Wood foreign matter space positioning system based on X-ray detection

Abstract

The invention discloses a wood foreign matter space positioning system based on X-ray detection, which relates to the technical field of wood detection and comprises a first X-ray light source, a second X-ray light source, a first detector, a second detector and a data acquisition board electrically connected with the first detector and the second detector. According to the method, the related imaging principle of X-rays is applied, a linear array scanning imaging system is adopted to carry out online detection on the wood plate blank passing through a channel, foreign matters in the wood plate blank can be rapidly detected through an AI algorithm, automatic analysis and calculation are carried out through respective imaging results of two angles, and the spatial position of the foreign matters on the wood plate blank is determined; the system adopts two X-ray emitters and two high-efficiency detectors, the accuracy is improved through cross validation and distortion correction of the double detectors, equipment can be guided to avoid foreign matters, safety is guaranteed, processing waste products can be reduced, and labor and equipment maintenance cost is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wood detection, in particular to a wood foreign matter space positioning system based on X-ray detection. BACKGROUND

[0002] Wood has been listed as an important raw material since ancient times, and the wood processing industry is one of the manufacturing industries with rapid development and good prospects in China. Wood processing mainly uses mechanical or chemical methods to process wood raw materials. During the processing, metal, jujube and other foreign matters are generated to varying degrees. The existence of these foreign matters will affect the quality of the board, and during the later processing of wood products, the board will be unable to be used, resulting in board returns, claims, etc., which will damage the reputation of the wood processing factory and even lose a large number of cooperation opportunities. The traditional detection technology can only detect whether there are foreign matters in the wood board blank, but cannot confirm the spatial position calibration of the foreign matter on the parallel plane of the object plane. To solve this problem, the present application provides a wood foreign matter space positioning system based on X-ray detection. SUMMARY

[0003] To solve the above technical problems, a wood foreign matter space positioning system based on X-ray detection is provided. The technical solution comprises a double light source, a double detector, and two angle imaging of vertical direct and 45-degree oblique. The system analyzes the wood board blank according to the detection signal, generates an image, and uses a corresponding algorithm to obtain the spatial position of the foreign matter in the wood board blank, thereby solving the problem of being unable to determine the spatial position calibration of the foreign matter on the parallel plane of the object plane.

[0004] To achieve the above purpose, the technical solution adopted by the present application is as follows: a wood foreign matter space positioning system based on X-ray detection, comprising an object table and a positioning structure, the positioning structure comprising: a first X-ray light source, a second X-ray light source, a first detector, a second detector, and a data acquisition board electrically connected with the first detector and the second detector. The first X-ray light source is installed directly above the object table, and the center line of the light source generated thereby is perpendicular to the object plane of the object table. The first X-ray light source is a fan-shaped ray beam. The fan-shaped ray beam passes through the wood board blank and the object table and is irradiated onto the first detector to obtain a first detection signal, which is received by the data acquisition board to obtain a first X-ray imaging coordinate system. The picture distortion is corrected and optimized by a computer vision algorithm to obtain a real first image coordinate system, which is the position calibration of the foreign matter in the wood on the parallel plane of the object table. The second X-ray light source is installed obliquely above the stage, and the center line of the light source generated by the second X-ray light source forms an angle of 45° with the width direction of the stage, and the X-ray formed is the fan-shaped beam, which is irradiated to the second detector through the wood blank and the stage, and a second detection signal is obtained, which is received by the data acquisition board, and the picture distortion is corrected and optimized by the computer vision algorithm, and a second image coordinate system is obtained, that is, the relative position calibration of the foreign matter at the 45° angle of the parallel plane of the stage; the first detector and the second detector are located below the stage. The image coordinate system is displayed, enlarged and AI analyzed and calculated by the upper computer software, and a digital image coordinate system is obtained, that is, the final spatial position calibration of the foreign matter on the parallel plane of the stage.

