Automatic shape follow-up coring method for dalbergia odorifera based on image recognition

Abstract

The invention provides a dalbergia odorifera automatic shape follow-up coring method based on image recognition. The method comprises the steps that S1, wood to be processed is fixed to a multi-axis processing platform; s2, the length and contour data of the to-be-processed wood are obtained through three-dimensional laser scanning, and the actual axis of the to-be-processed wood is obtained according to the contour data; s3, obtaining an image of a to-be-processed end face of the to-be-processed wood through the camera, and obtaining outer contour line data and inner contour line data of the outer cladding according to the image; s4, the outer contour line data and the inner contour line data serve as machining paths of the rotary cutting tool on the preset feeding distance, and rotary cutting is conducted along the machining paths with the actual axis as the machining axis; and the step S3 and the step S4 are repeated until the accumulated preset feeding distance reaches the length of the wood to be processed. According to the method, the problem that the current precious wood shape follow-up coring operation excessively depends on manual participation can be solved, and automation of wood coring operation is achieved based on the image scanning and recognition technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic wood processing, and particularly relates to a method for automatically taking cores of Huanghua pear based on image recognition. BACKGROUND

[0002] The core material of Huanghua pear is dense, has beautiful texture, unique fragrance and corrosion resistance, and is the only part with high economic value and practical value, while the edge material (white wood) is loose and easy to rot, and has no use value.

[0003] At present, the core material operation of Huanghua pear completely relies on manual operation, and each step from material selection depends on the long-term experience of artisans, the skilled and accurate use of tools, visual fatigue and physical consumption, which finally leads to low efficiency and great waste of core material. SUMMARY

[0004] The present application aims to provide a method for automatically taking cores of Huanghua pear based on image recognition, solve the problem that the core taking operation of precious wood excessively relies on manual operation, and realize the automation of wood core taking operation based on image scanning and recognition technology.

[0005] The above technical purposes of the present application are mainly realized by the following technical solutions.

[0006] On one hand, the present application provides a method for automatically taking cores of Huanghua pear based on image recognition, which comprises the following steps:

[0007] Step S1: fixing the wood to be processed with an outer cladding layer and a core layer on a rotating shaft of a multi-axis processing platform;

[0008] Step S2: obtaining the length and contour data of the wood to be processed by stereoscopic laser scanning, and obtaining the actual axis of the wood to be processed according to the contour data;

[0009] Step S3: obtaining the image of the end face to be processed of the wood to be processed by a camera, and obtaining the outer contour line data of the outer cladding layer and the inner contour line data of the junction of the outer cladding layer and the core layer according to the image;

[0010] Step S4: taking the outer contour line data and the inner contour line data as the processing path of a rotating cutting tool at a predetermined feed distance, and rotating and cutting the outer cladding layer along the processing path with the actual axis as the processing axis;

[0011] Repeating the above step S3 and step S4 until the accumulated predetermined feed distance reaches the length of the wood to be processed, and completing the cutting of the outer cladding layer.

[0012] In a preferred embodiment of the present application, the predetermined feeding distance is a length along the axial direction of the rotating shaft, and the predetermined feeding distance is 0.01mm-10mm.

[0013] In a preferred embodiment of the present application, the step S2 comprises the following steps:

[0014] Step S21: vertically projecting a straight laser onto the surface of the wood to be processed by using a linear laser;

[0015] Step S22: rotating the wood to be processed by the rotating shaft for one round, and acquiring the image by using the camera to capture the curve formed by the straight laser projected onto the surface of the wood to be processed in real time, so as to obtain the contour data of the wood to be processed;

[0016] Step S23: establishing a contour data model corresponding to the wood to be processed according to the obtained contour data, and obtaining the length and the actual axis according to the contour data model.

[0017] In a preferred embodiment of the present application, the step S3 comprises the following steps:

[0018] Step S31: rotating the wood to be processed by the rotating shaft for one round, and capturing the image on the upper side of the end face to be processed by the camera in real time;

[0019] Step S32: synthesizing the complete image of the end face to be processed according to the images on the upper side of the end face to be processed at different rotating angles;

[0020] Step S33: performing the grayscale, noise reduction and binarization processing on the complete image, so as to obtain the binary image data;

[0021] Step S34: extracting the outer contour line data and the inner contour line data from the binary image data;

[0022] Step S35: performing the polygon approximation or spline curve fitting on the extracted outer contour line data and inner contour line data, so as to convert them into the vector contour line data composed of straight lines and circular arc lines;

[0023] Step S36: performing the feature recognition and process planning on the vector contour line data according to the preset processing rule, so as to generate the processing path.

