Wood surface treatment detection method and system

Abstract

The invention discloses a wood surface treatment detection method and system, and relates to the technical field of wood detection, and the method comprises the following steps: S1, shooting wood through a high-definition camera to obtain a shot picture, judging whether the surface of the wood has scabs or not through the color and texture in the shot picture, and if the scabs exist, positioning and marking the scabs of the wood; according to the scheme, secondary shooting is carried out on the knocked fast knot, whether the scabs after knocking loosen and displace or not is judged by analyzing and comparing the pictures and the shot pictures, if the scabs loosen and displace due to knocking, the strength of the scabs is insufficient, and therefore it is indicated that the fast knot repairing difficulty is large, and the percent of pass after repairing is low; therefore, the image comparison effect after the slipknot and the hard knot are knocked can be used for conveniently judging whether the scabs can be reused through repair or not subsequently, thereby being beneficial to avoiding material waste and cost increase caused by directly discarding wood of the hard knot.

Description

Technical Field

[0001] This invention relates to the field of wood testing technology, specifically to a method and system for testing wood surface treatment. Background Technology

[0002] Wood surface inspection is a process of quickly identifying and assessing defects such as knots, cracks, decay, wormholes, and discoloration in wood through visual observation, tactile testing, optical instruments, or automated systems. Traditional methods rely on experience-based judgment, while modern technology combines high-resolution cameras, infrared thermal imaging, or laser scanning to accurately analyze texture, density, and structural anomalies, ensuring the quality of wood grading, processing, and use. Wood knots are divided into live knots and dead knots. The key difference between the two lies in their bonding state with the surrounding wood and their functional impact. Live knots are formed when a living branch is wrapped around the tree during its growth. They are lighter in color, close to the original color of the wood, have continuous grain and a tight structure, and are hard. They have little impact on strength and are often used for decorative purposes. Dead knots, on the other hand, are formed when a dead branch is wrapped around the tree. They are dark black or brown in color, have cracks at the edges, and are easy to loosen and fall off, which leads to a decrease in wood strength and difficulty in processing. Currently, the identification of knots on the surface of wood still mainly relies on human experience. The false judgment rate of automated identification through machine vision is relatively high. In the actual process, some wood surfaces may have dead knots but the knot condition is good. The defect problem can be avoided by simple repair work. Directly discarding wood when dead knots are identified not only wastes materials but also increases costs. In order to address the shortcomings of high false judgment rate and lack of intelligent identification in the detection and identification of knots on the surface of wood, we propose a wood surface treatment detection method and system. Summary of the Invention

[0003] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a method and system for detecting wood surface treatment, thereby solving the aforementioned problems in the prior art.

[0004] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: a method for detecting wood surface treatment, comprising the following steps: S1: Take pictures of the wood using a high-definition camera. Determine whether there are knots on the wood surface based on the color and texture of the pictures. If knots are present, locate and mark the position of the knots in the wood. S2: By tapping the knots in the wood, the sound sensor obtains the tapping audio of the knots, and by comparing the photographed image with the tapping audio, it is determined whether the knot in the wood is a dead knot or a live knot. S3: Take a second picture of the wood identified as dead knot using a high-definition camera to obtain a comparison picture. Analyze the comparison picture with the original picture to determine whether the dead knot is loose. If the dead knot is loose, mark the wood as scrap wood. If the dead knot is not loose, proceed to S4. S4: By comparing the location of the dead knot in the picture, the image of the dead knot, and the audio of tapping the knot, determine whether the dead knot can be repaired. If the dead knot can be repaired, fill and wrap the dead knot by injecting epoxy resin into it. If the dead knot cannot be repaired, mark the wood as scrap wood.

[0005] Preferably, the presence of knots on the wood surface is determined by examining the color and texture in photographs. Specifically: S101: Acquire a photographed image, process the photographed image into grayscale to obtain a grayscale photographed image, and capture low-brightness elliptical areas in the grayscale photographed image and mark them as areas to be confirmed. S102: Locate the area to be confirmed in the photographed image, obtain the color of the area to be confirmed in the photographed image, determine whether the color of the area to be confirmed is brown, gray, brown or yellow. If the color of the area to be confirmed is brown, gray, brown or yellow, then execute S103. If the color of the area to be confirmed is not brown, gray, brown or yellow, then mark the area to be confirmed as a stain. S103: Obtain the edge texture of the area to be confirmed. Obtain the texture of the area to be confirmed. Determine whether the edge texture of the area to be confirmed and the texture of the area to be confirmed are consistent and continuous. If the edge texture of the area to be confirmed and the texture of the area to be confirmed are consistent and continuous, then mark the area to be confirmed as a stain. If the edge texture of the area to be confirmed and the texture of the area to be confirmed are not consistent and continuous, then mark the area to be confirmed as a scar.

