A method and device for evaluating the amount of flying sparks based on visual images

Abstract

The application discloses a splash spark amount evaluation method and device based on visual images, wherein the method comprises the following steps: extracting the spark amount recognized by a plurality of detection positions set in the first cycle of the back-and-forth movement of a grinding wheel; comparing the actual number of sparks in the first cycle with the expected number of sparks, and storing the comparison result into a first set; extracting the spark amount recognized by a plurality of detection positions set in the second cycle of the back-and-forth movement of the grinding wheel; comparing the actual number of sparks with the expected number of sparks, and storing the comparison result into a second set; comparing the first set with the second set, and obtaining the evaluation result of the grinding quality according to the comparison result of the sets. The application can realize real-time detection, improve the accuracy of evaluation, replace the work of manually observing the grinding effect and evaluating the grinding effect, and reduce the cost.

Description

Technical Field

[0001] This invention belongs to the field of spatter detection and evaluation, specifically, it relates to a method and apparatus for evaluating the amount of spatter based on visual images. Background Technology

[0002] In the tobacco processing industry, the shredder is a crucial component of the process. Its main function is to cut tobacco into shreds of a specific width. To meet the required shredding width in tobacco processing, the sharpness of the blade is a critical factor. The shredder's grinding system uses a grinding wheel to continuously sharpen the blades, ensuring their sharpness while removing any tobacco residue. As the shredder continues to process tobacco, tobacco residue accumulates on the blades and the grinding wheel's surface. The shredder's electronic control system, according to pre-set parameters, controls the feed of the cutting blade and grinding wheel to replenish the consumed blades and grinding wheel.

[0003] For a long time, operators have judged the grinding effect based on the spark state and adjusted the settings of the shredder to control the feed of the cutter and grinding wheel to achieve the best grinding effect. Manual operation is highly experience-based, lacks standardized procedures, and is not real-time reliable, with a high degree of arbitrariness. This is not conducive to maintaining the sharpness of the blades of the shredder for a long time, so as to achieve the shredding width required by tobacco processing.

[0004] In view of this, the present invention is proposed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method and device for evaluating the amount of spatter based on visual images, realize standardized rules, detect the amount of grinding sparks in real time, and evaluate the effect of grinding cutting edge.

[0006] To solve the above-mentioned technical problems, the basic concept of the technical solution adopted by the present invention is: a method for evaluating the amount of splashing sparks based on visual images, comprising:

[0007] S1: Extract the amount of sparks identified at the detection location within the first cycle;

[0008] S2: Compare the actual number of sparks with the expected number of sparks, and store the comparison result in the first set;

[0009] S3: Extract the amount of sparks identified at the detection location during the second cycle;

[0010] S4: Repeat step S2 and store the comparison results in the second set;

[0011] S5: Compare the first set and the second set, and output the evaluation result based on the comparison result.

[0012] It should be noted that the cycle refers to one round trip of the grinding wheel, and multiple detection positions are set within this cycle.

[0013] Optionally, the method for detecting the spark quantity at the detection location in S1 includes:

[0014] S101: Stores image data of the detection location;

[0015] S102: Perform grayscale processing and data format conversion on image data;

[0016] S103: Filter noise from image data;

[0017] S104: Obtain spark geometry information and solve for the roundness value;

[0018] S105: Calculate the number of sparks in the image;

[0019] S106: Repeat the above steps to calculate the number of sparks in the other images.

[0020] Optionally, in S2, specific methods for storing the comparison results into a set include:

[0021] S201: Establish a set containing a number of elements equal to the number of detection positions;

[0022] S202: Compare the spark quantity results with the element values ​​in the set.

[0023] Optionally, in S5, the specific methods for comparing the first set and the second set include:

[0024] Take elements from two sets in order and with the same index, and compare the values ​​of the two elements. If the values ​​of all corresponding elements in the two sets are equal, then the sets are defined as equal; if there is an element with the same index in the two sets but with a different value, then the sets are defined as unequal.

[0025] Optionally, in S101, the specific method for storing the detection location image data includes:

[0026] S1011: Set the path variable corresponding to the detection location;

[0027] S1012: Input the path variables into the corresponding function to obtain the image information data;

[0028] S1013: Store the acquired image data.

[0029] Optionally, in S102, the methods for grayscale processing and data format conversion of image data specifically include:

[0030] S1021: Convert a color image to a grayscale image;

[0031] S1022: Convert the image data type to floating point type.

[0032] Optionally, in S103, the specific method for noise filtering of the image data is as follows:

[0033] A Gaussian low-pass filter is used to filter noise from image data.