[0024] In a preferred embodiment of the present application, in the step S3, a radial machining feed amount is set so that when the overcladding layer is rotationally cut, the outer surface portion of the core layer is cut, and the inner edge of the machining ring surface formed after a predetermined feed distance is completed shows the interface profile line of the overcladding layer and the core layer, and the machining ring surface serves as the image recognition basis for planning the machining path on the next predetermined feed distance.

[0025] In a preferred embodiment of the present application, the multi-axis machining platform has a moving Y-axis and a moving Z-axis perpendicular to the rotation axis, and the rotary cutting tool is arranged on the moving Z-axis.

[0026] The step S4 includes: the rotation axis drives the wood to be machined to rotate, the moving Y-axis and the moving Z-axis follow based on the spatial change of the actual axis of the wood to be machined so that the position of the rotary cutting tool relative to the actual axis remains unchanged, and at the same time, the moving Z-axis moves up and down based on the machining path generated in the step S3 so that the rotary cutting tool performs rotational cutting relative to the wood to be machined.

[0027] On the other hand, the present application also provides a wood contour core taking device for implementing the image recognition-based automatic contour core taking method for Huanghua Li as described above, and the wood contour core taking device is a four-axis machining platform.

[0028] In a preferred embodiment of the present application, the four-axis machining platform includes:

[0029] a base;

[0030] a first moving platform arranged on the base and capable of moving along the Y-axis direction under the drive of a first driving assembly;

[0031] a clamping and rotating assembly fixedly installed on the first moving platform, used for clamping the wood to be machined and capable of rotating around the X-axis direction under the drive of a second driving assembly;

[0032] a second moving platform arranged on the base and capable of moving along the X-axis direction under the drive of a third driving assembly;

[0033] a rotary cutting tool arranged on the second moving platform and capable of moving along the Z-axis direction under the drive of a fourth driving assembly;

[0034] a scanning assembly fixedly installed on the second moving platform, used for performing three-dimensional laser scanning on the wood to be machined;

[0035] A first camera is fixedly installed on the second moving platform and used to capture an image of the end face of the wood to be processed;

[0036] A control device is electrically connected with the first driving assembly, the second driving assembly, the third driving assembly, the fourth driving assembly, the scanning assembly and the first camera, and is configured to control the first moving platform, the clamping and rotating assembly, the second moving platform and the rotating cutting tool to act according to the scanning result of the scanning assembly and the capturing result of the first camera, so as to cut off the outer layer of the wood to be processed layer by layer along the X-axis direction.

[0037] The X-axis, the Y-axis and the Z-axis are three axes perpendicular to each other in a space coordinate system.

[0038] In a preferred embodiment of the present application, the image capturing direction of the first camera is arranged downwardly along the X-axis direction and faces the contact processing area of the rotating cutting tool and the wood to be processed, and the included angle between the image capturing direction and the X-axis direction is 45 degrees.

[0039] In a preferred embodiment of the present application, the scanning assembly has a linear laser and a second camera, both of which are fixedly installed on the second moving platform, the laser projection direction of the linear laser intersects with the image capturing direction of the second camera, and the included angle between the laser projection direction and the image capturing direction is 26 degrees.

[0040] Compared with the prior art, the technical solution of the present application has the following characteristics and advantages:

[0041] The automatic following shape coring method of Huanghua pear based on image recognition in the present application automatically generates a cutting tool machining path according to the surface contour features and line contour features of wood based on the results of image scanning and shooting analysis, replaces the current wood coring operation requiring manual participation with automatic operation of a mechanical machining platform, and has high accuracy, low waste rate and high efficiency, thereby realizing a qualitative leap in wood coring operation. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor. In the drawings:

[0043] The drawings described herein are for purposes of illustration only and are not intended to limit the scope of the present disclosure in any way. Additionally, the shapes and relative sizes of the various components in the drawings are exaggerated for clarity and are not necessarily drawn to scale. One skilled in the art will recognize that the various components in the drawings can be combined, divided, re-arranged, and / or otherwise altered.

