Vision-based method for identifying excess material in blind holes of solid rocket engine grain

Abstract

This invention discloses a vision-based method for identifying foreign matter within blind holes of solid rocket motor propellant grains, comprising: acquiring a calculated image of a model with no foreign matter within the blind hole of a reference propellant grain; and obtaining the perimeter P of the blind hole boundary of the reference propellant grain based on the image. m The maximum inscribed circle diameter D of the reference propellant column m Obtain the calculated image of the model containing foreign matter within the blind hole of the propellant column to be judged, and obtain the perimeter P of the blind hole boundary of the propellant column to be judged based on the image. y The maximum inscribed circle diameter D of the blind hole of the propellant column to be determined. y The area A of excess material inside the blind hole of the propellant column to be determined y And the blind hole area A of the propellant column to be determined m Establish a quantitative discrimination model for foreign matter in the blind hole of the propellant column; based on the quantitative discrimination model for foreign matter in the blind hole of the propellant column, use P m D m P y D y A y and A m The quantitative discrimination value of the foreign matter in the drug column to be judged is obtained. This invention is applicable to various foreign matter residues in actual drug column production, and has a high reliability of evaluation results.

Description

Technical Field

[0001] This invention belongs to the field of mass discrimination of foreign matter inside solid rocket motor propellant grains, specifically involving a vision-based method for quantitative discrimination of foreign matter inside blind holes of propellant grains. Background Technology

[0002] With the development of solid rocket motors and the in-depth research on solid rocket motor propellant grain processing systems, the advantages of solid rocket motor propellant grains, such as high energy, low cost, high thrust, long storage time, stable performance, and high production efficiency, have become more prominent. At the same time, the requirements for propellant grain processing quality are also gradually increasing. After blind hole processing, the residual silver wires and propellant debris inside directly affect the combustion performance of the propellant grain. With the increasing demands for quality control of these residues in online high-efficiency propellant grain processing systems, it is urgent to quantitatively determine the residual status of these residues within the blind holes of the propellant grain.

[0003] Currently, the evaluation method for residual material inside blind holes of propellant cartridges still relies on a single inscribed circle. This method can lead to serious judgment failures when assessing residual material inside the cartridge, especially at the center. Furthermore, it can cause serious judgment biases when assessing long and thin residual material at the center of the cartridge. This method cannot meet the requirements for accurate identification and evaluation of residual material inside various blind holes under real production conditions. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned defects and provide a vision-based method for identifying foreign matter in blind holes of solid rocket motor propellant grains. This method solves the technical problem that existing methods cannot accurately identify and evaluate the residual situation of various foreign matter in blind holes under real production conditions. This invention is applicable to various residual situations of propellant grains in real production conditions and has a high degree of reliability in evaluation results.

[0005] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0006] A vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains includes:

[0007] Obtain a model image of the blind hole of the reference propellant grain with no foreign matter inside, and calculate the perimeter P of the blind hole boundary based on this image. m The maximum inscribed circle diameter D of the reference propellant column m ;

[0008] Obtain a model image of the propellant column containing foreign matter in the blind hole, and obtain the perimeter P of the blind hole boundary of the propellant column based on the image. y The maximum inscribed circle diameter D of the blind hole of the propellant column to be determined. y The area A of excess material inside the blind hole of the propellant column to be determined yAnd the blind hole area A of the propellant column to be judged m ;

[0009] Establish a quantitative discrimination model for foreign matter inside blind holes of propellant cartridges;

[0010] Based on the quantitative discrimination model of foreign matter in the blind hole of the drug cartridge, using P m D m P y D y A y and A m The quantitative discrimination value of excess material in the drug column to be judged is obtained.

[0011] Furthermore, the method for obtaining a calculation image of a model with no foreign matter in the blind hole of a reference propellant column includes: acquiring a visual image of the blind hole of the reference propellant column using an industrial camera, and performing morphological closing operations on the visual image of the blind hole to obtain a calculation image of a model with no foreign matter in the blind hole of the propellant column; wherein the reference propellant column is a propellant column with no foreign matter in the blind hole.

