Cleaning system and cleaning method for oak barks

Abstract

The invention discloses a cleaning system and a cleaning method for oak barks, relates to the technical field of bark cleaning, and solves the problem that a traditional cleaning mode cannot effectively and fully clean sunken point positions on the surfaces of the barks. The method comprises the following steps of: selecting an edge contour, gradually confirming a corresponding associated area on the basis of the corresponding edge contour towards an inner ring, taking each associated area as a group of cleaning areas, fully covering the surface of the whole bark by the cleaning areas, and sequentially confirming associated line segments of the corresponding associated areas on the surface of the bark, according to the method, gradient changes between adjacent point locations are determined based on gradient features associated with different point locations in corresponding connection segments, numerical value confirmation is carried out based on specific gradient change conditions so as to lock the lowest pixel point, association determination of a cleaning route is carried out based on position features of a plurality of pixel points, and the cleaning route is determined based on the position features of the plurality of pixel points. Therefore, the overall cleaning effect of the surface of the corresponding bark is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of bark cleaning, in particular to a cleaning system and a cleaning method for oak bark. Background Art

[0002] Oak bark is widely used in many fields such as forestry processing, Chinese medicine processing and handicraft production; however, cleaning oak bark has always been a challenging problem. Traditional cleaning methods often rely on manual operation and lack precision and efficiency.

[0003] During manual cleaning, it is difficult for operators to accurately determine the areas on the bark that need to be cleaned, and they can usually only perform rough rinsing and wiping. This not only leads to uneven cleaning results, but also easily causes unnecessary damage to the bark. At the same time, due to the lack of understanding of the microscopic characteristics of the bark surface, manual cleaning cannot formulate an optimized cleaning path for the complex conditions such as different stain distributions and depressions and protrusions on the bark surface, making it difficult to completely remove some stains, especially deep in the bark texture or in depressions.

[0004] The requirements for cleaning quality and efficiency are increasing day by day. In modern industrial production and processing scenarios, a more scientific and accurate cleaning method and device are needed to meet the needs of oak bark cleaning. In addition, for some fields that have extremely high requirements for the quality of oak bark, such as high-end handicraft production and high-purity medicinal material processing, traditional cleaning methods cannot guarantee that the bark can still maintain its original physical and chemical properties after cleaning, which urgently requires a new cleaning technology to solve these problems.

