Num=Num+1; and the image Image[i,j,r,g,b] jumps to the next row, that is, row i=i+1;

[0025] Repeat the above pixel steps until i = N; traverse all the pixels in the image Image[i,j,r,g,b] and get Outline[k,i k ,j k ]The number Num of curves and curve array elements.

[0026] Furthermore, the S33 is specifically:

[0027] Create a curve first-order derivative array DaoShu[h,i h ,j h ,v h ], h is the serial number of the first-order derivative element, the derivative array has Num-1 elements, i h Indicates the row number of the derivative array element with serial number h in the Image array; j h Indicates the column number of the derivative array element with serial number h in the Image array; v h Represents the curve array Outline[k,i k ,j k ]The derivative value at point k = h, v h The value is equal to the curve array Outline[h+1,i h+1 ,j h+1 ] column number j h+1 Subtract Outline[h,i h ,j h ] column number j h The value of Outline[h+1,i h+1 ,j h+1 ]'s row number i h+1 Value minus Outline[h,i h ,j h ]'s row number i h value;

[0028] Get the average value MeanDaoShu of the derivative array DaoShu, MeanDaoShu is equal to all derivative arrays DaoShu[h,i h ,j h ,v h ] divided by the number of elements in the derivative array Num-1;

[0029] Create a defect array QueXian[p,i p ,j p ], p is the serial number of the defect array element, and its initial value is 0; i p is the row number of the defect array element with serial number p in the image array ImageImage picture array; j pis the column number of the defect array element with the serial number p in the image array ImageImage picture array;

[0030] Set the critical value C of the derivative array deviating from the critical point DaoShu , judge the first-order derivative array of the curve DaoShu[h,i h ,j h ,v h ] h Is the absolute value of the difference with MeanDaoShu greater than C? DaoShu , if its interpolation value is greater than C DaoShu Then QueXian[p,i p ,j p ]'s pth element i p =i h , j p =j h , and analyze each element in the DaoShu array according to the above method with p=p+1, and preliminarily obtain the defect element group QueXian; delete the first and last points in the defect element group QueXian, and obtain the defect array QueXian[p,i p ,j p ];

[0031] Define a total SumQueXian[t,i t ,j t ], used to count the defects on the surface of the entire wood chip, t is the defect number, i t is the row number of the defect point with serial number t in Image[i,j,r,g,b]; j t is the column number of the defect point with serial number t in Image[i,j,r,g,b]; a monochromatic line laser is used to gradually move from the lower left corner of the wood chip surface to the upper right corner of the wood chip at a certain angle, and each movement distance is 1-5mm relative to the previous position. Each time the line laser is moved, a high-definition camera is used to take a photo of the line laser projection on the wood chip surface. According to the above method, for each movement of the line laser, the defect array QueXian[p,i p ,j p ] to accumulate, and QueXian[p,i p ,j p ] are accumulated into SumQueXian[t,i p ,j p ], thereby obtaining the defect data of the entire wood chip surface.

[0032] Further, the S34 is specifically as follows:

[0033] Define the defect outline array QueXian_Outline[r,i r ,j r ], r is the number of the point in the defect contour array, and the initial value of r is 1; i r is the row number of the contour array with serial number r in the image array Image[i,j,r,g,b]; j r is the column number of the contour array with serial number r in the Image[i,j,r,g,b]e image array; take the defect array QueXian[p,i p ,j p ], QueXian[1,i1,j1], QueXian_Outline adds this point, r=r+1,i r =i p ,j r =j p , determine the distance d between the element and the defect array and the rest of the elements, d = (j p -j1)*(j p -j1)+(i p -i1)*(i p -i1) square root; take the defect array element with the smallest d as the next point, in QueXian[p,i p ,j p ] and record the element in QueXian_Outline;

[0034] Repeat the above method until all defect points QueXian are analyzed and the defect outline QueXian_Outline array is obtained.

[0035] Furthermore, since the outlines of wood chip surface defects are mostly convex polygons, we will use QueXian_Outline[r,i r ,j r ], there are k points in total, from QueXian_Outline[r,i r ,j r ] points to points other than Outline[1,2,i,j] and the end point, and k-1 triangles can be formed; for each triangle, calculate the lengths of the three sides: a, b, c, and the average perimeter p = (a+b+c) / 2, according to the formula: the area of ​​the triangle s = 1 / 2 of p(pa)(pb)(pc);

[0036] Count the areas of all k-1 triangles to get the defect area S, and count the number of defects Num = Num + 1 and the sum of the defect areas S sum =S sum +S;

[0037] After completing the above steps S31-34, find the points whose pixel values ​​r, g, and b are not zero in the three-dimensional array Image array again, and execute steps S31-34 until all pixel points in the Image array are 1, and finally obtain the number Num of all defects on the wood chip and the sum Ssum of all defect areas.