[0005] Preferably, the first X-ray light source and the second X-ray light source are line scanning X-ray detectors.

[0006] Preferably, the imaging of the same feature point of the first X-ray light source and the second X-ray light source in the same plane is on a line, and it is assumed that the first X-ray light source and the second X-ray light source emit light to the foreign matter at the same time, the imaging surface of the first detector and the second detector is the Y-axis reference surface, and the images of the same feature point (X z , Y z ) of the foreign matter under the left light source and the right light source are (X l , 0) and (X r , 0) respectively.

[0007] Preferably, it is known that the first detector and the second detector are in the same plane, the height of the left light source is H, the height of the right light source is H+a, the distance between the left and right light sources is B1, and the imaging point distance is B2, and based on the geometric relationship of a triangle, the following can be obtained: H / Y z =X l / (X l -X z ) (H+a) / Y z =(X r -B1) / (X r -X z ) The position X z and Y z of the feature point of the foreign matter are calculated by the formula, so that any point of the image signal received on the first detector can find a corresponding matching point on the imaging image of the second detector to determine the three-dimensional coordinates of the point, and the spatial position of the foreign matter is obtained.

[0008] Preferably, the stage is a channel conveyor belt for conveying wood blanks.

[0009] Preferably, the first X-ray light source and the second X-ray light source are both high-energy fan-shaped beams, the fan angle is matched with the maximum width of the wood blank and the effective detection size of the first detector, and the energy range is 50-160kV.

[0010] Preferably, the first detector and the second detector both adopt high-resolution linear array X-ray detectors, the pixel density is greater than 100μm / pixel, the pixel arrangement direction is perpendicular to the running direction of the conveying belt, and one-dimensional cross-sectional X-ray attenuation signals of the wood blank are obtained through single scanning; the first detector and the second detector are connected with the data acquisition board through a high-speed differential transmission interface.

[0011] Preferably, the computer vision algorithm obtains the internal and external parameters of the imaging system by adopting Zhang's calibration method based on a standard calibration board, calculates the distortion coefficient, and corrects the distortion and optimizes the contrast of the original image to obtain the real first image coordinate system.

[0012] Compared with the prior art, the wood foreign matter space positioning system based on X-ray detection has the following beneficial effects: The wood foreign matter space positioning system based on X-ray detection can accurately calibrate the length and width coordinates of the foreign matter through the vertical light source, position the thickness and depth through the 45° oblique light source combined with the algorithm, and output the three-dimensional coordinates through the double-viewing-angle cooperation, so that the three-dimensional positioning difficulty is solved, the continuous production is adapted, the conveying belt does not stop running, the line scanning detection is synchronized with the production, the data processing is fast, the efficiency loss is reduced, the accuracy is improved through the cross verification and distortion correction of the double detectors, the equipment can be guided to avoid the foreign matter, the safety is ensured, the processing waste, the labor cost and the equipment maintenance cost are reduced, the system is suitable for various types of wood and foreign matter, and can be widely applied in the field of furniture manufacturing. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 It is a positioning operation schematic diagram of the present application; Figure 2 It is a positioning operation frame schematic diagram of the present application; Figure 3 It is a positioning system frame diagram of the present application; Figure 4 It is an X-ray detection schematic diagram of the present application. DETAILED DESCRIPTION

[0014] The following description is used to disclose the present application so that those skilled in the art can implement the present application. The preferred embodiments in the following description are only used as examples, and other obvious modifications can be thought of by those skilled in the art.

