CN111402307B - Method for processing electric water level image - Google Patents

Abstract

The present application discloses a method for processing electric level images, which is characterized in that it includes the following steps: obtaining a preset number of original electric level images before and after the rotating shaft; Searching for the number of image spots; performing median filtering on the obtained original electrical level image to obtain a denoised electrical level image; according to the gray value, searching for a determined number of image spots in the denoised electrical level image, and determining respectively The central position of the searched image spot; in the denoising electric level image, sort and number according to the coordinates of the center position of the image spot in the x, y coordinate system; the same numbered image spot on the denoising electric level image Organize the data columns of the x-axis and y-axis coordinate values, and perform statistical analysis to determine whether there are abnormal data spots. A technical effect of the present invention is that, without adjusting the optical path and the mechanical structure, the image processing result can obtain a relatively stable processing result.

Description

A processing method of electric level image

technical field

The application belongs to the field of detection of astronomical instruments, and in particular relates to a processing method of electric level images.

Background technique

The multifunctional astronomical theodolite is a kind of astrometric instrument. In order to obtain high-precision measurement results, it is necessary to measure and deduct the influence of various instrument errors. The level difference is one of the main instrument errors, which mainly comes from the deflection of the instrument pier and the manufacturing error and installation error of the instrument. Also due to changes in ambient temperature and the effects of thermal deformation of the instrument, the level difference is constantly changing. Therefore, how to measure the level difference of astrometric instruments in real time is the key to improving the observation accuracy of the instruments. In order to measure the change of the level difference in real time, the multifunctional astronomical theodolite uses an electric level detector to measure the level difference. Its main principle is: using the mercury surface as the reference reflection surface, and using the analog CCD camera to obtain the electric level image before and after the rotating shaft of the instrument through self-collimation, the image usually contains a 3×3 image spot array (see Fig. 9 and Fig. 10).

However, due to the influence of the optical axis offset of the electric level autocollimation optical path, the oxidation of the mercury reflective surface, and the low quantum efficiency of the analog CCD camera configured, and the large background noise, the signal-to-noise ratio of the obtained electric level image is low, and the image The brightness of the spot is uneven, and the image is relatively inferior, which is not conducive to obtaining a stable processing result. Some stress changes in the optical-mechanical structure and the unsatisfactory adjustment results of the optical path can lead to the deviation of the optical path, resulting in inconsistent brightness of the image spots in the obtained electrical level images (as shown in Figure 1-8), which will cause some The signal-to-noise ratio of each image spot is too low, which leads to recognition errors (Figure 10), thus affecting the final centering result.

Therefore, it is necessary to provide a method for processing electrical level images.

Contents of the invention

An object of the present invention is to provide a new technical solution for an electrical level image processing method.

According to one aspect of the present invention, a method for processing an electrical level image provided by the present invention includes the following steps:

Obtain a preset number of raw level images before and after the pivot;

According to the original electrical level image, determine the x, y coordinate system and the number of image spots to be searched;

Perform median filtering on the obtained original electrical level images to obtain denoised electrical level images; the denoised electrical level images corresponding to the preset number of original electrical level images before the rotation are group A; the preset number of original electrical level images after the rotation The denoised electric level image corresponding to the level image is group B;

Searching for the determined number of image spots in the denoised electrical level image according to the gray value, and determining the center positions of the searched image spots respectively;

Sorting and numbering in the denoising electrical level image according to the coordinates of the central position of the image spot in the x, y coordinate system;

Organize the data columns of the x-axis and y-axis coordinate values of the same numbered image spots on the denoised electrical level images in group A and group B respectively, and perform statistical analysis to judge whether there are image spots with abnormal data; if they exist, remove them The x-axis and y-axis coordinate value data columns of the image spot, and the central position positioning data of the remaining image spots participate in the subsequent data processing; if they do not exist, the central position positioning data of all image spots participate in the subsequent data processing.

Optionally, the specific method for finding the determined number of image spots in the denoised electrical level image is: find the point with the largest gray value in the denoised electrical level image, and select the area of the centering area according to the size of the image spots , if the average value of the area is greater than the background value of the image, the point is considered as the target image spot, and the image spot is centered based on the x and y coordinate system data of the original electrical level image, and the pixel value of the image spot position in the filtered image is assigned 0 , and then look for the next image spot until the determined number is reached.

Optionally, the central position of the image spot is determined according to the following formula,

in:

x, y are the x, y coordinates on the image respectively, x ₀ , y ₀ are the center coordinate values after centering processing, I′(x, y) is the corrected pixel brightness value, I(x, y) is the original pixel brightness value, and T is the threshold value taken.