[0005] The rise of machine vision technology has made it possible to solve these problems; through machine vision, high-resolution images of the surface of objects can be obtained, thus providing detailed data for subsequent analysis and processing. By using image processing technology to analyze images obtained by machine vision, the outline, area and microscopic features of the bark can be identified, which is crucial for formulating accurate cleaning strategies; moreover, by analyzing the gradient and other features of the pixel points in the image, special areas such as depressions on the bark surface can be found. These areas are often places where stains are easy to accumulate and difficult to clean, thus providing a basis for determining the cleaning route. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a cleaning system and a cleaning method for oak bark, which solves the problem that the traditional cleaning method fails to effectively and fully clean the concave points on the bark surface.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A system for cleaning oak bark, comprising: The machine vision end acquires the surface image of the oak bark that needs to be involved in the cleaning process, and transmits the acquired surface image to the bark image processing end; The bark image processing end performs contour analysis on the acquired surface image, determines the associated area to be cleaned from the surface image, and transmits the determined associated areas to the associated area analysis end in the following manner: Based on the confirmed surface image, the overall edge contour of the oak bark is identified: based on a preset color value interval, a partial area image belonging to the interval is calibrated as a bark image, based on the calibrated bark image, the peripheral edge contour points of the bark image are extracted, and a curve covered by a plurality of adjacent peripheral edge contour points is calibrated as the overall edge contour of the oak bark; Based on the confirmed overall edge contour of the oak bark, the associated areas to be cleaned are confirmed in turn: based on the confirmed bark image and the peripheral edge contour point, the peripheral edge contour point is placed in a set of two-dimensional coordinate systems, and the two-dimensional coordinates associated with the peripheral edge contour point are confirmed from the two-dimensional coordinate system, and then the two-dimensional coordinates of several different peripheral edge contour points are averaged to confirm the average coordinate, and the position of the average coordinate is marked as the center point of the bark image; Based on the confirmed center point of the bark image, the overall edge contour of the oak bark is reduced according to this center point, and the correlation distance L between the same points is recorded during the reduction process. When L=Y1, the area reduced by the first reduction process is marked as the first group of associated areas, where Y1 is a preset value, and then the subsequent reduction processes are executed in sequence, and the correlation distance L between the same points in each subsequently confirmed associated area is equal to Y1, and then the subsequent corresponding reduction processes are gradually executed to confirm the subsequent associated areas. When the points of the internal contour of the previous group of associated areas are reduced to the center point, and the generated correlation distance L is less than Y1, the area reduced in this reduction process is marked as the last group of associated areas; The plurality of groups of associated regions marked in sequence are transmitted to the associated region analysis terminal; The associated area analysis end performs feature analysis on each confirmed associated area, prioritizes the connection line segments between the same points in the corresponding associated area, and then selects the standard feature points in the connection line segment based on the gradient features of the associated pixels covered by the connection line segment, and determines the cleaning route based on several groups of standard feature points confirmed in the corresponding associated area. The specific method is as follows: Based on the same points associated during the scaling process in the associated area, a connecting line segment between corresponding same points is determined; Then determine the groups of pixels associated with the corresponding connecting line segments, and determine the vertical gradient Tz of the corresponding pixels from the bark image. i and the lateral gradient Tci , where i represents different pixels, using Determine the gradient feature Xz of its pixel i , starting from one end of the connecting line segment to the other end, the gradient feature Xz determined by the corresponding pixel point i Sorting and confirming the feature sequence; Let i = 1, 2, ..., n, where n represents the total number of pixels in the corresponding connecting line segment, and the feature sequence is proposed as: {Xz1, Xz2, ..., Xz n}, assume j∈[2,n-1], use |Xz j -Xz j-1 |+|Xz j -Xz j+1 |=XD j Determine the gradient difference XD associated with the corresponding pixel point j j ; Different gradient differences XD confirmed from different pixel points in the corresponding connecting line segment j , select XD j The pixel point associated with max is used as the standard feature point of this connecting line segment; Confirm the standard feature points of other connecting line segments in this associated area in turn, connect several groups of standard feature points confirmed in this associated area, confirm the cleaning routes belonging to this associated area, and confirm the cleaning routes associated with each group of associated areas in turn; Transmitting the confirmed groups of cleaning videos to the control center; The control center controls the cleaning spray gun based on the confirmed cleaning route, so that the cleaning spray gun cleans the surface of the oak bark according to the confirmed cleaning route. When executing the specific cleaning process, the cleaning starts from the innermost circle cleaning route and processes the cleaning routes outside the innermost circle in turn.

[0008] Preferably, a method for cleaning oak bark comprises the following steps: Step 1: Use a machine vision terminal to acquire a surface image of the oak bark that needs to be cleaned, and transmit the acquired surface image to a bark image processing terminal; Step 2: Use the bark image processing end to perform contour analysis on the acquired surface image, prioritize determining the overall edge contour of the surface image and simultaneously confirm its center point, reduce the overall edge contour based on the associated center point, and monitor the distance of the real-time reduction process. When the scaled distance between the same points reaches Y1, determine the associated area; Step 3: Perform feature analysis on each confirmed associated area through the associated area analysis terminal, prioritize the connection line segments between the same points in the corresponding associated area, and then select the standard feature points in the connection line segment based on the gradient features of the associated pixels covered by the connection line segment. The standard feature points of other connection line segments in the associated area are confirmed and locked in turn for the cleaning route. Step 4: Based on the different cleaning routes confirmed for each different associated area, the cleaning spray gun is controlled by the corresponding control center so that the cleaning spray gun cleans the surface of the oak bark according to the confirmed cleaning routes.

[0009] The present invention provides a cleaning system and method for oak bark. Compared with the prior art, the system and method have the following beneficial effects: The present invention obtains and analyzes the surface image of the oak bark, confirms the edge contour of the corresponding bark, and then proceeds to the inner circle based on the corresponding edge contour to gradually confirm the corresponding associated area. Each associated area is used as a group of cleaning areas, and the cleaning area fully covers the entire bark surface, so as to ensure the overall cleaning effect of the corresponding bark surface and achieve a better bark treatment effect. Subsequently, the associated line segments of the corresponding associated areas of the bark surface are confirmed in turn, and the gradient changes between adjacent points are determined based on the gradient characteristics associated with different points in the corresponding connection segments. Then, the numerical confirmation is performed based on the specific gradient changes to lock the lowest pixel point. Then, based on the position characteristics of several pixel points, the associated cleaning route is determined to ensure the overall cleaning effect of the corresponding bark surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic diagram of the principle framework of the present invention; Figure 2 This is a schematic diagram for determining the associated area of ​​the present invention. DETAILED DESCRIPTION