[0006] Preferably, in S2, the sound sensor obtains the tapping audio of the knot by tapping the knot location of the wood. Specifically, the clamping position is marked on the wood surface, the wood is fixed at the clamping position by a clamp, the tapping distance and tapping force are set, and the same position of the wood knot is tapped three times by a tapping device according to the tapping distance and tapping force to obtain three audio values. The three audio values ​​are summed and the average value is taken to obtain the tapping audio.

[0007] Preferably, in S2, the determination of whether a wood knot is a dead knot or a live knot is made by taking a photograph and listening to the sound of tapping the knot. Specifically: S201: Set the audio judgment range, obtain the tapping audio, and determine whether the tapping audio is within the audio judgment range. If the tapping audio is within the audio judgment range, execute S203; if the tapping audio is not within the audio judgment range, execute S202. S202: Determine whether the tapping audio is less than the minimum value of the audio judgment range. If the tapping audio is less than or equal to the minimum value of the judgment range, mark the wood knot as a dead knot. If the tapping audio is greater than the minimum value of the judgment range, determine whether the tapping audio is greater than the maximum value of the audio judgment range. If the tapping audio is greater than or equal to the maximum value of the audio judgment range, mark the wood knot as a live knot. S303: Obtain a grayscale photograph, locate the knot in the grayscale photograph, randomly select 10 knot locations as measurement points, and obtain the brightness of each of the 10 measurement points. Sum the brightness of the 10 measurement points and take the average to obtain the average brightness of the knot. Set a preset brightness threshold and determine whether the average brightness of the knot is greater than the preset brightness threshold. If the average brightness of the knot is greater than or equal to the preset brightness threshold, mark the knot as a live knot. If the average brightness of the knot is less than the preset brightness threshold, mark the knot as a dead knot.

[0008] Preferably, in S3, a comparison image is obtained by taking a second picture of the wood identified as dead knot using a high-definition camera. Specifically, the camera position and wood position at the time the first picture was taken are obtained, and the wood identified as dead knot is re-picked using the same camera position and wood position as when the first picture was taken to obtain a comparison image.

[0009] Preferably, in S3, the location of the knot is determined by comparing the images with the captured images, specifically: S301: Obtain a photographed image, adjust the transparency of the photographed image to 50% to obtain a semi-transparent photographed image, obtain a comparison image, and adjust the transparency of the comparison image to 50% to obtain a semi-transparent comparison image; S302: Overlap the semi-transparent photograph and the semi-transparent comparison image to obtain a semi-transparent overlap image. Mark the outline of the scab in the photograph and the outline of the scab in the comparison image in the semi-transparent overlap image. Determine whether the outline of the scab in the photograph and the outline of the scab in the comparison image are completely overlapped. If the outline of the scab in the photograph and the outline of the scab in the comparison image are not completely overlapped, the result is that the knot is loose. If the outline of the scab in the photograph and the outline of the scab in the comparison image are completely overlapped, then execute S303. S303: Obtain the outline of the scar edge gap in the semi-transparent photograph, obtain the outline of the scar edge gap in the semi-transparent comparison image, and determine whether the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image completely overlap. If the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image completely overlap, the judgment result is marked as the dead knot position is not loose. If the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image do not completely overlap, the judgment result is marked as the dead knot position is loose.