Optionally, the sorting and numbering method of the image spots is: sort according to the x coordinate value from small to large, and the y value is sorted with the x value to determine the number of columns; according to the number of image spots contained in each column, the y coordinate value is small Sort to large.

Optionally, the method for judging whether there is an abnormal data spot is as follows:

Eigenvalue m=[std_n_x_flag)] ² +[data_n_v_flag)] ² ;

Among them, std() is the solution of the standard deviation, and flag is the identification of the data before the rotation axis or the data after the rotation axis;

The meaning of the symbol in data_n_x_flag is: data_n_x represents the x-coordinate column of the center point of the nth image spot;

The meaning of the symbol in data_n_y_flag is: data_n_y indicates the y coordinate column of the center point of the nth image spot;

Determine the threshold n in combination with conventional non-abnormal data;

Threshold n=2×(avg(m _1～k )+3×std(m _1～k ));

Among them, avg() is the solution for the mean value, std() is the solution for the standard deviation, k is the number of no abnormal points, and m _1~k is the eigenvalue column without abnormal points;

When m≥n, the data of this data point is abnormal, the data of this data point will be eliminated, and will not participate in the subsequent data processing work;

When m<n, the data point is a valid point, which is reserved and participates in subsequent data processing.

A technical effect of the present invention is that, without adjusting the optical path and the mechanical structure, the image processing result can obtain a relatively stable processing result.

Description of drawings

The drawings described here are used to provide a further understanding of the application and constitute a part of the application. The schematic embodiments and descriptions of the application are used to explain the application and do not constitute an improper limitation to the application. In the attached picture:

Figure 1 is an example 1 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 2 is an example 2 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 3 is an example 3 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 4 is an example 4 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 5 is an example 5 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 6 is an example 6 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 7 is an example 7 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Figure 8 is an example 8 of the uneven brightness distribution of the acquired image spots caused by the bias of the optical path;

Fig. 9 is an example of correct recognition of an electric level image;

Figure 10 is an example of wrong recognition of a poor-quality electric level image;

Figure 11 is part of the electrical level image data series of target No. 0221 acquired on December 27, 2018;

Fig. 12 is the electric level data result of choosing 7 points in some embodiments;

Fig. 13 is the electric level data result of 6 points selected in some embodiments.

Detailed ways

The implementation of the present application will be described in detail below with reference to the accompanying drawings and examples, so as to fully understand and implement the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects.

The method for processing the electric level image provided by the present invention is applied to the electric level to measure the level difference. For the electric level image sequence captured by the terminal CCD or CMOS detector of the electric level system before and after the rotating shaft of the instrument, the image spot array is caused by the bias of the optical path. The brightness distribution is uneven, and some image spots are difficult to identify due to the low signal-to-noise ratio of low-quality electric level images. In some embodiments, the following steps are included:

Read the image data of a single original electrical level image, determine the x, y coordinate system, and determine the number of image spots to be searched through visual inspection. Taking the 3×3 image spot array shown in FIG. 9 as an example, among the 9 image spots, only 8 brighter image spots are searched.

Obtain a preset number of original electrical level images before and after the rotating shaft, that is, for astronomical instruments, such as multi-functional astronomical theodolite, continuously obtain a preset number of original electrical level images before the rotating shaft rotates, and then continuously obtain the preset number after the rotating shaft rotates Quantity of raw electrical level images. The denoised electrical level images corresponding to the preset number of original electrical level images before the rotation are group A; the denoised electrical level images corresponding to the preset number of original electrical level images after the rotation are group B. In normal operation, 100 original electric level images are taken before and after the rotating shaft respectively, totaling 200 original electric level images.

According to the gray value, a determined number of image spots are searched in each denoised electric level image, and the center positions of the searched image spots are respectively determined. The specific method for finding the determined number of image spots in the denoised electrical level image is: find the point with the largest gray value in the denoised electrical level image, select the area of the centering area according to the size of the image spot, if the average value of the area is greater than Image background value, consider this point as the target image spot, center the image spot based on the x and y coordinate system data of the original electrical level image, assign the pixel value of the image spot position of the filtered image to 0, and then search for the next image spot Spots until the determined number is reached. The central position of the image spot is determined according to the following formula,

in:

x and y are the x and y coordinates on the image respectively, x ₀ and y ₀ are the center coordinate values after centering processing respectively, I'(x, y) is the corrected pixel brightness value, I(x, y) is the original pixel brightness value, and T is the threshold value taken.