[0011] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0012] First embodiment: Please refer to Figure 1 , the present application provides a cleaning system for oak bark, comprising a machine vision end, a bark image processing end, a correlation region analysis end, and a control center, wherein the machine vision end, the bark image processing end, the correlation region analysis end, and the control center are all electrically connected from an output node to an input node; Among them, the machine vision end acquires the surface image of the oak bark that needs to participate in the cleaning process, and transmits the acquired surface image to the bark image processing end. Specifically, the oak bark will be placed on the corresponding cleaning table surface during the cleaning process, and the corresponding machine vision device can directly acquire the bark image above the cleaning table surface. The machine vision end needs to ensure that the resolution of the image is high enough when acquiring the image; Among them, the bark image processing end performs contour analysis on the acquired surface image, determines the associated area to be cleaned from the surface image, and transmits the determined associated area in sequence to the associated area analysis end. Specifically, the associated area to be cleaned is the associated area that needs to be cleaned. Such associated area is the area that can be covered by a flushing route in the corresponding flushing process. Therefore, the corresponding surface image will be divided into multiple different related areas. The corresponding flushing route needs to be determined in each related area to complete the cleaning process of the corresponding oak bark. The specific method of contour analysis is: Based on the confirmed surface image, the overall edge contour of the oak bark is identified: based on a preset color value interval, the partial area image belonging to this interval is calibrated as a bark image (the color of the surface of the washing table is generally bright white or silver, which is clearly distinguished from the color of the bark surface), and the color value interval is prepared in advance by the relevant operator according to the color characteristics of the bark. Based on the calibrated bark image, the outer edge contour points of the bark image are extracted, and the curve covered by several adjacent outer edge contour points is calibrated as the overall edge contour of the oak bark; Based on the confirmed overall edge contour of the oak bark, the associated areas to be cleaned are confirmed in turn: based on the confirmed bark image and the peripheral edge contour point, the peripheral edge contour point is placed in a set of two-dimensional coordinate systems, and the two-dimensional coordinates associated with the peripheral edge contour point are confirmed from the two-dimensional coordinate system, and then the two-dimensional coordinates of several different peripheral edge contour points are averaged to confirm the mean coordinate, and the position of the mean coordinate is marked as the center point of the bark image (when the corresponding point positions are placed in the two-dimensional coordinate system, the associated features of adjacent points will not be changed); Based on the confirmed center point of the bark image, the overall edge contour of the oak bark is reduced according to this center point, and the correlation distance L between the same points is recorded during the reduction process. When L=Y1, the area reduced by the first reduction process is marked as the first group of associated areas, and Y1 is a preset value. The specific value is determined by the operator based on experience. Then the subsequent reduction processes are performed in sequence, and the correlation distance L between the same points in each subsequent confirmed associated area is equal to Y1 (the so-called same point means that the points of the original contour line will change to other positions after reduction, and these two groups of points belong to the same point). Then, when the corresponding reduction processes are gradually performed, when the points of the internal contour of the previous group of associated areas are reduced to the center point in the corresponding reduction process, and the generated correlation distance L is less than Y1, the area reduced in this reduction process is marked as the last group of associated areas; The marked groups of associated regions are transmitted to the associated region analysis terminal. Specifically, Figure 2 , the outer area of ​​the corresponding bark image is the corresponding edge contour. Based on the corresponding edge contour, the center point associated with the bark image is locked, and then the edge contour is reduced in sequence. The distance between the same point in each reduction process is Y1. Therefore, by executing the corresponding reduction process in sequence, the relevant associated areas can be confirmed from the periphery in sequence, and the corresponding associated areas can be confirmed from the outside to the inside in sequence.