[0015] Referring to Figures 1-3 As shown in the figure, a wood foreign matter space positioning system based on X-ray detection includes a support table and a positioning structure, the positioning structure includes a first X-ray light source, a second X-ray light source, a first detector, a second detector, and a data acquisition board electrically connected with the first detector and the second detector. The first X-ray light source is installed directly above the object table, and the light source center line generated thereby is perpendicular to the object plane of the object table, the first X-ray light source is a fan-shaped ray beam, the fan-shaped ray beam is irradiated to the first detector through the wood blank and the object table to obtain a first detection signal, the first detection signal is received by a data acquisition board to obtain a first X-ray imaging coordinate system, and a picture distortion correction optimization is performed on the first X-ray imaging coordinate system through a computer vision field algorithm to obtain a real first image coordinate system, that is, a position calibration of the foreign matter in the wood on a plane parallel to the object table; The second X-ray light source is installed obliquely above the object table, and the light source center line generated thereby forms a 45° angle with the width direction of the object table, an X-ray formed thereby is a fan-shaped ray beam, the fan-shaped ray beam is irradiated to the second detector through the wood blank and the object table to obtain a second detection signal, the second detection signal is received by the data acquisition board, and a picture distortion correction optimization is performed on the second image coordinate system through a computer vision field algorithm to obtain a second image coordinate system, that is, a relative position calibration of the foreign matter on the 45° angle parallel plane of the object table; the first detector and the second detector are both located below the object table; The image coordinate system is displayed, enlarged and AI analyzed and calculated through the upper computer software to obtain a digital image coordinate system, that is, a spatial position calibration of the foreign matter on the parallel plane of the object table.

[0016] The application applies the related imaging principle of X-rays, adopts a linear array scanning imaging system to perform online detection on the wood blank passing through the channel, can quickly detect the foreign matter in the wood blank through an AI algorithm, and performs automatic analysis and calculation on the respective imaging results of two angles to determine the spatial position of the foreign matter on the wood blank, the system adopts two X-ray emitters and two high-efficiency detectors, one of the X-ray emitters is installed above the wood blank, the X-rays are perpendicular to the wood blank, the other X-ray emitter forms a 45° angle with the detection channel; the two detectors are installed side by side and parallel to the channel, in this way, in the case that the origin and the coordinate system of the wood blank are set, the X-ray image of the X-axis movement direction of the wood blank and the X-ray image of the Y-axis thickness direction can be obtained at the same time, as the detected wood blank is uniformly sent into the channel by the channel conveying belt, the system sequentially scans the wood blank, and through picture optimization and distortion correction algorithms, two X-ray pictures of the foreign matter position of two angles are obtained, and the system calculates the spatial coordinate position of the foreign matter.

[0017] The computer vision field algorithm includes: Camera calibration and distortion model correction are based on a pinhole camera model, the intrinsic parameters and distortion parameters of the camera are calculated in advance through Zhang Zhengyou calibration method, and then the image pixels are inversely mapped and corrected through a formula, which is especially suitable for radial and tangential distortion caused by pixel arrangement errors of the detector or collimation deviation of the light source; For complex nonlinear distortion, the deep learning correction algorithm can directly learn the mapping relationship between the distorted image and the corrected image using the CNN network. By labeling a large number of pairs of distorted / real images to train the model, an end-to-end high-precision correction is realized, which is especially suitable for scenarios that are difficult to model using traditional geometric models. Gray correction and non-uniformity correction: For the uneven brightness of the image caused by the inconsistent response of the X-ray detector, the non-uniformity of the detector itself is eliminated through flat field correction, or the local contrast is enhanced using adaptive histogram equalization to improve the accuracy of subsequent feature extraction.

[0018] Reference Figure 4 As shown in the figure, the first X-ray light source and the second X-ray light source are line-scan X-ray detection.

[0019] The imaging of the same feature point of the first X-ray light source and the second X-ray light source in the same plane is on a line. Assuming that the first X-ray light source and the second X-ray light source emit light to the foreign object at the same time, the imaging planes of the first detector and the second detector are the Y-axis reference planes, and the images of the same feature point (X z , Y z ) of the foreign object under the left light source and the right light source are (X l , 0) and (X r , 0), respectively.