In the denoised electrical level image, the ordering and numbering are carried out according to the coordinates of the center positions of the image spots in the x, y coordinate system. The sorting and numbering method of the image spots is: sort according to the x coordinate value from small to large, and the y value is sorted according to the x value, and the number of columns is determined; according to the number of image spots contained in each column, the y coordinate value is sorted from small to large . Taking the 3×3 image spot array as an example, among the 9 image spots, only 8 brighter image spots are found as an example. First, the x coordinate values are sorted from small to large, and the y values are sorted according to the x values. In this way, the final sorting of the image spots is: the first column; the second column; the third column; after that, the number of image spots contained in each of the three columns is determined to be: 3, 3, and 2 respectively. According to the number of image spots contained in each column, the y coordinate values are sorted from small to large. The distribution of the final center point corresponds to the corresponding position in the 3×3 in the original image in an orderly manner.

Organize the data columns of the x-axis and y-axis coordinate values of the same numbered image spots on the denoised electrical level images in group A and group B respectively, and perform statistical analysis to judge whether there are image spots with abnormal data; if they exist, remove them The x-axis and y-axis coordinate value data columns of the image spot, and the central position positioning data of the remaining image spots participate in the subsequent data processing; if they do not exist, the central position positioning data of all image spots participate in the subsequent data processing.

The method of judging whether there is an abnormal data spot is as follows:

Eigenvalue m=[std(data_n_x_flag)] ² +[std(data_n_y_flag)] ² ;

Among them, std() is the standard deviation solution, and flag is the data before or after the rotation axis. For example, the data before the rotation is marked as zq, and the data after the rotation is marked as zh; the meaning of the symbols in data_n_x_flag is:

data_n_x indicates the x-coordinate column of the center point of the nth image spot; the meaning of the symbol in data_n_y_flag is:

data_n_y indicates the y coordinate column of the center point of the nth image spot;

Determine the threshold n in combination with conventional non-abnormal data;

Threshold n=2×(avg(m _1～k )+3×std(m _1～k ));

Among them, avg() is the solution for the mean value, std() is the solution for the standard deviation, k is the number of no abnormal points, and m _1~k is the eigenvalue column without abnormal points;

When m≥n, the data of this data point is abnormal, the data of this data point will be eliminated, and will not participate in the subsequent data processing work;

When m<n, the data point is a valid point, which is reserved and participates in subsequent data processing.

In other follow-up operations of the obtained data, those skilled in the art select the required data according to the image situation and the statistical analysis of the image spot data points.

Through the method of the present invention, it is possible to reduce the identification error caused by the low signal-to-noise ratio of a certain image spot, thereby affecting the final centering result, and reducing the influence of uneven brightness and darkness of the image spot caused by the bias of the optical path on the data processing result. Influenced, without adjusting the optical path and mechanical structure, a relatively stable processing result can be obtained.

In a specific embodiment 1, referring to Fig. 11, the electric level data of No. 0221 target acquired by the multifunctional astronomical theodolite of Yunnan Astronomical Observatory on December 27, 2018 is processed by using the processing method proposed by the present invention (only for illustration There are 19 original electrical level images before and after the rotating shaft, the actual data is 100 original electrical level images before and after the rotating shaft), and the threshold n is determined to be 30 according to the conventional data without abnormalities, as shown in Table 1.

Table 1 Statistics of the eigenvalues of the electricity level data of target No. 0221 on December 27, 2018

The calculated eigenvalue [std(data_7_x_zq)] ² + [std(data_7_y_zq)] ² before the rotation axis of data point 7 is 5434.9, which is far greater than the threshold value of 30. The value of this point is used in the subsequent calculation of the amount of change in the position of the image spot before and after the rotation axis. As shown in the results obtained by selecting 7 point data (data point 1 to data point 7) in Figure 12, it can be seen that the calculated electric level optical path The amount of change in the measurement coordinate position before and after the rotation axis of the image spot (Y-axis direction) has a large dispersion, showing obvious jumps. It can be judged that this point is an abnormal data point, which cannot truly reflect the change of the level difference before and after the rotation axis of the multi-functional astronomical theodolite , so it is necessary to remove the data of this point, see Figure 13, after the removal, use the data of 6 points (data point 1-data point 6) to calculate the change of the x-axis and y-axis coordinate values before and after the rotation axis, and the calculation result is shown in Figure 13 As shown, the results show that the dispersion of the changes in the x-axis and y-axis directions is small, which can more accurately reflect the change of the level difference of the multifunctional astronomical theodolite, and the calculation and processing results can further participate in subsequent data processing.