[0038] Furthermore, the S36 is specifically as follows: preset the critical defect area S of the grade A plate a , the value range is: 0.0001-0.2 square meters; B-level board critical defect area (S b ), the value range is: 0.0001-0.2 square meters; when the defect area S on the wood chip sum a It is judged as Grade A board; when the defect area S on the wood chip a sum b It is judged as Class B board; when the defect area S on the wood chip sum >S b It is judged to be a C-grade board.

[0039] The present invention provides an automatic identification system and method for wood chip surface defects, which is an efficient and reliable automatic grading system for wood chip surface defects, and can significantly improve the product quality of plywood and reduce production costs. When used, it can quickly identify the location and outline of defects and determine the grade of wood chips.

[0040] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A schematic diagram of the structure of an automatic identification system for wood chip surface defects provided by an embodiment of the present invention;

[0044] Figure 2 ​​​A distribution diagram of key points when scanning a wood chip in an automatic identification system for wood chip surface defects provided by an embodiment of the present invention;

[0045] Figure 3 A method for automatically identifying wood chip surface defects provided in an embodiment of the present invention is provided to obtain a black and white image of a wood chip surface projected by a laser;

[0046] Figure 4 A schematic diagram of calculating defect area in an automatic identification method of wood chip surface defects provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present invention. Instead, they are merely examples of systems consistent with some aspects of the present invention as detailed in the appended claims.

[0048] In order to solve the problem in the prior art that the wood chip surface defect category is judged and the veneer grade is evaluated only by artificial vision and work experience, firstly, this embodiment provides an automatic recognition system for wood chip surface defects, such as Figure 1-2 As shown, it includes a high-definition camera 1, a linear laser light source 2, a reflective lens 10 and a motor 12; the high-definition camera 1 is arranged above the vertical wood chip, and can capture the wood chip and the edge area. The reflective lens 10 is arranged on the motor shaft 11 on the motor 12, and the motor shaft 11 driven by the motor 12 rotates (the rotation angle is: 0-360 degrees), driving the reflective lens 10 to rotate, and projecting the projection light 5 on the surface of the wood chip, and the projection range of the light must exceed the boundary of the wood chip, so as to detect the edge and defects of the wood chip. As the angle of rotation of the motor 12 changes, the projection light 5 can continuously change from the lower left corner to the upper right corner of the wood chip, and the plane formed by the projection light 5 and the linear laser light source 2 forms an angle of about 30-70 degrees with the surface of the wood chip, thereby realizing the scanning of the surface of the wood chip, as shown in FIG. Figure 1 In the middle, the edge point of the wood chip is 6; the defect point is 7-8; the edge point of the wood chip is 9;

[0049] When the wood chip surface defect identification starts, the motor 12 drives the motor shaft 11 to drive the reflective lens 10 to rotate (the angle of each rotation is 0.001-0.2 degrees, preferably 0.01 degrees), and the plane formed by the projection light 5 and the linear laser light source 2 forms an angle of about 30-70 degrees with the wood chip surface, and the projection light 5 is projected to the lower left corner of the wood chip 3, and the wood chip surface is scanned line by line; the high-definition camera located above the wood chip collects the image of the wood chip surface area; if there is no defect on the wood chip surface, the projection of the light on the wood chip surface is a straight line; if there is a crack or hole defect on the wood chip surface, the straight line projected by the light on the wood chip surface will become a broken line; the wood chip surface defect rapid detection method can be used to quickly identify the location and contour of the defect. After completing the above action, the motor rotates at an angle of 0.01 degrees, driving the projection light 5 to move parallel to the upper right corner, and continue to repeat the above scanning until the projection light 5 leaves the wood chip from the upper right corner of the wood chip. Through the above scanning, the number and distribution pattern of defects on the surface of the wood chips can be obtained, and the grade of the wood chips can be determined.