[0012] The method for obtaining the calculation image of the model containing foreign matter in the blind hole of the propellant column to be judged includes: using an industrial camera to acquire a visual image of the blind hole of the propellant column to be judged, and performing morphological closing operation processing on the visual image of the blind hole to obtain the calculation image of the model containing foreign matter in the blind hole of the propellant column.

[0013] Furthermore, methods for acquiring blind-hole visual images of reference propellant charges using industrial cameras include:

[0014] An industrial camera is placed at the 1st, 2nd...nth shooting positions to capture images of the blind hole of the reference propellant column, resulting in n images. The visual image of the blind hole of the reference propellant column is obtained by weighted averaging the n images.

[0015] Methods for acquiring blind-hole visual images of the propellant charge to be judged using industrial cameras include:

[0016] An industrial camera is placed at the 1st, 2nd...nth shooting positions to capture images of the blind hole of the propellant to be judged, resulting in n images. The visual image of the blind hole of the propellant to be judged is obtained by weighted average image fusion of the n images.

[0017] The 1st, 2nd...nth shooting positions are different positions along the axis of the drug column, where n≥2.

[0018] Furthermore, the perimeter of the blind hole boundary is the set of blind hole contour boundary points obtained through the boundary tracing algorithm.

[0019] Furthermore, the area A of excess material inside the blind hole of the propellant column to be determined... y This is a set of points representing the extraneous material region inside the blind hole, obtained through a boundary tracing algorithm.

[0020] The blind hole area A of the propellant column to be determined m This is the set of blind hole region points obtained through the boundary tracing algorithm.

[0021] Furthermore, the quantitative discrimination model for excess material inside the blind hole of the propellant column is as follows:

[0022]

[0023] Where Q is the blind hole quality discrimination score of the drug column to be judged, that is, the quantitative discrimination value of the foreign matter in the drug column to be judged, and w is the weight vector.

[0024] Furthermore, the method for determining the weight vector w is as follows:

[0025]

[0026] In the formula, w p For the perimeter weight, w a For area weights, w d This is the weight of the largest inscribed circle diameter.

[0027] Furthermore, w p w a and w d Determined according to the following formula:

[0028]

[0029]

[0030]

[0031] In the formula, k p For the perimeter weight gain, α p k is the perimeter weighted index. a For the area weight gain, α a k is the area weighting index. d α is the weight gain for the maximum inscribed circle diameter. d It is the weight index of the largest inscribed circle diameter.

[0032] Furthermore, k p k a and k d The value of k ranges from 1 to 100, and k p k a and k d The sum of the three is equal to 1 to 100;

[0033] α p α a and α d The value ranges from 0 to 1.

[0034] Compared with the prior art, the present invention has at least one of the following advantages:

[0035] (1) This invention creatively proposes a vision-based method for identifying foreign matter in blind holes of solid rocket engine propellant grains. It reasonably establishes a multi-factor comprehensive quantitative identification model for foreign matter in blind holes of solid rocket engine propellant grains, and fully considers the influence of the perimeter of the blind hole boundary, the envelope area of ​​foreign matter in the blind hole and the diameter of the maximum inscribed circle on the quality identification of blind holes containing foreign matter. The evaluation model is more comprehensive and reasonable, and the evaluation results are more reliable.

[0036] (2) By acquiring multiple images of blind holes containing foreign objects with quantitative discrimination and performing fusion processing, this invention can reduce the image error caused by single image acquisition and further improve the credibility of the evaluation results.

[0037] (3) The present invention provides a method for determining parameters such as the perimeter of the blind hole boundary and the envelope area of ​​the foreign matter inside the blind hole, which can effectively improve the discrimination efficiency and accuracy;

[0038] (4) The quantitative evaluation method of the present invention can be adapted to the various residual substances in the actual production of pharmaceutical columns. Attached Figure Description

[0039] Figure 1 This is a flowchart illustrating the vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains according to the present invention.