[0013] The associated region analysis end performs feature analysis on each associated region confirmed, preferentially determines the connecting line segments between the same points in the corresponding associated regions, and then selects the standard feature points in the connecting line segments based on the gradient features of the associated pixels covered by the connecting line segments. The specific method of selection is as follows: Based on the same points associated during the scaling process in the associated area (in the last group of associated areas, the center point is the same point of all points of the outer circle contour), the connecting line segments between the corresponding same points are determined; Then determine the groups of pixels associated with the corresponding connecting line segments, and determine the vertical gradient Tz of the corresponding pixels from the bark image. i and the lateral gradient Tc i (This type of data can be directly obtained from the image and belongs to the corresponding pixel features), where i represents different pixel points, using Determine the gradient feature Xz of its pixel i , starting from one end of the connecting line segment to the other end, the gradient feature Xz determined by the corresponding pixel point i Sorting and confirming the feature sequence; Let i = 1, 2, ..., n, where n represents the total number of pixels in the corresponding connecting line segment, and the feature sequence is proposed as: {Xz1, Xz2, ..., Xz n}, assume j∈[2,n-1], use |Xz j -Xz j-1 |+|Xz j -Xz j+1 |=XD j Determine the gradient difference XD associated with the corresponding pixel point j j , its gradient difference XD j The larger the value, the greater the change between the point and the previous and next points. On the bark surface, the point with such characteristics is generally the point with the largest degree of depression, and the point with the largest change between the previous and next points will appear. That is, the point with the largest change can be used as the point to be cleaned, so as to achieve the best cleaning effect for the designated area. Different gradient differences XD confirmed from different pixel points in the corresponding connecting line segment j , select XD j The pixel point associated with max is used as the standard feature point of this connecting line segment; Then, the standard feature points of other connecting line segments in the associated area are confirmed in turn, and several groups of standard feature points confirmed in the associated area are connected to confirm the cleaning routes belonging to the associated area, and the cleaning routes associated with each group of associated areas are confirmed in turn; Transmit the confirmed groups of cleaning videos to the control center.

[0014] Specifically, the internal contour line of the corresponding associated area is obtained based on the process of shrinking the external contour line. Therefore, there are identical points between the corresponding contour lines, and the corresponding connecting line segments can be confirmed between the identical points. The connecting line segments will cover the corresponding related pixel points on the bark surface. Each different pixel point has different gradient characteristics. The gradient change between adjacent points is determined based on the gradient characteristics associated with different points in the corresponding connecting segment, and then the numerical confirmation is performed based on the specific gradient change situation to lock the lowest pixel point. Then, based on the position characteristics of several pixels, the cleaning route is associated and determined to ensure the overall cleaning effect of the corresponding bark surface.

[0015] Its control center controls the cleaning spray gun based on the confirmed cleaning route, so that the cleaning spray gun cleans the surface of the oak bark according to the confirmed cleaning route. In the specific implementation process, the cleaning process is carried out from the inner cleaning route to the outer cleaning route in sequence, so as to prevent dirt from falling to other locations during the cleaning process.

[0016] Second embodiment: A method for cleaning oak bark, comprising the following steps: Step 1: Use a machine vision terminal to acquire a surface image of the oak bark that needs to be cleaned, and transmit the acquired surface image to a bark image processing terminal; Step 2: Use the bark image processing end to perform contour analysis on the acquired surface image, prioritize determining the overall edge contour of the surface image and simultaneously confirm its center point, reduce the overall edge contour based on the associated center point, and monitor the distance of the real-time reduction process. When the scaled distance between the same points reaches Y1, determine the associated area; Step 3: Perform feature analysis on each confirmed associated area through the associated area analysis terminal, prioritize the connection line segments between the same points in the corresponding associated area, and then select the standard feature points in the connection line segment based on the gradient features of the associated pixels covered by the connection line segment. The standard feature points of other connection line segments in the associated area are confirmed and locked in turn for the cleaning route. Step 4: Based on the different cleaning routes confirmed for each different associated area, the cleaning spray gun is controlled by the corresponding control center so that the cleaning spray gun cleans the surface of the oak bark according to the confirmed cleaning routes.

[0017] The third embodiment: This embodiment includes the entire implementation process of the above two groups of embodiments in its specific implementation process.

[0018] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0019] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A system for cleaning oak bark, characterized in that: include: The machine vision end acquires the surface image of the oak bark that needs to be involved in the cleaning process, and transmits the acquired surface image to the bark image processing end; The bark image processing end performs contour analysis on the acquired surface image, determines the associated area to be cleaned from the surface image, and transmits the determined associated areas to the associated area analysis end; The associated area analysis end performs feature analysis on each confirmed associated area, preferentially determines the connecting line segments between the same points in the corresponding associated area, and then selects the standard feature points in the connecting line segments based on the gradient features of the associated pixels covered by the connecting line segments, and determines the cleaning route based on several groups of standard feature points confirmed in the corresponding associated area; The control center controls the cleaning spray gun based on the confirmed cleaning route, so that the cleaning spray gun cleans the surface of the oak bark according to the confirmed cleaning route.