[0044] Figure 1 Flow chart of the automatic core taking method along the shape of Huanghua Li based on image recognition according to the present disclosure;

[0045] Figure 2 Flow chart of step S2 in the automatic core taking method along the shape of Huanghua Li based on image recognition according to the present disclosure;

[0046] Figure 3 Flow chart of step S3 in the automatic core taking method along the shape of Huanghua Li based on image recognition according to the present disclosure;

[0047] Figure 4 Model diagram of the wood to be processed according to the present disclosure;

[0048] Figure 5 Cross-sectional structure diagram of the wood to be processed according to the present disclosure;

[0049] Figure 6 Figure 5 Enlarged diagram of part A in FIG. 1;

[0050] Figure 7 Axonometric view of the end surface to be processed of the wood to be processed after preliminary processing according to the present disclosure;

[0051] Figure 8 Diagram of the end surface to be processed of the wood to be processed from the perspective of the camera according to the present disclosure;

[0052] Figure 9 Figure 8 Enlarged diagram of part B in FIG. 1;

[0053] Figure 10 Structural diagram of the wood core taking device according to the present disclosure.

[0054] Legend of reference signs:

[0055] 10, core layer; 11, outer cladding layer; 12, end surface to be processed; 13, outer contour line; 14, inner contour line;

[0056] 20, base; 21, first moving platform; 22, first driving assembly; 23, clamping and rotating assembly; 24, second driving assembly; 25, second moving platform; 26, third driving assembly; 27, rotating cutting tool; 28, fourth driving assembly;​​

[0057] 30, first camera; 31, linear laser; 32, second camera. DETAILED DESCRIPTION

[0058] In order to make the personnel in the technical field better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work should belong to the scope of protection of the present application.

[0059] It should be noted that when an element is referred to as being "disposed on" another element, it can be directly on the other element or there can be an intervening element. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or there can be an intervening element. The terms "vertical", "horizontal", "left", "right", and similar expressions used herein are for illustrative purposes only and are not intended to be limiting.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0061] Embodiment one:

[0062] The present application provides an automatic following core taking method for huanghuali based on image recognition, as shown in the figure, which comprises the following steps: Figure 1 As shown in the figure, which comprises the following steps:

[0063] Step S1: fixing the wood to be processed with an outer cladding layer 11 and a core layer 10 on the rotating shaft of the multi-axis processing platform;

[0064] Step S2: obtaining the length and contour data of the wood to be processed by stereoscopic laser scanning, and obtaining the actual axis of the wood to be processed according to the contour data;

[0065] Step S3: obtaining the image of the end face 12 to be processed of the wood to be processed by a camera, and obtaining the outer contour line data of the outer cladding layer 11 and the inner contour line data of the junction of the outer cladding layer 11 and the core layer 10 according to the image;

[0066] Step S4: taking the outer contour line data and the inner contour line data as a machining path of the rotary cutting tool 27 on a predetermined feed distance, and performing rotary cutting on the outer cladding layer 11 along the machining path with the actual axis as the machining center;

[0067] Steps S3 and S4 are repeated until the accumulated predetermined feed distance reaches the length of the wood to be machined, and the cutting of the outer cladding layer 11 is completed.

[0068] The image recognition-based automatic follow-up core taking method for huanghuali wood provided by the present application automatically generates a tool cutting machining path according to the surface contour features and line contour features of the wood based on the results of image scanning and shooting analysis, replacing the current manual wood core taking operation with automatic operation of a mechanical machining platform, which is accurate, low-waste, and efficient, thereby realizing a qualitative leap in wood core taking operations.

[0069] The specific process of each step of the image recognition-based automatic follow-up core taking method for huanghuali wood provided by the present application will be described in detail below.

[0070] In step S1, a multi-axis machining platform is used to perform core taking operations on the wood to be machined.