[0040] Figure 2 This is a schematic diagram of the blind hole of the propellant column to be identified in this invention. Detailed Implementation

[0041] The features and advantages of the present invention will become clearer and more apparent from the following detailed description.

[0042] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments. Although various aspects of embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated otherwise.

[0043] The purpose of this invention is to provide a vision-based method for quantitatively identifying foreign matter inside blind holes in solid rocket motor propellant grains. This method establishes a comprehensive quantitative discrimination model consisting of multiple factors, including the perimeter of the blind hole boundary, the envelope area of ​​foreign matter inside the blind hole, and the diameter of the maximum inscribed circle. By processing the collected images of blind holes with no foreign matter and images with foreign matter remaining, the calculation parameters in the quantitative discrimination model are extracted, thereby obtaining the mass score value of the blind hole inside the solid rocket motor propellant grain to be quantitatively identified.

[0044] This invention provides a vision-based method for quantitatively identifying foreign matter inside blind holes of solid rocket motor propellant grains, the specific steps of which are as follows:

[0045] Step 1. Use an industrial camera to take pictures to obtain a visual template image of the blind hole of the solid rocket motor propellant grain without any foreign objects. After morphological closing operation processing, obtain the calculated image of the blind hole of the propellant grain without any foreign objects (the calculated image of the blind hole of the reference propellant grain without any foreign objects).

[0046] Step 2. Use an industrial camera to take pictures to obtain visual images of the blind holes of the solid rocket motor propellant grains to be quantitatively identified. After morphological closing operation processing, obtain the calculated image of the model containing foreign matter in the blind holes of the propellant grains to be quantitatively identified (the calculated image of the model containing foreign matter in the blind holes of the propellant grains to be identified).

[0047] Step 3. Establish a quantitative discrimination model for foreign matter inside the blind hole of the solid rocket motor propellant grain;

[0048] Step 4. Calculate the model image based on the model obtained in Step 1 and Step 2, and calculate the parameters in the quantitative discrimination model in Step 3, including the perimeter of the blind hole boundary of the template image and the image to be quantitatively discriminated, the area of ​​the blind hole and the area of ​​the excess material in the blind hole of the image to be quantitatively discriminated, and the maximum inscribed circle diameter of the template image and the image to be quantitatively discriminated.

[0049] Step 5. Substitute the parameters obtained in Step 4, such as the perimeter of the blind hole boundary, the area of ​​the blind hole and the excess material inside the blind hole, and the diameter of the maximum inscribed circle, into the membership function to obtain the weight vector in the quantitative discrimination model of Step 3.

[0050] Step 6. Substitute the parameters obtained in Step 4 and Step 5 into the quantitative discrimination model established in Step 3 to obtain the mass score of the blind hole inside the blind hole of the solid rocket motor propellant grain to be quantitatively discriminated.

[0051] In one specific implementation, the visual template image of no extraneous objects inside the blind hole of the solid rocket motor propellant grain in step 1 is obtained by weighted averaging image fusion of two camera images acquired at different locations.

[0052] In one specific implementation, the visual image of the blind hole of the solid rocket motor propellant grain to be quantitatively identified in step 2 is obtained by weighted averaging image fusion of two camera images acquired at different locations.

[0053] In one specific implementation, the perimeter of the blind hole boundary in step 4 is the set of blind hole contour boundary points obtained by the boundary tracing algorithm.

[0054] In one specific implementation, the area of ​​excess material inside the blind hole in step 4 is a set of points of excess material region inside the blind hole obtained by the boundary tracing algorithm, and the area of ​​the blind hole is a set of points of blind hole region obtained by the boundary tracing algorithm.

[0055] In one specific implementation, the membership functions in step 5 include the perimeter membership function, the area membership function, and the maximum inscribed circle diameter membership function.