[0050] The above-mentioned method for rapid detection of wood chip surface defects specifically comprises the following steps:

[0051] (1) Obtaining the laser projection pattern on the wood chip surface

[0052] A monochrome line laser is projected onto the surface of the wood chip at a certain angle, and a high-definition camera set in the vertical direction of the wood chip is used to shoot the surface of the wood chip. The image data obtained by the camera shooting is pixel data containing red, green and blue colors. The number of row pixels and column pixels obtained is N, and the number of column pixels is M. It can be described by a five-dimensional array Image[i,j,r,g,b], where i represents the row number of the current pixel; j represents the column number of the current pixel; r represents the red value of the pixel; g represents the green value of the pixel; and b represents the blue value of the pixel.

[0053] The collected wood chip surface image Image[i,j,r,g,b] is color screened. The specific method is to set the critical values ​​of the red, green and blue colors of the image (Cr, Cg, Cb), and judge the red, green and blue colors of all pixels in the image. If the red, green and blue color values ​​of the pixel point are all higher than the critical values ​​of the red, green and blue colors (Cr, Cg, Cb), then the r, g, and b pixel values ​​of the point are set to 0, otherwise the r, g, and b pixel values ​​of the point are set to 1. After completing the setting of all pixels, a black and white image is obtained, as shown in Figure 3 As shown, the image is composed of pixels in both row and column directions. The types of pixels are pixels where both RGB are 1 (110) and pixels where both RGB are 0 (111).

[0054] (2) Obtaining the defect outer contour points

[0055] Create a curve array Outline[k,i k ,j k ], k is the serial number of the point on the curve, and its initial value is 1; i k Indicates the row number of the current pixel in the Image array; j k Indicates the column number of the current pixel in the Image array; creates a counter Num to record the number of curve array elements, with an initial value of 0.

[0056] Take the first row of data of Image[i,j,r,g,b], that is, i=1, and analyze the information of the pixels column by column from the first column to the last column, that is, from j=1 (that is, Image[1,1,r,g,b]) to j=M (that is, Image[1,M,r,g,b]), and judge whether the rgb values ​​of the point Image[i,j,r,g,b] are all equal to 0. If not, then j=j+1, and continue to judge Are the rgb values ​​all equal to 0 until j=M; if the rgb values ​​of Image[i,j,r,g,b] are all equal to 0, then record the i and j values ​​of the current Image[i,j,r,g,b] pixel in the Outline[k,i,j] array, and set the ordinal k=k+1, and increment the counter Num value, that is, Num=Num+1; and the image Image[i,j,r,g,b] jumps to the next row, that is, row i=i+1.

[0057] Repeat the above pixel steps until i = N. After traversing all the pixels in the image Image[i,j,r,g,b], we get Outline[k,i k ,j k ]The number Num of curves and curve array elements.

[0058] (3) Obtaining defect points

[0059] Create a curve first-order derivative array DaoShu[h,i h ,j h ,v h ], h is the serial number of the first-order derivative element, and the elements of the derivative array have Num-1, that is, the number range of h values ​​is 1 to Num-1, i h Indicates the row number of the derivative array element with serial number h in the Image array; j h Indicates the column number of the derivative array element with sequence number h in the Image array. h Represents the curve array Outline[k,i k ,j k ]The derivative value at point k = h, vh The value is equal to the curve array Outline[h+1,i h+1 ,j h+1 ] column number j h+1 Subtract Outline[h,i h ,j h ] column number j h The value of Outline[h+1,i h+1 ,j h+1 ]'s row number i h+1 Value minus Outline[h,i h ,j h ]'s row number i h Take the point h=1 as an example, DaoShu[1,i1,j1,v1]=(j2-j1) / (i2--i1);

[0060] Get the average value (MeanDaoShu) of the derivative array DaoShu, MeanDaoShu is equal to all derivative arrays DaoShu[h,i h ,j h ,v h ] divided by the number of elements in the derivative array Num-1. Create a defect array QueXian[p,i p ,j p ] is used to record the coordinates of the defect, p is the serial number of the defect array element, and its initial value is 0; i p is the row number of the defect array element with serial number p in the image array ImageImage picture array; j p is the column number of the defect array element with serial number p in the image array ImageImage picture array. Set the critical value C of the derivative array deviating from the critical point DaoShu (Its value is based on the analysis accuracy requirement, and is between 0.01 and 1, preferably 0.2), and the first-order derivative array DaoShu[h,i h ,j h ,v h ] h Is the absolute value of the difference with MeanDaoShu greater than C? DaoShu , if its interpolation value is greater than C DaoShu Then QueXian[p,i p ,j p ]'s pth element i p =i h , j p =j h, and p=p+1, analyze each element in the DaoShu array according to the above method, and preliminarily obtain the defect element group QueXian. Delete the first and last points in the defect element group QueXian (these two points correspond to the edge of the wood chip, not the defect point), and obtain the defect array QueXian[p,i p ,j p ].