[0034] Optionally, in S104, the specific method for obtaining spark geometry information and solving for the roundness value is as follows:

[0035] S1041: Perform edge calculations on image data to obtain spark geometry information;

[0036] S1042: The spark roundness value is calculated using a roundness algorithm.

[0037] Optionally, the specific method for calculating the number of sparks based on the roundness value in S105 includes:

[0038] S1051: Locate image data boundaries;

[0039] S1052: Calculate the actual roundness value of all image data;

[0040] S1053: Calculate the total number of sparks.

[0041] The present invention also provides a splash spark quantity evaluation device based on visual images. The device includes an industrial lens, an industrial camera, an industrial control computer, and a program. The industrial control computer executes the program to implement the above-described method.

[0042] By adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0043] 1. Excellent real-time performance and significantly improved processing speed;

[0044] 2. Replaces manual observation and evaluation of grinding effects, reducing initial labor costs;

[0045] 3. Calculations using functions are highly accurate.

[0046] The specific embodiments of the present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0047] The accompanying drawings, as part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention, but do not constitute an undue limitation of the invention. Obviously, the drawings described below are merely some embodiments, and those skilled in the art can obtain other drawings based on these drawings without creative effort. In the drawings:

[0048] Figure 1 This is a schematic flowchart of the spark quantity evaluation method of the present invention. Figure 1 ;

[0049] Figure 2 This is a schematic flowchart of the spark quantity detection method of the present invention;

[0050] Figure 3 This is a schematic flowchart of the spark quantity evaluation method of the present invention. Figure 2 ;

[0051] Figure 4 This is a schematic diagram of the spark quantity evaluation device of the present invention.

[0052] It should be noted that these accompanying drawings and textual descriptions are not intended to limit the scope of the invention in any way, but rather to illustrate the concept of the invention to those skilled in the art by referring to specific embodiments. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below with reference to the accompanying drawings. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0054] In the description of this invention, it should be noted that the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0055] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0056] A method for evaluating the amount of splash sparks based on visual images is provided. This method is stored as a computer program and can be executed by a computer system, such as... Figure 1 , 3 As shown, steps S1 to S5 are included:

[0057] In step S1, the amount of sparks identified at the detection position within the first cycle is extracted.

[0058] Specifically, the reciprocating motion of the grinding wheel is defined as one cycle, and multiple detection positions are set. The specific method is as follows: the blade length is marked using five points, namely position 1, position 2, position 3, position 4, and position 5. Further, position 1 is defined as the position where the grinding wheel just contacts the blade, position 2 as the midpoint of the blade length, position 3 as the position where the grinding wheel is about to leave the blade, position 4 as the midpoint of the line segment between positions 2 and 3, and position 5 as the midpoint of the line segment between positions 1 and 2.

[0059] Methods for detecting the number of sparks at the detection location, such as... Figure 2 Specifically, this includes steps S101 to S106.

[0060] In step S101, image data of the detection location is stored.

[0061] Specifically, an industrial camera is used to capture images of the detection locations, and then the captured images are stored. Specifically, steps S1011 to S1013 are executed. First, based on the five detection locations set in S1, five path variables are defined as "Location1", "Location2", "Location3", "Location4", and "Location5". Then, the five path variables are input into the imread() function to obtain the information data of the five images. Finally, the obtained image data is stored.

[0062] In step S102, the image data undergoes grayscale processing and data format conversion.

[0063] Specifically, steps S1021 to S1022 are executed.

[0064] In step S1021: the color image is converted into a grayscale image.

[0065] Specifically, the rgb2gray() function is used to convert a color image (RGB) to a grayscale image. Its calling format is as follows: X = rgb2gray(I), where X is the converted grayscale image and I is the original color image;

[0066] In step S1022: the image data type is converted to floating point type.

[0067] Specifically, the double() function is used to convert data types to floating-point types, and its calling format is as follows:

[0068] dX = double(X), where dX is the image after data transformation.

[0069] In step S103, noise filtering is performed on the image data.

[0070] Specifically, noise filtering is performed on the image data after grayscale processing. The specific method is as follows: using a Gaussian low-pass filter, according to the formula:

[0071]

[0072] It should be noted that: H represents the image information after low-pass filtering, and D0 is the cutoff frequency of the ideal low-pass filter, which is set to D0 = 80 in this system.

[0073] In step S104, the spark geometry information is obtained and the roundness value is calculated.

[0074] The specific steps are steps S1041 to S1042.

[0075] In step S1041, edge calculation is performed on the image data to obtain spark set information.

[0076] Specifically, the edge() function is used, with the Canny operator selected, to solve for the geometric information of the spark object in the image. Its calling format is as follows:

[0077] BW = edge(H,'canny',sigma), where H is the original grayscale image. This system uses the Canny operator, where sigma = 0.8.