[0071] Specifically, in the present embodiment, the multi-axis machining platform can be a four-axis machining platform, which has three moving axes and one rotating axis, where the three moving axes are a moving X axis, a moving Y axis, and a moving Z axis, and the rotating axis is capable of rotating around the moving X axis direction. A clamp is provided on the rotating axis, and a rotary cutting tool 27 is installed on the moving Z axis; before starting machining, one end of the wood to be machined is fixed on the clamp of the rotating axis. Under the coordinated action of the multi-axis, the wood to be machined can move up and down, left and right, forward and backward, and rotate relative to the rotary cutting tool 27. Among them, as shown in Figure 4 the wood to be machined is a huanghuali log with an irregular shape, which has an outer cladding layer 11 that is white and a core layer 10 that is red, and in order to obtain the core layer 10 with higher value, the outer cladding layer 11 needs to be removed.

[0072] In step S2, a three-dimensional laser scanner is used to perform three-dimensional laser scanning on the fixed wood to be machined to obtain part of the features of the wood to be machined.

[0073] Specifically, as shown in Figure 2 , step S2 includes the following steps:

[0074] Step S21: a linear laser is projected vertically onto the surface of the wood to be machined using a linear laser 31, and the linear laser appears as a curved line due to the unevenness of the surface of the wood to be machined.

[0075] Step S22: The rotating shaft drives the wood to be processed to rotate one revolution. At the same time, the camera is used to acquire the curve formed by the linear laser beam projected onto the surface of the wood to be processed in real time to obtain the contour data of the wood to be processed.

[0076] Step S23: Establish a contour data model corresponding to the wood to be processed based on the obtained contour data, and obtain the length and actual axis of the wood to be processed based on the contour data model.

[0077] Based on the material data of the wood to be processed and the contour data obtained by the scanning component, a contour data model of the wood to be processed is established, and the length and actual axis of the wood to be processed are obtained from this data model. The length of the wood to be processed is the length along the X-axis of movement, which is also the length of the wood to be processed along the rotation axis. The actual axis is the direction extending from the actual axis of the wood to be processed, and it is related to the shape of the wood to be processed. Since the wood to be processed is usually not cylindrical, its actual axis does not coincide with the axis of rotation.

[0078] In step S3, a camera is used in conjunction with image recognition technology to obtain some features on the end face 12 of the wood to be processed, and thereby generate a processing path for the rotary cutting tool 27 at a predetermined feed distance.

[0079] Specifically, such as Figure 3 As shown, step S3 includes the following steps:

[0080] Step S31: The rotating shaft drives the wood to be processed to rotate one revolution, while the camera captures a real-time image of the upper side of the end face 12 to be processed. One end of the wood to be processed is fixed on the clamp, and the camera is set above the other end of the wood to be processed and tilted downwards.

[0081] Step S32: Synthesize a complete image of the end face 12 to be processed based on multiple images of its upper side taken at different rotation angles. For example... Figure 4 to Figure 9 As shown, before the start of processing, the end face of the outer cladding layer 11 and the end face of the core layer 10 are in the same plane, at which time it is relatively easy to obtain an image of the entire end face 12 to be processed; after the rotary cutting tool 27 rotates and travels at least a predetermined feed distance along the moving X-axis, the end face of the outer cladding layer 11 and the end face of the core layer 10 are no longer in the same plane, and the exposed core layer 10 may obstruct the end face of the outer cladding layer 11. Therefore, multiple images are taken during the rotation of the wood to be processed by taking images at a downward tilting shooting angle, and a complete image of the end face 12 to be processed is synthesized by image compositing technology.

[0082] Step S33: Perform grayscale conversion, noise reduction, and binarization on the synthesized complete image to obtain binary image data.

[0083] Step S34: Extract outer contour data and inner contour data from the binary image data. The outer contour data represents the outer contour line 13 of the outer cladding layer 11 on the end face 12 to be processed; the inner contour data represents the boundary line between the outer cladding layer 11 and the core layer 10 on the end face 12 to be processed, that is, the inner contour line 14 of the outer cladding layer 11 on the end face 12 to be processed.

[0084] Step S35: Perform polygon approximation or spline curve fitting on the extracted outer contour data and inner contour data to convert them into vector contour data composed of straight lines and arcs.

[0085] Step S36: Based on the preset processing rules (grayscale level parameters, continuous gap parameters, longitudinal discontinuity parameters, etc.), perform feature recognition and process planning on the vector contour data to generate a processing path.