[0010] Preferably, in S4, the ability to repair a dead knot in wood is determined by comparing the location of the dead knot in the image, the image of the dead knot, and the audio recording of tapping the knot. Specifically: S401: Measure the minimum distance between the edge of the dead knot and the edge of the wood to obtain the dead knot edge distance. Determine whether the dead knot edge distance is less than 3 cm. If the dead knot edge distance is less than or equal to 3 cm, the result is that the dead knot cannot be repaired. If the dead knot semicircle distance is greater than 3 cm, then execute S402. S402: Obtain the image of the dead knot in the wood by comparing the images, obtain the outline shape of the dead knot in the image, and determine whether the outline shape of the dead knot is radial. If the outline shape of the dead knot is radial, the result is that the dead knot cannot be repaired. If the outline shape of the dead knot is not radial, then execute S403. S403: Measure the maximum diameter of the dead knot in the wood to obtain the width of the dead knot, and determine whether the width of the dead knot is greater than 2 cm. If the width of the dead knot is greater than or equal to 2 cm, the result is that the dead knot cannot be repaired. If the width of the dead knot is less than 2 cm, then proceed to S404. S404: Obtain the tapping audio of wood knots, obtain the minimum value of the set audio judgment range, set the preset threshold for safe tapping audio of dead knots, take the preset threshold for safe tapping audio of dead knots as the minimum value, take the minimum value of the set audio judgment range as the maximum value, obtain the safe audio range of dead knots, determine whether the tapping audio of wood knots is within the safe audio range of dead knots, if the tapping audio of wood knots is within the safe audio range of dead knots, the judgment result is that the dead knot can be repaired, if the tapping audio of wood knots is not within the safe audio range of dead knots, the judgment result is that the dead knot cannot be repaired.

[0011] A wood surface treatment inspection system includes the following modules: The wood knot marking module takes pictures of the wood with a high-definition camera, and determines whether there are knots on the wood surface by analyzing the color and texture of the pictures. If knots are present, the module locates and marks the position of the knots. The wood knot / live knot identification module uses a sound sensor to capture the sound of tapping on the knot location and then compares the captured image with the sound of the tapping to determine whether the knot is a dead or live knot. The timber dead knot loosening judgment module uses a high-definition camera to take a second picture of the timber identified as dead knots to obtain a comparison picture. By comparing the comparison picture with the captured picture, it is determined whether the dead knot position is loose. If the dead knot position is loose, the timber is marked as scrap timber. If the dead knot position is not loose, the timber dead knot repair judgment module is executed. The wood knot repair judgment module determines whether a wood knot can be repaired by comparing the location of the wood knot in the image, the image of the wood knot, and the audio of tapping the knot. If the wood knot can be repaired, epoxy resin is injected into the knot to fill and wrap it. If the wood knot cannot be repaired, the wood is marked as scrap wood.

[0012] (III) Beneficial Effects This invention provides a method and system for detecting wood surface treatment, which has the following beneficial effects: (1) In this scheme, the knot location is tapped and the image is taken for comprehensive analysis to determine whether the knot is dead or live. The comprehensive judgment by sight and hearing can improve the accuracy of the judgment result. Because the wood of dead knot is loose and there are gaps between the knot and the wood, the density is relatively lower. Therefore, the sound of tapping dead knot will be more muffled. The wood of live knot is hard and the live knot is tightly connected with the wood fibers of the wood. The relative density is higher. Therefore, the sound of tapping live knot will be more crisp. Therefore, by analyzing the tapping audio of knot, it can be determined whether the knot is dead or live. Because dead knot is darker in color and the edge outline of the knot is loose, it is easy to produce gaps. Live knot is lighter in color and the edge of the knot has no gaps. Therefore, by tapping audio and image, it can be determined whether the knot is dead or live. This facilitates the screening and detection of wood and reduces the misjudgment rate of knot detection and identification on the wood surface.

[0013] (2) In this scheme, the dead knot after being knocked is photographed a second time. By analyzing and comparing the images with the photographed images, it is determined whether the knot after being knocked has loosened or shifted. If the knot is loosened or shifted by knocking, it indicates that the strength of the knot is insufficient. Therefore, it indicates that the dead knot is difficult to repair and the pass rate after repair is low. Therefore, by comparing the images after knocking the live knot and the dead knot, it is easier to determine whether the knot can be reused through repair. This helps to avoid directly discarding the wood due to dead knots, which would waste materials and increase costs. Attached Figure Description

[0014] Figure 1 This is a flowchart of a wood surface treatment testing method according to the present invention; Figure 2 This is a schematic diagram of the structure of a wood surface treatment detection system according to the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figures 1-2 This invention provides a method for detecting wood surface treatment, comprising the following steps: S1: Take pictures of the wood using a high-definition camera. Determine whether there are knots on the wood surface based on the color and texture of the pictures. If knots are present, locate and mark the position of the knots in the wood. S2: By tapping the knots in the wood, the sound sensor obtains the tapping audio of the knots, and by comparing the photographed image with the tapping audio, it is determined whether the knot in the wood is a dead knot or a live knot. S3: Take a second picture of the wood identified as dead knot using a high-definition camera to obtain a comparison picture. Analyze the comparison picture with the original picture to determine whether the dead knot is loose. If the dead knot is loose, mark the wood as scrap wood. If the dead knot is not loose, proceed to S4. S4: By comparing the location of the dead knot in the picture, the image of the dead knot, and the audio of tapping the knot, determine whether the dead knot can be repaired. If the dead knot can be repaired, fill and wrap the dead knot by injecting epoxy resin into it. If the dead knot cannot be repaired, mark the wood as scrap wood.