[0061] Define a total SumQueXian[t,i t ,j t ], used to count the defects on the surface of the entire wood chip, t is the defect number, i t is the row number of the defect point with serial number t in Image[i,j,r,g,b]; j t is the column number of the defect point with serial number t in Image[i,j,r,g,b]. Use a monochromatic line laser to gradually move from the lower left corner of the wood chip surface to the upper right corner of the wood chip at a certain angle. Each movement distance is 1-5mm relative to the previous position (that is, the distance between two parallel light rays is 1-5mm, preferably 1mm). Each time the line laser is moved, a high-definition camera is used to take a photo of the line laser projection on the wood chip surface. According to the above method, for each movement of the line laser, the defect array QueXian[p,i p ,j p ] to accumulate, and QueXian[p,i p ,j p ] are accumulated into SumQueXian[t,i p ,j p ], thereby obtaining the defect data of the entire wood chip surface.

[0062] (4) Obtaining defect area

[0063] Define the defect outline array QueXian_Outline[r,i r ,j r ], r is the number of the point in the defect contour array, and the initial value of r is 1; i r is the row number of the contour array with serial number r in the image array Image[i,j,r,g,b]; j r is the column number of the contour array with serial number r in the Image[i,j,r,g,b]e image array. p ,j p ], QueXian[1,i1,j1], QueXian_Outline adds this point, r=r+1,i r =i p,j r =j p , determine the distance d between the element and the defect array and the rest of the elements, d = (j p -j1)*(j p -j1)+(i p -i1)*(i p -i1). Take the defect array element with the smallest d as the next point, and in QueXian[p,i p ,j p ] and record the element in QueXian_Outline.

[0064] Repeat the above method until all defect points QueXian are analyzed and the defect outline QueXian_Outline array is obtained.

[0065] (5) Calculation of defect area

[0066] Since the outlines of wood chip surface defects are mostly convex polygons, we use QueXian_Outline[r,i r ,j r ] (a total of k points), from QueXian_Outline[r,i r ,j r ] points to points other than Outline[1,2,i,j] and the end point, k-1 triangles can be formed. For each triangle, calculate the lengths of the three sides: a, b, c, and the average perimeter p = (a+b+c) / 2, according to the formula: triangle area s = p(pa)(pb)(pc) to the power of 1 / 2. Count the areas of all k-1 triangles to get the defect area S, count the number of defects Num = Num+1 and the sum of the defect areas S sum =S sum +S, such as Figure 4 .

[0067] After completing steps 1-4 above, find points in the three-dimensional array Image whose pixel values ​​r, g, and b are not zero again, and execute steps 1-4 until all pixels in the Image array are 1. Finally, the number of all defects on the wood chip Num and the sum of all defect areas S are obtained. sum .

[0068] (6) Determine the grade of wood chips

[0069] Preset critical defect area of ​​Class A plate (S a ), the value range is: 0.0001-0.2 square meters, preferably 0.001 square meters; B-grade plate critical defect area (S b), the value range is: 0.0001-0.2 square meters, preferably 0.01 square meters; when the defect area S on the wood chip sum a It is judged as Grade A board; when the defect area S on the wood chip a sum b It is judged as Class B board; when the defect area S on the wood chip sum >S b It is judged to be a C-grade board.

