CN112508945A - Nondestructive testing method for solid rocket engine based on computer vision - Google Patents

Abstract

The invention provides a nondestructive detection method of a solid rocket engine based on computer vision, which utilizes a data dichotomy to process image edge pixel points into a series of discontinuous line segments according to the gray value, selects the vertexes of polygons in an image or the edge points of an object with large curvature change as characteristic points, and utilizes the characteristic points and the series of discontinuous line segments to determine the starting point and the ending point of a defect; and calculating the pixel distance between the starting point and the ending point, and calculating the actual size according to a certain scale. The invention solves the problems of low manual detection precision and large human resource investment, can replace manual measurement in the prior art, and greatly improves the measurement efficiency and the measurement precision.

Description

Nondestructive testing method for solid rocket engine based on computer vision

Technical Field

The invention relates to the field of nondestructive testing of solid rocket engines, in particular to a nondestructive testing method of a solid rocket engine based on computer vision.

Background

The nondestructive detection and evaluation technology of the solid rocket engine is one of important technical means for ensuring the quality and the reliability of the solid rocket engine. The development of modern solid rocket engines leaves advanced nondestructive testing and evaluation technology, so that the quality of the solid rocket engines cannot be judged, the product failure and service life cannot be analyzed and predicted, and the products cannot be delivered from factories and delivered, so the modern nondestructive testing and evaluation technology is very important in the technical development of the solid rocket engines. The measurement result of the nondestructive testing image of the solid rocket engine directly influences the interpretation result of the subsequent nondestructive testing image, so that the measurement result has higher precision, an instrument with higher measurement precision is required during measurement, and the professional technical capability of a measurer is higher. In the past, nondestructive testing image measurement of the solid rocket engine is manually completed. At present, no case of applying the computer vision technology to nondestructive testing image measurement of the solid rocket engine exists in China.

Disclosure of Invention

The invention provides a nondestructive testing method for a solid rocket engine based on computer vision, which solves the problems of low manual testing precision and large human resource investment. The invention can replace manual measurement in the prior art and greatly improve the measurement efficiency and the measurement precision.

In order to solve the technical problem, the invention provides a nondestructive testing method of a solid rocket engine based on computer vision, which comprises the following steps:

s1: selecting engine image data;

s2: a noise reduction operation, wherein a Gaussian filter is adopted to reduce noise;

s3: performing convolution operation, namely performing convolution operation by using templates in the horizontal direction and the vertical direction respectively to obtain pixel values F (x) and F (y) of the center of the template;

s4: after the pixel value is obtained, the gradient change amplitude value is calculated by using a square root method, and the calculation formula is as follows:

s5: removing non-edge pixel points, extracting edges by adopting a high threshold and a low threshold, wherein the ratio of the high threshold to the low threshold is 2: 1, then keeping points which are larger than the high threshold value and between the high threshold value and the low threshold value, and removing points which are smaller than the low threshold value;

s6: calculating a characteristic parameter, wherein energy and moment of inertia represent the thickness degree of a characteristic texture, and are represented by F1 and F2, the larger the value is, the thicker the image texture vein is, the smaller the value is, the finer the image texture vein is, and the calculation formula is as follows:

the entropy represents the complexity of the texture of the engine image, and is represented by F3, the texture is simpler when the value is smaller, and the texture is more complex and more textured when the value is larger, and the calculation formula is as follows:

the correlation represents the degree of similarity between image rows and columns, denoted by F4, and μ and σ represent the mean and variance of the corresponding parameters, and is calculated as:

s7: characterizing engine image features, and characterizing the solid engine nondestructive testing image features through the mean value and the variance of Fl, F2, F3 and F4;

s8: judging high-order semantics through image characteristics, stopping working if the high-order semantics are judged to be defect-free, and measuring feature points if the high-order semantics are judged to be defect;

s9: and measuring the distance of the characteristic points, and calculating the pixel distance between the starting point and the ending point according to the selected starting point and the selected ending point to finish the measurement.

According to the nondestructive testing method for the solid rocket engine based on computer vision, the computer automatic measurement technology is used for replacing the manual measurement process, the investment of human resources is greatly reduced, the manual measurement precision is about 5mm in the aspect of measurement precision, the measurement precision is 0.2mm, and the measurement precision is improved by about 25 times compared with the manual measurement precision.