[0056] This invention reduces image errors caused by single-image acquisition by acquiring multiple images of blind holes containing foreign matter and performing fusion processing. By rationally establishing a multi-factor comprehensive quantitative discrimination model for foreign matter in blind holes of solid rocket motor propellant grains, it can fully consider the influence of the blind hole boundary perimeter, the envelope area of ​​foreign matter within the blind hole, and the diameter of the maximum inscribed circle on the quality discrimination of blind holes containing foreign matter. This quantitative evaluation method is adaptable to various foreign matter residue situations in actual propellant grain production.

[0057] As can be seen from the above technical solution, compared with the currently used methods, the present invention discloses a multi-factor comprehensive quantitative identification method for foreign matter in blind holes of solid rocket motor propellant grains based on vision. This quantitative evaluation method can adapt to various foreign matter residue situations in real production, the evaluation model is more comprehensive and reasonable, and the evaluation results are more reliable.

[0058] Example:

[0059] This invention provides a vision-based method for quantitatively identifying foreign matter inside blind holes of solid rocket motor propellant grains, such as... Figure 1 This includes the following steps:

[0060] Step 1: Use an industrial camera to take pictures to obtain a visual template image of no foreign matter inside the blind hole of the solid rocket motor propellant grain. After morphological closing operation processing, obtain a calculated image of the model of no foreign matter inside the blind hole of the propellant grain.

[0061] The visual template image of the solid rocket motor propellant grain blind hole without any foreign objects is obtained by weighted averaging of two camera images acquired at different locations. This weighted averaging image fusion processing can be implemented using the cv2.addWeighted function in OpenCV software, and the morphological closing operation of the image can be implemented using the morphologyEx function in OpenCV software.

[0062] Step 2: Use an industrial camera to take pictures to obtain visual images of the blind holes of the solid rocket motor propellant grains to be quantitatively identified, and obtain the calculated image of the model containing foreign matter in the blind holes of the propellant grains to be quantitatively identified after morphological closure operation processing.

[0063] The visual image of the blind hole of the solid rocket motor propellant grain to be quantitatively identified is obtained by weighted averaging of two camera images acquired at different locations. The weighted averaging image fusion process can be implemented by the cv2.addWeighted function in OpenCV software, and the morphological closing operation of the image can be implemented by the morphologyEx function in OpenCV software.

[0064] Step 3: Establish a quantitative discrimination model for foreign matter inside the blind hole of the solid rocket motor propellant grain;

[0065] The quantitative discrimination model for excess material inside the blind hole of the propellant column in this step is as follows:

[0066]

[0067] In the formula, Q is the quality discrimination score of the blind hole of the propellant column containing excess material, and P y For the model of a drug column containing foreign matter in a blind hole, calculate the perimeter of the blind hole boundary in the image (i.e., the perimeter of the blind hole boundary of the drug column to be identified), P m For a model where there is no excess material inside the blind hole of a propellant grain, calculate the perimeter of the blind hole boundary in the image (i.e., the perimeter of the blind hole boundary of the reference propellant grain), A y To calculate the area of ​​foreign matter inside the blind hole of the drug column to be quantified (i.e., the area of ​​foreign matter inside the blind hole of the drug column to be quantified), A m To calculate the blind hole area (i.e., the blind hole area of ​​the drug column to be identified) in the image of the drug column blind hole containing foreign matter, D y For the model of the blind hole containing foreign matter in the propellant column to be quantitatively determined, calculate the maximum inscribed circle diameter of the image (i.e., the maximum inscribed circle diameter of the blind hole of the propellant column to be determined), D. m Calculate the maximum inscribed circle diameter (i.e., the maximum inscribed circle diameter of the reference propellant grain) in the image for a model with no excess material inside the blind hole of the propellant grain, where w is the weight vector. For example... Figure 2 In the blind hole of the propellant column to be identified, the black border represents the perimeter P of the blind hole boundary. y The entire area inside the black border (including the white and gray areas) represents the blind hole area A of the drug cartridge to be determined. m Gray represents excess material, and the area of ​​the gray region is the area A of excess material inside the blind hole of the drug cartridge to be judged. y The maximum inscribed circle diameter of the white area inside the black border is D, which is the maximum inscribed circle diameter of the blind hole of the drug cartridge to be determined. y .