[0086] Furthermore, in step S36, the generated machining path is provided with a radial machining feed amount so that when the outer cladding layer 11 is rotated and removed, the outer surface portion of the core layer 10 is removed. After completing the predetermined feed distance, the inner edge of the machining torus formed shows the boundary contour line between the outer cladding layer 11 and the core layer 10. The machining torus serves as the image recognition basis for planning the machining path at the next predetermined feed distance.

[0087] In step S4, based on the identification results in steps S2 and S3, the rotary cutting tool 27 completes the removal of the outer cladding layer 11 over a predetermined feed distance.

[0088] Specifically, the rotating axis drives the wood to be processed to rotate, and the moving Y-axis and moving Z-axis follow the spatial changes of the actual axis of the wood to be processed at the end face to be processed, so that the position of the rotating cutting tool 27 relative to the actual axis of the wood to be processed at the end face 12 to be processed remains unchanged. At the same time, the moving Z-axis moves up and down based on the processing path generated in step S3, so that the rotating cutting tool 27 rotates and cuts relative to the wood to be processed until a predetermined feed distance is completed.

[0089] Preferably, the predetermined feed distance is the length along the axial direction of the rotation axis, and the predetermined feed distance is 0.01mm-10mm.

[0090] Repeat steps S3 and S4 above, as follows: Figure 6 As shown, the outer layer 11 of the wood to be processed is removed layer by layer along the direction of the moving X-axis until all outer layers 11 of the wood to be processed are removed in the length direction.

[0091] Implementation Method Two:

[0092] The application further provides a wood contour core taking device for implementing the image recognition-based automatic contour core taking method of Huanghua Li as described in the first embodiment, and the wood contour core taking device is a four-axis machining platform.

[0093] As shown in the drawings, Figure 10 The four-axis machining platform comprises a base 20, a first moving platform 21 arranged on the base 20 and capable of moving along a Y-axis direction under the drive of a first driving assembly 22, a clamping and rotating assembly 23 fixedly installed on the first moving platform 21 and used for clamping a wood to be processed and capable of rotating around an X-axis direction under the drive of a second driving assembly 24, a second moving platform 25 arranged on the base 20 and capable of moving along an X-axis direction under the drive of a third driving assembly 26, a rotating cutting tool 27 arranged on the second moving platform 25 and capable of moving along a Z-axis direction under the drive of a fourth driving assembly 28, a scanning assembly fixedly installed on the second moving platform 25 and used for performing three-dimensional laser scanning on the wood to be processed, a first camera 30 fixedly installed on the second moving platform 25 and used for shooting an image of a processed end surface 12 of the wood to be processed, and a control device electrically connected with the first driving assembly 22, the second driving assembly 24, the third driving assembly 26, the fourth driving assembly 28, the scanning assembly and the first camera 30, and configured to control the first moving platform 21, the clamping and rotating assembly 23, the second moving platform 25 and the rotating cutting tool 27 to act according to the scanning result of the scanning assembly and the shooting result of the first camera 30, so as to remove the outer cover layer 11 of the wood to be processed layer by layer along the X-axis direction, wherein the X-axis, the Y-axis and the Z-axis are three axes perpendicular to each other in a spatial coordinate system, and the first driving assembly 22, the second driving assembly 24, the third driving assembly 26 and the fourth driving assembly 28 are all step motors.

[0094] Further, as shown in the drawings, Figure 10 The image collection direction of the first camera 30 is arranged downwardly along the X-axis direction and faces the contact processing area of the rotating cutting tool 27 and the wood to be processed, and the included angle between the image collection direction and the X-axis direction is 45 degrees.

[0095] Further, as shown in the drawings, Figure 10 The scanning assembly has a linear laser 31 and a second camera 32, both of which are fixedly installed on the second moving platform 25, the laser projection direction of the linear laser 31 intersects with the image collection direction of the second camera 32, and the included angle between the laser projection direction and the image collection direction is 26 degrees.