Drawings

In order to illustrate the embodiments of the present invention more clearly, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained by those skilled in the art without inventive effort.

FIG. 1, an original image;

FIG. 2, an image with edge detection;

FIG. 3, a partially enlarged image with edge detection;

FIG. 4, image (length) after edge detection at partial magnification;

FIG. 5, partially enlarged image (width) with edge detection;

FIG. 6, a horizontal convolution template graph;

FIG. 7, vertical convolution template diagram.

Detailed Description

The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the obtained embodiments. All other embodiments that can be derived from the embodiments of the present invention by a person of ordinary skill in the art are within the scope of the present invention.

As shown in fig. 1-5, a nondestructive testing method for a solid rocket engine based on computer vision comprises the following steps:

s1: selecting engine image data;

s2: a noise reduction operation, wherein a Gaussian filter is adopted to reduce noise;

s3: performing convolution operation, namely performing convolution operation by using templates in the horizontal direction and the vertical direction respectively to obtain pixel values F (x) and F (y) of the center of the template; as shown in fig. 6-7:

s4: after the pixel value is obtained, the gradient change amplitude value is calculated by using a square root method, and the calculation formula is as follows:

s5: removing non-edge pixel points, extracting edges by adopting a high threshold and a low threshold, wherein the ratio of the high threshold to the low threshold is 2: 1, then keeping points which are larger than the high threshold value and between the high threshold value and the low threshold value, and removing points which are smaller than the low threshold value;

s6: calculating a characteristic parameter, wherein energy and moment of inertia represent the thickness degree of a characteristic texture, and are represented by F1 and F2, the larger the value is, the thicker the image texture vein is, the smaller the value is, the finer the image texture vein is, and the calculation formula is as follows:

the entropy represents the complexity of the texture of the engine image, and is represented by F3, the texture is simpler when the value is smaller, and the texture is more complex and more textured when the value is larger, and the calculation formula is as follows:

the correlation represents the degree of similarity between image rows and columns, denoted by F4, and μ and σ represent the mean and variance of the corresponding parameters, and is calculated as:

s7: characterizing engine image features, and characterizing the solid engine nondestructive testing image features through the mean value and the variance of Fl, F2, F3 and F4;

s8: judging high-order semantics through image characteristics, stopping working if the high-order semantics are judged to be defect-free, and measuring feature points if the high-order semantics are judged to be defect;

s9: and measuring the distance of the characteristic points, and calculating the pixel distance between the starting point and the ending point according to the selected starting point and the selected ending point to finish the measurement.

Claims (1)

1. A nondestructive testing method for a solid rocket engine based on computer vision is characterized by comprising the following steps:

s1: selecting engine image data;

s2: a noise reduction operation, wherein a Gaussian filter is adopted to reduce noise;

s3: performing convolution operation, namely performing convolution operation by using templates in the horizontal direction and the vertical direction respectively to obtain pixel values F (x) and F (y) of the center of the template;

s4: after the pixel value is obtained, the gradient change amplitude value is calculated by using a square root method, and the calculation formula is as follows:

s5: removing non-edge pixel points, extracting edges by adopting a high threshold and a low threshold, wherein the ratio of the high threshold to the low threshold is 2: 1, then keeping points which are larger than the high threshold value and between the high threshold value and the low threshold value, and removing points which are smaller than the low threshold value;

s6: calculating a characteristic parameter, wherein energy and moment of inertia represent the thickness degree of a characteristic texture, and are represented by F1 and F2, the larger the value is, the thicker the image texture vein is, the smaller the value is, the finer the image texture vein is, and the calculation formula is as follows:

the entropy represents the complexity of the texture of the engine image, and is represented by F3, the texture is simpler when the value is smaller, and the texture is more complex and more textured when the value is larger, and the calculation formula is as follows:

the correlation represents the degree of similarity between image rows and columns, denoted by F4, and μ and σ represent the mean and variance of the corresponding parameters, and is calculated as:

s7: characterizing engine image features, and characterizing the solid engine nondestructive testing image features through the mean value and the variance of Fl, F2, F3 and F4;

s8: judging high-order semantics through image characteristics, stopping working if the high-order semantics are judged to be defect-free, and measuring feature points if the high-order semantics are judged to be defect;

s9: and measuring the distance of the characteristic points, and calculating the pixel distance between the starting point and the ending point according to the selected starting point and the selected ending point to finish the measurement.

Images