[0068] Step 4: Calculate the model image obtained in Step 1 and Step 2, and calculate the parameters in the quantitative discrimination model in Step 3, including the perimeter of the blind hole boundary, the area of ​​the blind hole and the excess material inside the blind hole, and the maximum inscribed circle diameter of the template image and the image to be quantitatively discriminated.

[0069] The perimeter of the blind hole boundary in this step is obtained from the set of blind hole contour boundary points using a boundary tracing algorithm. The calculation method is as follows:

[0070]

[0071]

[0072] In the formula, x y Calculate the abscissa (y) of the image pixels for the model of the drug column with foreign matter in the blind hole to be quantitatively determined. y To calculate the ordinate of the image pixels for the model of a drug column with blind holes containing excess material to be quantitatively determined, R y To calculate the set of blind hole boundary points in the image of the model containing foreign matter in the blind hole of the drug column to be quantitatively determined, x m Calculate the x-coordinate of the image pixels for a model with no excess material inside the blind hole of the propellant grain. m Calculate the ordinate of the image pixels for a model with no excess material inside the blind hole of the propellant grain, R m The set of blind hole boundary points in the image is calculated for a model where there is no excess material inside the blind hole of the propellant grain.

[0073] The envelope area of ​​the foreign matter within the blind hole in this step is obtained by using a boundary tracing algorithm to set the points of the foreign matter region within the blind hole. The calculation method is as follows:

[0074]

[0075]

[0076] In the formula, Ω y Ω is the set of points in the image of the foreign matter region in the blind hole model of the drug column to be quantitatively determined. m The set of points in the blind hole region of the model for the drug column containing excess material in the blind hole is calculated.

[0077] The maximum inscribed circle diameter in this step is calculated as follows:

[0078] D y =max(D ymin )

[0079] D m =max(D mmin )

[0080] In the formula, D ymin For the model of a drug column containing excess material in a blind hole, calculate the diameter D of the inscribed circle inside the blind hole in the image. mmin The diameter of the inscribed circle inside the blind hole in the image is calculated for a model with no excess material inside the blind hole of the propellant grain. ymin and D mminThis can be achieved using the cv2.pointPolygonTest function in the OpenCV software.

[0081] Step 5: Substitute the parameters obtained in Step 4, such as the perimeter of the blind hole boundary, the area of ​​the blind hole and the excess material inside the blind hole, and the maximum inscribed circle diameter, into the membership function to obtain the weight vector in the quantitative discrimination model in Step 3.

[0082] The weight vector in this step is calculated as follows:

[0083]

[0084] In the formula, w p For the perimeter weight, w a For area weights, w d This is the weight of the largest inscribed circle diameter.

[0085] The membership functions in this step include the perimeter membership function, the area membership function, and the membership function for the largest inscribed circle diameter, and their expressions are as follows:

[0086]

[0087]

[0088]

[0089] In the formula, k p For the perimeter weight gain, α p k is the perimeter weighted index. a For the area weight gain, α a k is the area weighting index. d α is the weight gain for the maximum inscribed circle diameter. d It is the weight index of the largest inscribed circle diameter.

[0090] Step 6: Substitute the parameters obtained in Step 4 and Step 5 into the quantitative discrimination model established in Step 3 to obtain the mass score of the excess material in the blind hole of the solid rocket motor propellant grain to be quantitatively judged. The larger the Q value, the better the quality of the blind hole.

[0091] The implementation methods described in this specification are interconnected and cannot be achieved independently of any single step. For details on the differences and similarities between each implementation method, please refer to the specification. The methods involved in the implementation are mainly aimed at solid rocket motor propellant grains, and their application in the blind hole drilling of propellant grains is already well-documented. Therefore, this invention introduces a quality assessment method based on the residue of foreign matter in the blind hole after the operation.

[0092] The present invention has been described in detail above with reference to specific embodiments and exemplary examples; however, these descriptions should not be construed as limiting the present invention. Those skilled in the art will understand that various equivalent substitutions, modifications, or improvements can be made to the technical solutions and embodiments of the present invention without departing from the spirit and scope of the invention, and all such modifications and improvements fall within the scope of the present invention. The scope of protection of the present invention is defined by the appended claims.