[0070] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the claims.​​​

Claims

1. A method for automatically identifying surface defects of wood chips, characterized in that: An automated identification system is used to identify surface defects of wood chips, wherein the automated identification system comprises a high-definition camera (1), a linear laser light source (2), a reflective lens (10), and a motor (12); The linear laser light source (2) is arranged in front of the reflective lens (10), the drive motor shaft (11) of the motor (12) is connected to the reflective lens (10), and the projection light (5) formed on the surface of the wood chip emitted by the linear laser light source (2) and the plane formed by the linear laser light source (2) form an angle of 30-70 degrees with the surface of the wood chip; The high-definition camera (1) is arranged vertically above the wood chip; The motor shaft (11) of the motor (12) can rotate at an angle of 0-360 degrees; The motor (12) drives the motor shaft (11) to drive the reflective lens (10) to rotate at an angle, and the angle of each rotation is 0.001-0.2 degrees; The method for automatically identifying surface defects of wood chips using the system comprises the following steps: S1: The motor (12) drives the motor shaft (11) to drive the reflective lens (10) to rotate at an angle, and the angle of each rotation is 0.001-0.2 degrees. The plane formed by the projection light (5) and the linear laser light source (2) is adjusted to form an angle of about 30-70 degrees with the surface of the wood chip, and the surface of the wood chip is scanned line by line. The high-definition camera located above the wood chip collects an image of the surface area of ​​the wood chip; S2: When identifying defects on the wood chip surface, the motor (12) drives the motor shaft (11) to drive the reflective lens (10) to rotate, projecting the projection light (5) onto the lower left corner of the wood chip (3), and the high-definition camera located above the wood chip collects an image of the surface area of ​​the wood chip; if there are no defects on the surface of the wood chip, the projection of the light on the surface of the wood chip is a straight line; if there are cracks or holes on the surface of the wood chip, the straight line projected by the light on the surface of the wood chip will become a broken line; S3: For wood chips with cracks or holes, a rapid detection method for wood chip surface defects is used to quickly identify the location and outline of the defects; S4: After completing the above steps, the motor rotation angle is set to 0.01 degrees, driving the projection light (5) to move parallel to the upper right corner, and the above scanning is repeated until the projection light (5) leaves the wood chip from the upper right corner of the wood chip; through the above scanning, the number and distribution of defects on the surface of the wood chip are obtained, and the grade of the wood chip is determined; The method for rapid detection of wood chip surface defects comprises: S31: obtaining the laser projection pattern on the wood chip surface; S32: obtaining the outer contour points of the defects; S33: obtaining the defect points; S34: obtaining the defect area; S35: calculating the defect area; S36: determining the grade of the wood chip; The S31 is specifically as follows: the image data obtained by the camera is pixel data containing red, green and blue colors, and the number of pixel points in the image row is N, and the number of pixel points in the column is M; the image data is described using a five-dimensional array Image[i,j,r,g,b], where i represents the row number of the current pixel point, j represents the column number of the current pixel point, r represents the red value of the pixel point, g represents the green value of the pixel point, and b represents the blue value of the pixel point; Perform color screening on the collected wood chip surface image Image[i,j,r,g,b]: set the critical values ​​Cr, Cg, and Cb of the red, green, and blue colors of the image, and judge the red, green, and blue colors of all pixels in the image. If the red, green, and blue color values ​​of the pixel point are all higher than the critical values ​​Cr, Cg, and Cb of the red, green, and blue colors, then set the r, g, and b pixel values ​​of the point to 0. Otherwise, set the r, g, and b pixel values ​​of the point to 1. After completing the setting of all pixels, a black and white image is obtained. The S32 is specifically: Create a curve array Outline[k,i k ,j k ], k is the serial number of the point on the curve, and its initial value is 1; i k Indicates the row number of the current pixel in the Image array, j k Indicates the column number of the current pixel in the Image array; Create a counter Num to record the number of curve array elements, with an initial value of 0; Take the first row of data of Image[i,j,r,g,b], that is, i=1, and analyze the information of the pixels column by column from the first column to the last column, that is, from j=1 to j=M, and judge whether the rgb values ​​of the point Image[i,j,r,g,b] are all equal to 0. If not, then j=j+1, and continue to judge whether the rgb values ​​are all equal to 0 until j=M; if the rgb values ​​of Image[i,j,r,g,b] are all equal to 0, then record the i and j values ​​of the current Image[i,j,r,g,b] pixel in the Outline[k,i,j] array, and set the ordinal k=k+1, and increment the counter Num value, that is, Num=Num+1; and the image Image[i,j,r,g,b] jumps to the next row, that is, row i=i+1; Repeat the above pixel steps until i=N; traverse all the pixels in the image Image[i,j,r,g,b] and get Outline[k,i k ,j k ]The number of curves and curve array elements Num; The S33 is specifically: Create a curve first-order derivative array DaoShu[h,i h ,j h ,v h ], h is the serial number of the first-order derivative element, the derivative array has Num-1 elements, i h Indicates the row number of the derivative array element with serial number h in the Image array; j h Indicates the column number of the derivative array element with serial number h in the Image array; v h Represents the curve array Outline[k,i k ,j k ]The derivative value at point k=h, v h The value is equal to the curve array Outline[h+1,i h+1 ,j h+1 ] column number j h+1 Subtract Outline[h,i h ,j h ] column number j h The value of Outline[h+1,i h+1 ,j h+1 ]'s row number i h+1 Value minus Outline[h,i h ,j h ]'s row number i h value; Get the average value MeanDaoShu of the derivative array DaoShu, MeanDaoShu is equal to all derivative arrays DaoShu[h,i h ,j h ,v h ] divided by the number of elements in the derivative array Num-1; Create a defect array QueXian[p,i p ,j p ], p is the serial number of the defect array element, and its initial value is 0; i p is the row number of the defect array element with the serial number p in the Image array; j p is the column number of the defect array element with the serial number p in the Image array; Set the critical value C of the derivative array deviating from the critical point DaoShu , judge the first-order derivative array of the curve DaoShu[h,i h ,j h ,v h ] h Is the absolute value of the difference with MeanDaoShu greater than C? DaoShu , if its interpolation value is greater than C DaoShu Then QueXian[p,i p ,j p ]'s pth element i p =i h , j p =j h , and analyze each element in the DaoShu array according to the above method, and preliminarily obtain the defective element group QueXian; Delete the first and last points in the defect element group QueXian, and obtain the defect array QueXian[p,i p ,j p ]; Define a total SumQueXian[t,i t ,j t ], used to count the defects on the surface of the entire wood chip, t is the defect number, i t is the row number of the defect point with serial number t in Image[i,j,r,g,b]; j t is the column number of the defect point with serial number t in Image[i,j,r,g,b]; a monochromatic line laser is used to gradually move from the lower left corner of the wood chip surface to the upper right corner of the wood chip at a certain angle, and each movement distance is 1-5mm relative to the previous position. Each time the line laser is moved, a high-definition camera is used to take a photo of the line laser projection on the wood chip surface. According to the above method, for each movement of the line laser, the defect array QueXian[p,i p ,j p ] to accumulate, and QueXian[p,i p ,j p ] are accumulated into SumQueXian[t,i p ,j p ], thereby obtaining the defect data of the entire wood chip surface; The S34 is specifically: Define the defect outline array QueXian_Outline[r,i r ,j r ], r is the number of the point in the defect contour array, and the initial value of r is 1; i r is the row number of the contour array with serial number r in the image array Image[i,j,r,g,b]; j r is the column number of the contour array with serial number r in the image array Image[i,j,r,g,b]; take the defect array QueXian[p,i p ,j p ], QueXian[1,i1,j1], QueXian_Outline adds this point, r=r+1,i r =i p ,j r =j p , determine the distance d between this element and the rest of the defect array, d = (j p -j1)*(j p -j1)+(i p -i1)*(i p -i1) square root; take the defect array element with the smallest d as the next point, in QueXian[p,i p ,j p ] and record the element in QueXian_Outline; Repeat the above method until all defect points are analyzed and the defect outline QueXian_Outline array is obtained.