[0017] In this embodiment, the wood is photographed to obtain images. Since the color and texture of the knots are different from those of the non-knotted areas of the wood, image analysis technology is used to determine whether there are knots on the wood surface based on the wood color and texture. Finally, the knot locations are marked to facilitate subsequent determination of whether the knots are dead or live. This solution combines tapping on the knot location with photographic images for comprehensive analysis to determine whether a knot is a dead or live knot. This visual and auditory approach improves the accuracy of the judgment. Dead knots, with their looser wood texture and gaps between the knot and the wood, have a lower density and therefore produce a duller sound when tapped. Live knots, on the other hand, are made of harder wood with a tighter connection to the wood fibers, resulting in a higher density and a crisper sound. Therefore, analyzing the tapping audio can determine whether a knot is dead or live. Dead knots are darker in color and have looser edges, making them more prone to gaps, while live knots are lighter in color and have no gaps. Thus, by combining the tapping audio and images, it is possible to determine whether a knot is dead or live, facilitating the screening and detection of wood and reducing the misjudgment rate in detecting and identifying knots on the wood surface. In this solution, the dead knot is photographed a second time after being knocked down. By analyzing and comparing the images with the original images, it is determined whether the knot has loosened or shifted after being knocked down. If the knot has loosened or shifted after being knocked down, it indicates that the strength of the knot is insufficient. Therefore, it indicates that the dead knot is difficult to repair and the success rate of repair is low. Thus, comparing the images of dead and live knots after being knocked down can help determine whether the knot can be repaired and reused. This helps to avoid directly discarding the wood due to dead knots, which would waste materials and increase costs. This solution determines whether dead knots in wood can be repaired by analyzing their location, images, and the audio recordings of tapping on the knots. Since the location and size of knots affect the strength of the wood, the location and size of dead knots can help determine if repair is necessary. Surface image detection only shows surface problems; the presence of internal hollowness or rot within the dead knot cannot be accurately determined visually. Therefore, the audio recordings of tapping on the wood can help determine if there is significant internal hollowness or rot within the dead knot, thus facilitating the assessment of whether repair is necessary. Because dead knots with internal hollowness, rot, or wormholes often appear as rings radiating outwards and are relatively dark in color, images of the dead knots can help determine if repair is necessary and whether the repaired knots will meet strength requirements. This improves the accuracy of knot detection and identification while avoiding wood waste. It is worth mentioning that the value of the preset threshold in this scheme can be obtained through weight analysis, which will not be elaborated on here.

[0018] Determine whether there are knots on the wood surface by examining the color and texture in photographs. S101: Acquire a photographed image, process the photographed image into grayscale to obtain a grayscale photographed image, and capture low-brightness elliptical areas in the grayscale photographed image and mark them as areas to be confirmed. S102: Locate the area to be confirmed in the photographed image, obtain the color of the area to be confirmed in the photographed image, determine whether the color of the area to be confirmed is brown, gray, brown or yellow. If the color of the area to be confirmed is brown, gray, brown or yellow, then execute S103. If the color of the area to be confirmed is not brown, gray, brown or yellow, then mark the area to be confirmed as a stain. S103: Obtain the edge texture of the area to be confirmed. Obtain the texture of the area to be confirmed. Determine whether the edge texture of the area to be confirmed and the texture of the area to be confirmed are consistent and continuous. If the edge texture of the area to be confirmed and the texture of the area to be confirmed are consistent and continuous, then mark the area to be confirmed as a stain. If the edge texture of the area to be confirmed and the texture of the area to be confirmed are not consistent and continuous, then mark the area to be confirmed as a scar.

[0019] In this embodiment, photographic images are obtained by taking pictures of the wood. Since the color and texture of the knotted area are different from those of the non-knotted area, the color of the knotted area is relatively darker than that of the non-knotted area. The texture of the knotted area and the non-knotted area are discontinuous and independent. Therefore, image analysis technology is used to determine whether there are knots on the wood surface by the color and texture of the wood. Finally, the knotted area is marked to facilitate subsequent determination of whether the knot is a dead knot or a live knot.