[0020] The present application adopts a line-scan X-ray light source, which can realize dynamic detection in combination with the continuous movement of the channel conveyor belt, complete the comprehensive scanning of the wood board without pausing the conveying, synchronize the detection speed with the production rhythm, avoid the problem of "stopping piece by piece and low efficiency" of traditional face-scan detection, and adapt to the continuous needs of wood processing; The limitation of the double light source to be collinear in the same plane and at the same feature point can eliminate the positioning error caused by the plane deviation in the double-view detection. At the same time, the correspondence between the feature point of the foreign object and the imaging point of the double detector is clear, which provides accurate geometric reference for subsequent triangular geometric calculation and avoids the coordinate calculation deviation caused by chaotic imaging position; The correspondence between the feature point of the foreign object (X z , Y z ) and the imaging point of the double detector (X l , 0) (X r , 0) is defined in advance, and the imaging plane of the detector is set as the Y-axis reference plane, so that the coordinate formula can be derived directly based on the known geometric parameters, without additional processing of complex coordinate conversion, reducing the data processing steps and improving the real-time and accuracy of spatial positioning.

[0021] Given that the first detector and the second detector are in the same plane, the height of the left light source is H, the height of the right light source is H+a, the distance between the left and right is B1, and the imaging point distance is B2, based on the triangular geometric relationship, we can get: H / Y z =X l / (X l -X z ) (H+a) / Y z =(X r -B1) / (X r -X z ) The position X of the foreign matter feature point is calculated by a formula z and Y z , so that any point receiving an image signal on the first detector finds a corresponding matching point on the imaging diagram of the second detector to determine the three-dimensional coordinates of the point, thereby obtaining the spatial position of the foreign matter positioning.

[0022] The application establishes a mathematical correlation between the light source height, the detector spacing, and the imaging point coordinates, converts the abstract foreign matter position into calculable (X z , Y z ) coordinates, breaks through the limitations of traditional qualitative detection, upgrades the foreign matter positioning from rough area judgment to accurate coordinate output, and controls the positioning error to be within millimeters; the direct derivation based on triangular geometry avoids complex spatial coordinate conversion, uses the geometric constraints of two light sources and detectors, solves the three-dimensional coordinates through a simple binary linear equation system, reduces the data processing operation amount, ensures real-time performance, adapts to the high-speed detection needs of the wood production line; the corresponding rules of the imaging point of the first detector and the matching point of the second detector are clear, the matching accuracy is verified by geometric formulas, the positioning deviation caused by feature point mismatching is avoided, a numerical verification basis is provided for double-view image fusion, and the reliability of three-dimensional positioning is improved.

[0023] The object table is a channel conveyor belt for transmitting wood blanks; to improve the detection efficiency.

[0024] The first X-ray source and the second X-ray source are both high-energy fan-shaped ray beams, and the fan angle is matched with the maximum width of the wood blank and the effective detection size of the first detector, and the ray energy range is 50-160kV.

[0025] The first detector and the second detector both adopt high-resolution linear array X-ray detectors, the pixel density is greater than 100μm / pixel, the pixel arrangement direction is perpendicular to the running direction of the conveyor belt, and one-dimensional cross-sectional ray attenuation signals of the blank are obtained by single scanning; the first detector and the second detector are connected with the data acquisition board through a high-speed differential transmission interface.

[0026] The computer vision field algorithm obtains the intrinsic and extrinsic parameters of the imaging system by adopting Zhang's calibration method based on a standard calibration board, calculates the distortion coefficient, and performs distortion correction and contrast optimization on the original image to obtain the real first image coordinate system.

[0027] The foregoing merely illustrates the principles of the application and application of its more prominent features. Those skilled in the art will appreciate that the application is not limited to the embodiments described and illustrated and that various modifications and improvements can be made thereto without departing from the spirit and scope of the application. The scope of the application is delimited by the appended claims and their equivalents.