[0096] The above-described specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above-described specific embodiments are merely examples of the present application and are not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An automatic Huanghua pear wood shape-following coring method based on image recognition, characterized in that, The method comprises the following steps: Step S1: fixing the wood to be processed with an outer layer (11) and a core layer (10) on a rotating shaft of a multi-axis processing platform; Step S2: obtaining length and profile data of the wood to be processed by stereoscopic laser scanning, and obtaining an actual axis of the wood to be processed according to the profile data; Step S3: obtaining an image of a processing end face (12) of the wood to be processed by a camera, and obtaining outer profile line data of the outer layer (11) and inner profile line data of a junction of the outer layer (11) and the core layer (10) according to the image; Step S4: taking the outer profile line data and the inner profile line data as a processing path of a rotating cutting tool (27) at a predetermined feed distance, and performing rotating cutting on the outer layer (11) along the processing path with the actual axis as a processing axis; Repeating the above steps S3 and S4 until the accumulated predetermined feed distance reaches the length of the wood to be processed, and the cutting of the outer layer (11) is completed.

2. The image recognition based automatic rosewood conforming coring method according to claim 1, wherein, The predetermined feed distance is the length in the axial direction of the rotating shaft, and the predetermined feed distance is 0.01mm-10mm.

3. The image recognition based automatic rosewood conforming coring method according to claim 1, wherein, The step S2 comprises the following steps: Step S21: vertically projecting a linear laser onto the surface of the wood to be processed by a linear laser (31); Step S22: rotating the wood to be processed by the rotating shaft for one revolution, and simultaneously acquiring an image of a curve formed by the linear laser projected onto the surface of the wood to be processed by a camera in real time to obtain profile data of the wood to be processed; Step S23: establishing a profile data model corresponding to the wood to be processed according to the obtained profile data, and obtaining the length and the actual axis according to the profile data model.

4. The image recognition based automatic rosewood conforming coring method according to claim 1, wherein, The step S3 comprises the following steps: Step S31: rotating the wood to be processed by the rotating shaft for one revolution, and simultaneously shooting an image of the upper side of the processing end face (12) by the camera in real time; Step S32: synthesizing a complete image of the processing end face (12) according to a plurality of images of the upper side of the processing end face (12) at different rotation angles; Step S33: performing grayscale, noise reduction and binarization processing on the complete image to obtain binary image data; Step S34: extracting the outer profile line data and the inner profile line data from the binary image data; Step S35: performing polygon approximation or spline curve fitting on the extracted outer profile line data and inner profile line data, and converting them into vector profile line data composed of straight lines and circular arc lines; Step S36: performing feature recognition and process planning on the vector profile line data according to a preset processing rule to generate the processing path.

5. The image recognition based automatic rosewood conforming coring method according to claim 1 or 4, characterized in that, In the step S3, a radial machining feed amount is set so that when the overcladding layer (11) is rotationally cut, the outer surface portion of the core layer (10) is cut, and the inner edge of the machining ring surface formed after a predetermined feed distance is completed shows the interface profile line of the overcladding layer (11) and the core layer (10), and the machining ring surface serves as the image recognition basis for planning the machining path on the next predetermined feed distance.

6. The image recognition based automatic rosewood conforming coring method according to claim 5, wherein, The multi-axis machining platform has a moving Y-axis and a moving Z-axis perpendicular to the rotation axis, and the rotary cutting tool (27) is arranged on the moving Z-axis; The step S4 includes: the rotation axis drives the wood to be machined to rotate, the moving Y-axis and the moving Z-axis follow based on the spatial change of the actual axis of the wood to be machined, so that the relative position of the rotary cutting tool (27) to the actual axis remains unchanged, while the moving Z-axis moves up and down based on the machining path generated in the step S3, so that the rotary cutting tool (27) performs rotational cutting on the wood to be machined.

7. The image recognition based automatic rosewood conforming coring method according to claim 3, wherein, The laser projection direction of the linear laser (31) intersects with the image acquisition direction of the camera, and the included angle between the laser projection direction and the image acquisition direction is 26 degrees.

8. The image recognition based automatic rosewood conforming coring method according to claim 1 or 4, characterized in that, The image acquisition direction of the camera is arranged obliquely downward along the axial direction of the rotation axis and towards the contact machining area of the rotary cutting tool (27) and the wood to be machined, and the included angle between the image acquisition direction and the axis of the rotation axis is 45 degrees.

Images