For example, certain terms are used in the description and claims to refer to specific components or methods. Those skilled in the art should understand that different regions may use different terms to refer to the same component. The description and claims do not use the difference in name as a way to distinguish components. As mentioned throughout the specification and claims, "comprising" is an open term, so it should be interpreted as "including but not limited to". "Approximately" means that within an acceptable error range, those skilled in the art can solve the technical problem within a certain error range and basically achieve the technical effect. The subsequent description of the specification is a preferred implementation mode for implementing the application, but the description is for the purpose of illustrating the general principle of the application, and is not intended to limit the scope of the application. The scope of protection of the present application should be defined by the appended claims.

It should also be noted that the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a good or system comprising a set of elements includes not only those elements but also includes items not expressly listed. other elements of the product, or elements inherent in the commodity or system. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the article or system comprising said element.

The above description shows and describes several preferred embodiments of the invention, but as previously stated, it should be understood that the invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other embodiments. Combinations, modifications and circumstances, and can be modified within the scope of the inventive concept described herein, by the above teachings or by skill or knowledge in the relevant field. However, changes and changes made by those skilled in the art do not depart from the spirit and scope of the invention, and should be within the protection scope of the appended claims of the invention.

Claims (1)

1. A method for processing an electric level image is characterized by comprising the following steps:

acquiring a preset number of original electric water level images in front of and behind the rotating shaft;

determining an x coordinate system, a y coordinate system and the number of image spots to be searched according to the original electric level image;

carrying out median filtering processing on the obtained original electric level image to obtain a de-noised electric level image; the denoising electric level images corresponding to the original electric level images with the preset number in front of the rotating shaft are A groups; b, taking the denoising electric water level images corresponding to the original electric water level images with the preset number behind the rotating shaft as a group B;

searching the determined number of image spots in the de-noising electric level image according to the gray value, and respectively determining the central positions of the searched image spots;

sequencing and numbering the de-noised electric quasi-images according to the coordinates of the central positions of the image spots in an x coordinate system and a y coordinate system;

respectively carrying out statistical analysis on data columns of coordinate values of an x axis and a y axis of the image spots with the same number on the de-noised electric quasi-images in the group A and the group B, and judging whether image spot points with abnormal data exist or not; if the image spot exists, the x-axis coordinate value data array and the y-axis coordinate value data array of the image spot are removed, and the central position positioning data of the residual image spots participate in the subsequent data processing work; if not, the central position positioning data of all the image spots participate in the subsequent data processing work;

the specific method for finding the determined number of image spots in the denoised electrical level image is as follows: finding a point with the maximum gray value in the de-noised electric level image, selecting the area of a centering area according to the size of an image spot, if the average value of the area is larger than the background value of the image, regarding the point as a target image spot, centering the image spot based on x and y coordinate coefficients of the original electric level image, assigning 0 to the pixel value of the image spot position of the filtered image, and then searching the next image spot until the determined number is reached;

the central position of the image spot is determined according to the following formula,

wherein:

x and y are x and y coordinates on the image, respectively ₀ 、y ₀ Respectively, the central coordinate values after centeringI' (x, y) is the modified pixel luminance value, I (x, y) is the original pixel luminance value, T is the threshold taken;

the image spot sequencing and numbering method comprises the following steps: sorting according to the x coordinate values from small to large, sorting the y values along with the x values, and determining the number of columns; sorting the image spots in each row from small to large according to the number of the image spots in each row;

the method for judging whether the image spots with abnormal data exist comprises the following steps:

characteristic value m- [ std (data _ n _ x _ flag)] ² +[std(data_n_y_flag)] ² ；

Wherein std () is standard deviation solution, and flag is data before the rotating shaft or data identification after the rotating shaft; the symbol meaning in data _ n _ x _ flag is: data _ n _ x represents an x coordinate column of the center point of the nth image spot; the symbol meaning in data _ n _ y _ flag is: data _ n _ y represents the y coordinate column of the center point of the nth image spot;

determining a threshold value n by combining conventional abnormal-free data;

threshold n =2 × (avg (m) _1～k )+3×std(m _1～k ))；

Wherein avg () is the mean solution, std () is the standard deviation solution, k is the number of outliers, m _1～k A characteristic value column without abnormal points;

when m is larger than or equal to n, the data point data is abnormal, the data point data is removed, and subsequent data processing work is not involved;

when m is less than n, the data point is a valid point and is reserved and participates in the subsequent data processing work.

Images