2. The method for automatically identifying surface defects of wood chips according to claim 1, characterized in that: The plane extends beyond the boundaries of the wood chip.

3. The method for automatically identifying surface defects of wood chips according to claim 1, characterized in that: The S35 is specifically as follows: since the outlines of the defects on the wood chip surface are mostly convex polygons, QueXian_Outline[r,i r ,j r ], there are k points in total, from QueXian_Outline[r,i r ,j r ] points to points other than Outline[1,2,i,j] and the end point, and k-1 triangles can be formed; for each triangle, calculate the lengths of the three sides: a, b, c, and the average perimeter p=(a+b+c) / 2, according to the formula: triangle area s = p(pa)(pb)(pc) to the 1 / 2 power; Count the areas of all k-1 triangles to get the defect area S, and count the number of defects Num = Num+1 and the sum of the defect areas S sum = S sum + S; After completing the above steps S31-34, find the points whose pixel values ​​r, g, and b are not zero in the three-dimensional array Image array again, and execute steps S31-34 until all pixel points in the Image array are 1, and finally obtain the number Num of all defects on the wood chip and the sum Ssum of all defect areas.

4. The method for automatically identifying surface defects of wood chips according to claim 1, characterized in that: The S36 is specifically as follows: preset the critical defect area S of the A-level plate a , its value range is: 0.0001-0.2 square meters; B-grade plate critical defect area S b , its value range is: 0.0001-0.2 square meters; When the defect area S on the wood chip sum < S a It is judged as Grade A board; when the defect area S on the wood chip a sum < S b It is judged as Class B board; when the defect area S on the wood chip sum >S b It is judged to be a C-grade board.​

Images