[0093] The contents not described in detail in this specification are common knowledge to those skilled in the art.

Claims

1. A vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains, characterized in that, include: Obtain a model image of the blind hole of the reference propellant grain with no foreign matter inside, and calculate the perimeter of the blind hole boundary based on this image. and the maximum inscribed circle diameter of the reference propellant column ; Obtain a model image of the propellant column containing foreign matter within the blind hole, and calculate the perimeter of the blind hole boundary based on this image. The maximum inscribed circle diameter of the blind hole of the propellant column to be determined. Area of ​​excess material inside the blind hole of the propellant column to be determined and the blind hole area of ​​the propellant column to be determined ; Establish a quantitative discrimination model for foreign matter inside blind holes of propellant cartridges; Based on the quantitative discrimination model of foreign matter in the blind hole of the drug cartridge, using , , , , and Obtain the quantitative discrimination value of the excess material in the drug column to be judged; Methods for acquiring blind-hole visual images of reference propellant charges using industrial cameras include: An industrial camera is placed at the 1st, 2nd...nth shooting positions to capture images of the blind hole of the reference propellant column, resulting in n images. The visual image of the blind hole of the reference propellant column is obtained by weighted averaging the n images. Methods for acquiring blind-hole visual images of the propellant charge to be judged using industrial cameras include: An industrial camera is placed at the 1st, 2nd...nth shooting positions to capture images of the blind hole of the propellant to be judged, resulting in n images. The visual image of the blind hole of the propellant to be judged is obtained by weighted average image fusion of the n images. The 1st, 2nd...nth shooting positions are different positions along the axis of the drug column, where n≥2; The quantitative discrimination model for foreign matter in the blind hole of the propellant column is as follows: in, This is the quality discrimination score for the blind hole of the propellant column to be judged, that is, the quantitative discrimination value of the foreign matter inside the propellant column to be judged. It is a weight vector; The method for obtaining a calculation image of a model with no foreign matter in the blind hole of a reference propellant grain includes: acquiring a visual image of the blind hole of the reference propellant grain using an industrial camera, and performing morphological closing operations on the visual image of the blind hole to obtain a calculation image of a model with no foreign matter in the blind hole of the propellant grain; wherein the reference propellant grain is a propellant grain with no foreign matter in the blind hole; The method for obtaining the calculation image of the model containing foreign matter in the blind hole of the propellant column to be judged includes: using an industrial camera to acquire a visual image of the blind hole of the propellant column to be judged, and performing morphological closing operation processing on the visual image of the blind hole to obtain the calculation image of the model containing foreign matter in the blind hole of the propellant column.

2. The vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains according to claim 1, characterized in that, The perimeter of the blind hole boundary is the set of boundary points of the blind hole contour obtained by the boundary tracing algorithm.

3. The vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains according to claim 1, characterized in that, Area of ​​excess material inside the blind hole of the propellant column to be determined This is a set of points representing the extraneous material region inside the blind hole, obtained through a boundary tracing algorithm. Area of ​​blind hole of the propellant column to be determined This is the set of blind hole region points obtained through the boundary tracing algorithm.

4. The vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains according to claim 1, characterized in that, weight vector The method for determining it is as follows: In the formula, For the perimeter weight, For area weights, This is the weight of the largest inscribed circle diameter.

5. The vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains according to claim 4, characterized in that, , and Determined according to the following formula: In the formula, For the perimeter weight gain, Perimeter weighted index, For area weight gain, It is the area weighting index. The weight gain is the maximum inscribed circle diameter. It is the weight index of the largest inscribed circle diameter.

6. The vision-based method for identifying foreign matter inside blind holes of solid rocket motor propellant grains according to claim 5, characterized in that, , and The value ranges from 1 to 100, and , and The sum of the three is equal to 1 to 100; , and The value can range from 0 to 1.

Images