2. A system for cleaning oak bark according to claim 1, characterized in that: The specific method of performing contour analysis on the surface image at the bark image processing end is as follows: Based on the confirmed surface image, the overall edge contour of the oak bark is identified: based on a preset color value interval, a partial area image belonging to the interval is calibrated as a bark image, based on the calibrated bark image, the peripheral edge contour points of the bark image are extracted, and a curve covered by a plurality of adjacent peripheral edge contour points is calibrated as the overall edge contour of the oak bark; Based on the confirmed overall edge contour of the oak bark, the associated areas to be cleaned are confirmed in turn: based on the confirmed bark image and the peripheral edge contour point, the peripheral edge contour point is placed in a set of two-dimensional coordinate systems, and the two-dimensional coordinates associated with the peripheral edge contour point are confirmed from the two-dimensional coordinate system, and then the two-dimensional coordinates of several different peripheral edge contour points are averaged to confirm the average coordinate, and the position of the average coordinate is marked as the center point of the bark image; Based on the confirmed center point of the bark image, the overall edge contour of the oak bark is reduced according to this center point, and the associated distance L between the same points is recorded during the reduction process. When L=Y1, the area reduced by the first reduction process is marked as the first group of associated areas, where Y1 is a preset value, and then the subsequent reduction processes are executed in sequence, and the associated distance L between the same points in each subsequently confirmed associated area is equal to Y1, and then the subsequent corresponding reduction processes are gradually executed to confirm the subsequent associated areas; The plurality of groups of associated regions marked in sequence are transmitted to the associated region analysis terminal.

3. A system for cleaning oak bark according to claim 2, characterized in that: In the shrinking process: when the points of the inner contour of the previous set of associated regions are shrunk to the center point and the associated distance L generated is less than Y1, the area shrunk in this shrinking process is marked as the last set of associated regions.

4. A system for cleaning oak bark according to claim 1, characterized in that: The specific method of selecting the standard feature point in the connecting line segment at the associated region analysis end is: Based on the same points associated during the scaling process in the associated area, a connecting line segment between corresponding same points is determined; Then determine the groups of pixels associated with the corresponding connecting line segments, and determine the vertical gradient Tz of the corresponding pixels from the bark image. i and the lateral gradient Tc i , where i represents different pixels, using Determine the gradient feature Xz of its pixel i , starting from one end of the connecting line segment to the other end, the gradient feature Xz determined by the corresponding pixel point i Sorting and confirming the feature sequence; Let i = 1, 2, ..., n, where n represents the total number of pixels in the corresponding connecting line segment, and the feature sequence is proposed as: {Xz1, Xz2, ..., Xz n }, assume j∈[2,n-1], use |Xz j -Xz j-1 |+|Xz j -Xz j+1 |=XD j Determine the gradient difference XD associated with the corresponding pixel point j j ; Different gradient differences XD confirmed from different pixel points in the corresponding connecting line segment j , select XD j The pixel point associated with max is used as the standard feature point of this connecting line segment.

5. A system for cleaning oak bark according to claim 4, characterized in that: The specific method of determining the cleaning route at the associated area analysis end is: Confirm the standard feature points of other connecting line segments in this associated area in turn, connect several groups of standard feature points confirmed in this associated area, confirm the cleaning routes belonging to this associated area, and confirm the cleaning routes associated with each group of associated areas in turn; Transmit the confirmed groups of cleaning videos to the control center.

6. A system for cleaning oak bark according to claim 1, characterized in that: When executing a specific cleaning process, the control center starts cleaning from the innermost cleaning route and sequentially processes the cleaning routes outside the innermost circle.

7. A method for cleaning oak bark, the method being used to perform the cleaning process of the system for cleaning oak bark as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: Use a machine vision terminal to acquire a surface image of the oak bark that needs to be cleaned, and transmit the acquired surface image to a bark image processing terminal; Step 2: Use the bark image processing end to perform contour analysis on the acquired surface image, prioritize determining the overall edge contour of the surface image and simultaneously confirm its center point, reduce the overall edge contour based on the associated center point, and monitor the distance of the real-time reduction process. When the scaled distance between the same points reaches Y1, determine the associated area; Step 3: Perform feature analysis on each confirmed associated area through the associated area analysis terminal, prioritize the connection line segments between the same points in the corresponding associated area, and then select the standard feature points in the connection line segment based on the gradient features of the associated pixels covered by the connection line segment. The standard feature points of other connection line segments in the associated area are confirmed and locked in turn for the cleaning route. Step 4: Based on the different cleaning routes confirmed for each different associated area, the cleaning spray gun is controlled by the corresponding control center so that the cleaning spray gun cleans the surface of the oak bark according to the confirmed cleaning routes.

Images