[0020] In S2, the sound sensor obtains the tapping audio of the knot by tapping the knot location of the wood. Specifically, the clamping position is marked on the wood surface, the wood is fixed at the clamping position by a clamp, the tapping distance and tapping force are set, and the same position of the wood knot is tapped three times by a tapping device according to the tapping distance and tapping force to obtain three audio values. The three audio values ​​are summed and the average value is taken to obtain the tapping audio.

[0021] In this embodiment, since the striking force, striking distance, and fixture position all affect the tapping audio of the scar, this solution uses the controlled variable method to measure three audio values ​​with the same striking force, striking distance, and furniture position, and then takes the average to obtain the tapping audio. This facilitates data cleaning of the acquired data and improves the accuracy of the detection results.

[0022] In S2, the determination of whether a wood knot is a dead knot or a live knot is made by taking a photo and listening to the sound of tapping the knot. Specifically: S201: Set the audio judgment range, obtain the tapping audio, and determine whether the tapping audio is within the audio judgment range. If the tapping audio is within the audio judgment range, execute S203; if the tapping audio is not within the audio judgment range, execute S202. S202: Determine whether the tapping audio is less than the minimum value of the audio judgment range. If the tapping audio is less than or equal to the minimum value of the judgment range, mark the wood knot as a dead knot. If the tapping audio is greater than the minimum value of the judgment range, determine whether the tapping audio is greater than the maximum value of the audio judgment range. If the tapping audio is greater than or equal to the maximum value of the audio judgment range, mark the wood knot as a live knot. S303: Obtain a grayscale photograph, locate the knot in the grayscale photograph, randomly select 10 knot locations as measurement points, and obtain the brightness of each of the 10 measurement points. Sum the brightness of the 10 measurement points and take the average to obtain the average brightness of the knot. Set a preset brightness threshold and determine whether the average brightness of the knot is greater than the preset brightness threshold. If the average brightness of the knot is greater than or equal to the preset brightness threshold, mark the knot as a live knot. If the average brightness of the knot is less than the preset brightness threshold, mark the knot as a dead knot.

[0023] In this embodiment, the type of knot (dead or live) is determined by comparing the tapping audio with the audio judgment range. For knots with unclear tapping audio that are difficult to determine, the brightness of the knot in the grayscale image is used for differentiation. This combined visual and auditory judgment improves the accuracy of the judgment. Because dead knots have loose wood texture and gaps between the knot and the wood, their density is relatively lower, resulting in a duller tapping sound. Live knots have hard wood and are tightly connected to the wood fibers, resulting in a higher relative density, resulting in a crisper tapping sound. Therefore, analyzing the tapping audio of a knot can determine whether it is a dead or live knot. Dead knots are darker in color and have looser edges, making them prone to gaps. Live knots are lighter in color and have no gaps at the edges. Thus, by combining the tapping audio and image, it is possible to determine whether a knot is dead or live, which facilitates the screening and detection of wood and reduces the misjudgment rate of knot detection and identification on the wood surface.

[0024] In S3, a high-definition camera is used to take a second picture of the wood identified as dead knots to obtain a comparison picture. Specifically, the camera position and the wood position at the time the first picture was taken are obtained, and the wood identified as dead knots is re-shot using the same camera position and wood position as when the first picture was taken to obtain a comparison picture.

[0025] In this embodiment, since it is necessary to compare the captured images with the comparison images later, it is necessary to ensure that the position of the wood and the position of the camera are the same during the two shootings, so as to ensure that the distance between the wood and the camera and the shooting angle are the same, which facilitates the comparison and judgment of the changes in the position of the knot after knocking.

[0026] In S3, the location of the knot is determined by comparing and analyzing images with the captured images. Specifically: S301: Obtain a photographed image, adjust the transparency of the photographed image to 50% to obtain a semi-transparent photographed image, obtain a comparison image, and adjust the transparency of the comparison image to 50% to obtain a semi-transparent comparison image; S302: Overlap the semi-transparent photograph and the semi-transparent comparison image to obtain a semi-transparent overlap image. Mark the outline of the scab in the photograph and the outline of the scab in the comparison image in the semi-transparent overlap image. Determine whether the outline of the scab in the photograph and the outline of the scab in the comparison image are completely overlapped. If the outline of the scab in the photograph and the outline of the scab in the comparison image are not completely overlapped, the result is that the knot is loose. If the outline of the scab in the photograph and the outline of the scab in the comparison image are completely overlapped, then execute S303. S303: Obtain the outline of the scar edge gap in the semi-transparent photograph, obtain the outline of the scar edge gap in the semi-transparent comparison image, and determine whether the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image completely overlap. If the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image completely overlap, the judgment result is marked as the dead knot position is not loose. If the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image do not completely overlap, the judgment result is marked as the dead knot position is loose.