[0078] In step S1042, the spark roundness value is calculated using a roundness algorithm.

[0079] Specifically, based on the geometric information of the sparks in the image, a roundness algorithm is used to traverse all edges and calculate the circle of the marker. Its calling format is as follows:

[0080] Metric = (4π * area) / (perimeter^2), where area is the area of ​​the circle, perimeter is the circumference of the circle, and metric is the roundness (0 Ω·cm). <Metric<1)。

[0081] In step S105, the number of sparks in the image is calculated.

[0082] Specifically, steps S1051 to S1053 are executed.

[0083] Step S1051: Locate the image data boundaries.

[0084] Specifically, based on the roundness value obtained by the roundness algorithm, the bwboundaries() function is used to find the boundaries of the data image.

[0085] In step S1051, the actual roundness value of all image data is calculated.

[0086] Specifically, the regionprops() function is used to calculate the area estimate of all objects to obtain the actual roundness value;

[0087] In step S1053, the total number of sparks is calculated.

[0088] Specifically, iterate through all objects whose boundaries conform to the roundness of the edge to determine the number of sparks.

[0089] In step S106, the above steps are repeated to calculate the number of sparks in the other images.

[0090] In step S2, the actual number of sparks is compared with the expected number of sparks, and the result is stored in the first set.

[0091] Specifically, the expected value and deviation value of the number of sparks at each detection position are set. When setting the expected value and deviation value of each detection point, the contact sequence between the grinding wheel and the cutting edge should be considered, as well as the normal spark amount attenuation at different detection points caused by the wear of the grinding wheel and the cutting edge. The setting should be based on these conditions.

[0092] Optionally, the comparison results between the actual spark quantity and the expected quantity are calculated. The calculated comparison results include: Result 1: Normal; Result 2: Exceeding the upper limit; Result 3: Exceeding the lower limit.

[0093] The specific steps for storing the comparison results into the set are S201 to S202.

[0094] In step S201, a set containing a number of elements equal to the number of detection positions is established.

[0095] Specifically, a set is established with the same number of elements as the number of detection positions. Each element represents the result of comparing the actual number of sparks at a detection position with the expected number of sparks at that position.

[0096] In step S202, the spark quantity comparison result is matched with the element value in the set.

[0097] Specifically, the array contains the same number of elements as the number of detection positions, and the comparison result of the spark quantity at each detection position is represented by the corresponding element value.

[0098] Optionally, the relationship between the detection result and the element value includes: value = -1, indicating that the lower limit is exceeded; value = 0, indicating that it is normal; value = 1, indicating that the upper limit is exceeded.

[0099] In step S3, the amount of sparks identified at the detection position within the second cycle is extracted.

[0100] In step S4, step S2 is repeated, and the comparison results are stored in the second set.

[0101] Specifically, the elements in the first set are copied to another set, that is, a new array set is created, and the elements in the first set are copied to the new array set.

[0102] Optionally, the element values ​​of this set can be cleared and the calculation results of the new cycle can be stored, which is called the second set.

[0103] Specifically, the elements in the existing set are cleared, and the result of comparing the actual number of sparks at the detection position in the new cycle with the expected number of sparks at that position is stored in this set, which is called the second set.

[0104] In step S5, the first set and the second set are compared, and the evaluation result is output based on the comparison result.

[0105] Specifically, the comparison result set from the previous period is compared with the comparison result set from the new period. Elements with the same index are retrieved from both arrays, and their values ​​are compared to see if they are equal. If the values ​​of elements with the same index in both arrays are equal, then the arrays are defined as equal. If at least one element with the same index in both arrays has a different value, then the arrays are defined as unequal.

[0106] Optional, specific comparison results include: Result 1: Considering the order of elements in the sets, set A is equal to set B, i.e., A = B; Result 2: Considering the order of elements in the sets, set A is not equal to set B, i.e., A ≠ B.

[0107] Specifically, a preset relationship is established between different arrays and evaluation results. The array results calculated in step S5 are mapped to the corresponding evaluation results. If the set calculation results are equal, it indicates that the grinding trends of the two cycles are the same, and an evaluation can be given based on the current results. If the set calculation results are not equal, it indicates that the grinding trends of the two cycles are inconsistent, and an evaluation can also be given based on the current results.

[0108] Optional evaluation results include: Result 1, grinding is normal; Result 2, grinding is strong; Result 3, grinding is weak; Result 4, grinding is abnormal.

[0109] If the trends of the two grinding cycles are the same, the evaluation result can be: normal grinding, strong grinding, or weak grinding; if the trends of the two grinding cycles are inconsistent, the evaluation result is: abnormal grinding.