Claims

1. A spatial positioning system for foreign objects in wood based on X-ray detection, comprising a stage and a positioning structure, characterized in that, The positioning structure includes: a first X-ray source, a second X-ray source, a first detector, a second detector, and a data acquisition board electrically connected to the first detector and the second detector; The first X-ray source is installed directly above the stage, and the center line of the source is perpendicular to the surface of the stage. The first X-ray source is a fan-shaped beam. The fan-shaped beam passes through the wood board and the stage and irradiates the first detector to obtain a first detection signal. The signal is received by the data acquisition board to obtain the first X-ray imaging coordinate system. The image distortion is corrected and optimized by computer vision algorithms to obtain the true first image coordinate system, which is the position of the foreign object in the wood on the plane parallel to the stage. The second X-ray source is installed diagonally above the stage, with its center line forming a 45° angle with the width of the stage. The resulting X-rays are fan-shaped beams that pass through the wood blank and the stage before illuminating the second detector, generating a second detection signal. This signal is received by the data acquisition board, and image distortion is corrected and optimized using computer vision algorithms to obtain a second image coordinate system. This coordinate system represents the relative position of the foreign object at a 45° angle on the parallel plane of the stage. Both the first and second detectors are located below the stage. The image coordinate system is displayed, magnified, and analyzed and calculated by the host computer software to obtain the digital image coordinate system, which is the final spatial position calibration of the foreign object on the parallel plane of the stage.

2. The spatial positioning system for foreign objects in wood based on X-ray detection according to claim 1, characterized in that: The first and second X-ray sources are line scan X-ray detectors.

3. The spatial positioning system for foreign objects in wood based on X-ray detection according to claim 1, characterized in that: The images of the first X-ray source and the second X-ray source at the same feature point in the same plane are projected onto a line. Assuming the first and second X-ray sources emit light onto the foreign object simultaneously, and the imaging planes of the first and second detectors are taken as the Y-axis reference plane, the images at the same feature point on the foreign object (X...) are projected onto a line. z Y z The images under the left and right light sources are (X) l ,0) and (X) r ,0).

4. The spatial positioning system for foreign objects in wood based on X-ray detection according to claim 3, characterized in that: Given that the first detector and the second detector are on the same plane, the height of the left light source is H, the height of the right light source is H+a, the distance between the left and right light sources is B1, and the distance between the imaging points is B2, based on trigonometric relationships, we can obtain: H / Y z =X l / (X l -X z ) (H+a) / Y z =(X r -B1) / (X r -X z ) The location X of the foreign object's characteristic point is calculated using a formula. z With Y z Therefore, any point on the first detector that receives the image signal can find a corresponding matching point on the second detector's image map to determine the three-dimensional coordinates of that point and obtain the spatial location of the foreign object.

5. A spatial positioning system for foreign objects in wood based on X-ray detection according to claim 1, characterized in that: The platform is a conveyor belt used to transport timber boards.

6. The spatial positioning system for foreign objects in wood based on X-ray detection according to claim 1, characterized in that: Both the first and second X-ray sources are high-energy fan-shaped beams, with the fan angle matching the maximum width of the wood board and the effective detection size of the first detector. The radiation energy range is 50-160kV.

7. A spatial positioning system for foreign objects in wood based on X-ray detection according to claim 1, characterized in that: Both the first and second detectors are high-resolution linear X-ray detectors with a pixel density greater than 100 μm / pixel. The pixel arrangement direction is perpendicular to the conveyor belt running direction. A single scan acquires the one-dimensional cross-sectional X-ray attenuation signal of the slab. Both the first and second detectors are connected to the data acquisition board through a high-speed differential transmission interface.

8. A spatial positioning system for foreign objects in wood based on X-ray detection according to claim 1, characterized in that: In the field of computer vision, algorithms use Zhang's calibration method based on a standard calibration board to obtain the intrinsic and extrinsic parameters of the imaging system and calculate the distortion coefficients. The original image is then subjected to distortion correction and contrast optimization to obtain the true first image coordinate system.

Images