[0027] In this embodiment, the analysis and comparison of images with photographed images are used to determine whether the knot has loosened or shifted after being struck. If the knot has loosened or shifted after being struck, it indicates that the strength of the knot is insufficient, which means that the dead knot is difficult to repair and the success rate of the repair is low. Therefore, comparing the images of live and dead knots after being struck can help determine whether the knot can be repaired and reused. By comprehensively comparing the outline of the knot and the outline of the gaps at the edge of the knot, the accuracy of the judgment can be improved.

[0028] In S4, the ability to repair dead knots is determined by comparing the location of the dead knots in the images, the images of the dead knots, and the audio recordings of tapping on the knots. Specifically: S401: Measure the minimum distance between the edge of the dead knot and the edge of the wood to obtain the dead knot edge distance. Determine whether the dead knot edge distance is less than 3 cm. If the dead knot edge distance is less than or equal to 3 cm, the result is that the dead knot cannot be repaired. If the dead knot semicircle distance is greater than 3 cm, then execute S402. S402: Obtain the image of the dead knot in the wood by comparing the images, obtain the outline shape of the dead knot in the image, and determine whether the outline shape of the dead knot is radial. If the outline shape of the dead knot is radial, the result is that the dead knot cannot be repaired. If the outline shape of the dead knot is not radial, then execute S403. S403: Measure the maximum diameter of the dead knot in the wood to obtain the width of the dead knot, and determine whether the width of the dead knot is greater than 2 cm. If the width of the dead knot is greater than or equal to 2 cm, the result is that the dead knot cannot be repaired. If the width of the dead knot is less than 2 cm, then proceed to S404. S404: Obtain the tapping audio of wood knots, obtain the minimum value of the set audio judgment range, set the preset threshold for safe tapping audio of dead knots, take the preset threshold for safe tapping audio of dead knots as the minimum value, take the minimum value of the set audio judgment range as the maximum value, obtain the safe audio range of dead knots, determine whether the tapping audio of wood knots is within the safe audio range of dead knots, if the tapping audio of wood knots is within the safe audio range of dead knots, the judgment result is that the dead knot can be repaired, if the tapping audio of wood knots is not within the safe audio range of dead knots, the judgment result is that the dead knot cannot be repaired.

[0029] In this embodiment, since the location and size of knots affect the strength of wood, the location and size of dead knots can be used to determine whether wood knots need repair. Since surface image detection can only see surface problems, it is impossible to accurately determine whether there are hollow or rotten parts inside the dead knots using only surface images. Therefore, the sound of tapping the wood can be used to determine whether there are large hollows and rotten parts inside the dead knots, thus making it easier to determine whether the wood with dead knots needs repair. Since dead knots with internal hollowness, rot, or wormholes often appear as rings radiating outwards on the surface of the dead knot and are relatively dark in color, the image of the wood dead knot can be used to determine whether the wood dead knots need repair and whether the repair can meet the strength requirements. This improves the accuracy of wood surface knot detection and identification while avoiding wood waste.

[0030] Please see Figures 1-2 This invention provides a wood surface treatment detection system, comprising the following modules: The wood knot marking module takes pictures of the wood with a high-definition camera, and determines whether there are knots on the wood surface by analyzing the color and texture of the pictures. If knots are present, the module locates and marks the position of the knots. The wood knot / live knot identification module uses a sound sensor to capture the sound of tapping on the knot location and then compares the captured image with the sound of the tapping to determine whether the knot is a dead or live knot. The timber dead knot loosening judgment module uses a high-definition camera to take a second picture of the timber identified as dead knots to obtain a comparison picture. By comparing the comparison picture with the captured picture, it is determined whether the dead knot position is loose. If the dead knot position is loose, the timber is marked as scrap timber. If the dead knot position is not loose, the timber dead knot repair judgment module is executed. The wood knot repair judgment module determines whether a wood knot can be repaired by comparing the location of the wood knot in the image, the image of the wood knot, and the audio of tapping the knot. If the wood knot can be repaired, epoxy resin is injected into the knot to fill and wrap it. If the wood knot cannot be repaired, the wood is marked as scrap wood.