[0110] The present invention also provides an apparatus for applying the above-described evaluation method, such as... Figure 4 As shown, the system includes: an industrial lens, an industrial camera, an industrial control unit, and a program. The industrial camera captures sparks from the grinding wheel and cutting edge during the grinding process, and then inputs the captured images into the industrial control unit. The industrial control unit calls the corresponding program to process the input images and count the number of sparks in the images. The industrial control unit calls the corresponding program to compare the expected value and deviation value set for each detection point with the actual spark quantity at that detection point, calculates the corresponding results, places them in a set, and represents them using an array of elements with the same number of detection positions. Then, the elements in the set are copied to a new set, the first set is cleared, and the spark quantity at the detection position in the new cycle is extracted. The expected value and deviation value set for each detection point in the new cycle are compared with the actual spark quantity at that detection point, and the corresponding results are calculated and placed in the new set. The corresponding program is called to compare the two set arrays, and the comparison result is mapped to the evaluation result and output.

[0111] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-described technical content to create equivalent embodiments without departing from the scope of the present invention. The implementation schemes in the above embodiments can also be further combined or replaced. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for evaluating the amount of splashing sparks based on visual images, characterized in that, include: S1: Extract the amount of sparks identified at the detection positions within the first cycle, wherein the cycle refers to one round trip of the grinding wheel as one cycle, and multiple detection positions are set within this cycle; S2: Compare the actual number of sparks with the expected number of sparks and store the result in the first set; Specifically, for each detection location, an expected value and a deviation value are set for the number of sparks. Based on the set expected value and deviation value for each detection location, the actual number of sparks is compared with the expected number of sparks, and the comparison result between the actual number of sparks and the expected number is calculated. The comparison result includes: Result 1: Normal; Result 2: Exceeding the upper limit; Result 3: Exceeding the lower limit. The specific method for storing the comparison result in the first set is executed in steps S201 to S202. S201: Establish a set containing a number of elements equal to the number of detection positions; S202: Match the results of the spark quantity comparison with the values ​​of the elements in the set; S3: Extract the amount of sparks identified at the detection location during the second cycle; S4: Repeat step S2 and store the comparison results into the second set; S5: Compare the first set and the second set, and output the evaluation result based on the comparison result; The process involves comparing the first set as the comparison result set for the previous cycle and the second set as the comparison result set for the new cycle. Elements with the same index are extracted from both sets, and their values ​​are compared. If the results are equal, it indicates that the grinding trends of the two cycles are the same. If the results are not equal, it indicates that the grinding trends of the two cycles are inconsistent. If the grinding trends of the two cycles are the same, the evaluation results include: normal grinding, strong grinding, and weak grinding. If the grinding trends of the two cycles are inconsistent, the evaluation result is: abnormal grinding.

2. The method for evaluating the amount of splashing sparks based on visual images according to claim 1, characterized in that, The method for detecting the amount of spark at the detection location extracted by S1 specifically includes: S101: Stores image data of the detection location; S102: Perform grayscale processing and data format conversion on image data; S103: Filter noise from image data; S104: Obtain spark geometry information and solve for the roundness value; S105: Calculate the number of sparks in the image; S106: Repeat the above steps to calculate the number of sparks in the other images.

3. The method for evaluating the amount of splashing sparks based on visual images according to claim 2, characterized in that, S101 specifically includes: S1011: Set the path variable corresponding to the detection location; S1012: Input the path variables into the corresponding function to obtain the image information data; S1013: Store the acquired image data.

4. The method for evaluating the amount of splashing sparks based on visual images according to claim 2, characterized in that, S102 specifically includes: S1021: Convert a color image to a grayscale image; S1022: Convert the image data type to a floating-point type.

5. The method for evaluating the amount of splashing sparks based on visual images according to claim 2, characterized in that, S103 specifically includes: A Gaussian low-pass filter is used to filter noise from image data.

6. The method for evaluating the amount of splashing sparks based on visual images according to claim 2, characterized in that, S104 specifically includes: S1041: Perform edge calculations on image data to obtain spark geometry information; S1042: The spark roundness value is calculated using a roundness algorithm.

7. The method for evaluating the amount of splashing sparks based on visual images according to claim 2, characterized in that, Specifically, S105 includes: S1051: Locate image data boundaries; S1052: Calculate the actual roundness value of all image data; S1053: Calculate the total number of sparks.

8. A device for evaluating the amount of splashing sparks based on visual images, characterized in that, include: The invention comprises an industrial lens, an industrial camera, an industrial control unit, and a program, wherein the industrial control unit executes the program to implement the visual image-based spatter quantity evaluation method as described in any one of claims 1-7.

Images