[0031] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0032] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0033] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for detecting wood surface treatment, characterized in that, Includes the following steps: S1: Take pictures of the wood using a high-definition camera. Determine whether there are knots on the wood surface based on the color and texture of the pictures. If knots are present, locate and mark the position of the knots in the wood. S2: By tapping the knots in the wood, the sound sensor obtains the tapping audio of the knots, and by comparing the photographed image with the tapping audio, it is determined whether the knot in the wood is a dead knot or a live knot. S3: Take a second picture of the wood identified as dead knot using a high-definition camera to obtain a comparison picture. Analyze the comparison picture with the original picture to determine whether the dead knot is loose. If the dead knot is loose, mark the wood as scrap wood. If the dead knot is not loose, proceed to S4. S4: By comparing the location of the dead knot in the picture, the image of the dead knot, and the audio of tapping the knot, determine whether the dead knot can be repaired. If the dead knot can be repaired, fill and wrap the dead knot by injecting epoxy resin into it. If the dead knot cannot be repaired, mark the wood as scrap wood.

2. The method for detecting wood surface treatment according to claim 1, characterized in that: In S1, the presence of knots on the wood surface is determined by analyzing the color and texture in photographs. Specifically: S101: Acquire a photographed image, process the photographed image into grayscale to obtain a grayscale photographed image, and capture low-brightness elliptical areas in the grayscale photographed image and mark them as areas to be confirmed. S102: Locate the area to be confirmed in the photographed image, obtain the color of the area to be confirmed in the photographed image, determine whether the color of the area to be confirmed is brown, gray, brown or yellow. If the color of the area to be confirmed is brown, gray, brown or yellow, then execute S103. If the color of the area to be confirmed is not brown, gray, brown or yellow, then mark the area to be confirmed as a stain. S103: Obtain the edge texture of the area to be confirmed. Obtain the texture of the area to be confirmed. Determine whether the edge texture of the area to be confirmed and the texture of the area to be confirmed are consistent and continuous. If the edge texture of the area to be confirmed and the texture of the area to be confirmed are consistent and continuous, then mark the area to be confirmed as a stain. If the edge texture of the area to be confirmed and the texture of the area to be confirmed are not consistent and continuous, then mark the area to be confirmed as a scar.

3. The method for detecting wood surface treatment according to claim 1, characterized in that: In S2, the sound sensor obtains the tapping audio of the knot by tapping the knot location of the wood. Specifically, the clamping position is marked on the wood surface, the wood is fixed at the clamping position by a clamp, the tapping distance and tapping force are set, and the same position of the wood knot is tapped three times by a tapping device according to the tapping distance and tapping force to obtain three audio values. The three audio values ​​are summed and the average value is taken to obtain the tapping audio.

4. The method for detecting wood surface treatment according to claim 2, characterized in that: In S2, the determination of whether a wood knot is a dead knot or a live knot is made by taking a photo and listening to the sound of tapping the knot. Specifically: S201: Set the audio judgment range, obtain the tapping audio, and determine whether the tapping audio is within the audio judgment range. If the tapping audio is within the audio judgment range, execute S203; if the tapping audio is not within the audio judgment range, execute S202. S202: Determine whether the tapping audio is less than the minimum value of the audio judgment range. If the tapping audio is less than or equal to the minimum value of the judgment range, mark the wood knot as a dead knot. If the tapping audio is greater than the minimum value of the judgment range, determine whether the tapping audio is greater than the maximum value of the audio judgment range. If the tapping audio is greater than or equal to the maximum value of the audio judgment range, mark the wood knot as a live knot. S303: Obtain a grayscale photograph, locate the knot in the grayscale photograph, randomly select 10 knot locations as measurement points, and obtain the brightness of each of the 10 measurement points. Sum the brightness of the 10 measurement points and take the average to obtain the average brightness of the knot. Set a preset brightness threshold and determine whether the average brightness of the knot is greater than the preset brightness threshold. If the average brightness of the knot is greater than or equal to the preset brightness threshold, mark the knot as a live knot. If the average brightness of the knot is less than the preset brightness threshold, mark the knot as a dead knot.

5. The method for detecting wood surface treatment according to claim 1, characterized in that: In S3, a high-definition camera is used to take a second picture of the wood identified as dead knots to obtain a comparison picture. Specifically, the camera position and the wood position at the time the first picture was taken are obtained, and the wood identified as dead knots is re-shot using the same camera position and wood position as when the first picture was taken to obtain a comparison picture.

6. The method for detecting wood surface treatment according to claim 1, characterized in that: In S3, the location of the knot is determined by comparing and analyzing images with the captured images. Specifically: S301: Obtain a photographed image, adjust the transparency of the photographed image to 50% to obtain a semi-transparent photographed image, obtain a comparison image, and adjust the transparency of the comparison image to 50% to obtain a semi-transparent comparison image; S302: Overlap the semi-transparent photograph and the semi-transparent comparison image to obtain a semi-transparent overlap image. Mark the outline of the scab in the photograph and the outline of the scab in the comparison image in the semi-transparent overlap image. Determine whether the outline of the scab in the photograph and the outline of the scab in the comparison image are completely overlapped. If the outline of the scab in the photograph and the outline of the scab in the comparison image are not completely overlapped, the result is that the knot is loose. If the outline of the scab in the photograph and the outline of the scab in the comparison image are completely overlapped, then execute S303. S303: Obtain the outline of the scar edge gap in the semi-transparent photograph, obtain the outline of the scar edge gap in the semi-transparent comparison image, and determine whether the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image completely overlap. If the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image completely overlap, the judgment result is marked as the dead knot position is not loose. If the outline of the scar edge gap in the semi-transparent photograph and the outline of the scar edge gap in the semi-transparent comparison image do not completely overlap, the judgment result is marked as the dead knot position is loose.

7. The method for detecting wood surface treatment according to claim 4, characterized in that: In S4, the ability to repair dead knots is determined by comparing the location of the dead knots in the images, the images of the dead knots, and the audio recordings of tapping on the knots. Specifically: S401: Measure the minimum distance between the edge of the dead knot and the edge of the wood to obtain the dead knot edge distance. Determine whether the dead knot edge distance is less than 3 cm. If the dead knot edge distance is less than or equal to 3 cm, the result is that the dead knot cannot be repaired. If the dead knot semicircle distance is greater than 3 cm, then execute S402. S402: Obtain the image of the dead knot in the wood by comparing the images, obtain the outline shape of the dead knot in the image, and determine whether the outline shape of the dead knot is radial. If the outline shape of the dead knot is radial, the result is that the dead knot cannot be repaired. If the outline shape of the dead knot is not radial, then execute S403. S403: Measure the maximum diameter of the dead knot in the wood to obtain the width of the dead knot, and determine whether the width of the dead knot is greater than 2 cm. If the width of the dead knot is greater than or equal to 2 cm, the result is that the dead knot cannot be repaired. If the width of the dead knot is less than 2 cm, then proceed to S404. S404: Obtain the tapping audio of wood knots, obtain the minimum value of the set audio judgment range, set the preset threshold for safe tapping audio of dead knots, take the preset threshold for safe tapping audio of dead knots as the minimum value, take the minimum value of the set audio judgment range as the maximum value, obtain the safe audio range of dead knots, determine whether the tapping audio of wood knots is within the safe audio range of dead knots, if the tapping audio of wood knots is within the safe audio range of dead knots, the judgment result is that the dead knot can be repaired, if the tapping audio of wood knots is not within the safe audio range of dead knots, the judgment result is that the dead knot cannot be repaired.

8. A wood surface treatment detection system, applied to the wood surface treatment detection method according to any one of claims 1-7, characterized in that, Includes the following modules: The wood knot marking module takes pictures of the wood with a high-definition camera, and determines whether there are knots on the wood surface by analyzing the color and texture of the pictures. If knots are present, the module locates and marks the position of the knots. The wood knot / live knot identification module uses a sound sensor to capture the sound of tapping on the knot location and then compares the captured image with the sound of the tapping to determine whether the knot is a dead or live knot. The timber dead knot loosening judgment module uses a high-definition camera to take a second picture of the timber identified as dead knots to obtain a comparison picture. By comparing the comparison picture with the captured picture, it is determined whether the dead knot position is loose. If the dead knot position is loose, the timber is marked as scrap timber. If the dead knot position is not loose, the timber dead knot repair judgment module is executed. The wood knot repair judgment module determines whether a wood knot can be repaired by comparing the location of the wood knot in the image, the image of the wood knot, and the audio of tapping the knot. If the wood knot can be repaired, epoxy resin is injected into the knot to fill and wrap it. If the wood knot cannot be repaired, the wood is marked as scrap wood.

Images