CN117649414B

CN117649414B - Textile auxiliary production wastewater treatment equipment

Info

Publication number: CN117649414B
Application number: CN202410123797.8A
Authority: CN
Inventors: 徐进云; 刘燕军; 吴玲; 李琼; 刘博建
Original assignee: Tianjin Industry University For Textile Auxiliaries Co ltd
Current assignee: Tianjin Industry University For Textile Auxiliaries Co ltd
Priority date: 2024-01-30
Filing date: 2024-01-30
Publication date: 2024-04-09
Anticipated expiration: 2044-01-30
Also published as: CN117649414A

Abstract

The invention relates to the technical field of wastewater treatment, in particular to textile auxiliary production wastewater treatment equipment, which further comprises a wastewater treatment result monitoring device, wherein the wastewater treatment result monitoring device comprises an image collector and a data controller, the image collector is in signal connection with the data controller, and the image collector is used for collecting water surface images after wastewater treatment is completed; the data controller is used for dividing the water surface image according to the brightness value of each pixel point in the received water surface image to obtain a highlight region; respectively acquiring gradient distribution characteristic values, gray level concentration degrees and texture direction distribution values of each highlight region; screening each highlight region to determine a non-water spray region; and monitoring the textile auxiliary production wastewater treatment result according to the non-water-bloom area. The invention ensures that the effect of the wastewater treatment result of textile auxiliary production wastewater treatment equipment is better.

Description

Textile auxiliary production wastewater treatment equipment

Technical Field

The invention relates to the technical field of wastewater treatment, in particular to textile auxiliary production wastewater treatment equipment.

Background

Textile assistants are a class of chemicals used to improve the performance of processes such as fiber, textile processing and dyeing. In the production of textile assistants, various chemicals and process steps may be involved, resulting in wastewater containing various pollutants. In order to meet the specifications, it is generally desirable to treat textile auxiliary process wastewater to reduce adverse environmental effects. The existing textile auxiliary production wastewater treatment equipment is used for treating textile auxiliary production wastewater usually through a plurality of stages, so that wastewater treatment results meet the specifications, but due to the complexity of pollutants in wastewater, the condition that the wastewater is not met with the specifications after treatment can occur, and further, the wastewater treatment results of the textile auxiliary production wastewater treatment equipment are poor.

Disclosure of Invention

In order to solve the technical problem that the waste water treatment effect of the existing textile auxiliary production waste water treatment equipment is poor, the invention aims to provide the textile auxiliary production waste water treatment equipment, and the adopted technical scheme is as follows:

in a first aspect, the invention provides a textile auxiliary waste water treatment apparatus comprising a processor and a memory, the processor being arranged to process instructions stored in the memory to effect the following monitoring process:

Acquiring a water surface image after wastewater treatment is completed, and dividing the water surface image according to the brightness value of each pixel point in the water surface image to obtain a highlight region; obtaining a gradient distribution characteristic value of each highlight region according to the edge gradient distribution condition of each highlight region and the gradient distribution condition of pixel points in the region;

obtaining the gray level concentration degree of each highlight region according to the concentration degree of the gray level distribution of the pixel points on the edge of each highlight region;

obtaining a texture direction distribution value of each highlight region according to the texture direction distribution in each highlight region and the texture distribution conditions of pixel points under different gray levels;

screening each highlight region according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value to determine a non-water-bloom region; and monitoring the textile auxiliary production wastewater treatment result according to the non-water-bloom area.

Preferably, the obtaining the texture direction distribution value of each highlight region according to the texture direction distribution in each highlight region and the texture distribution condition of the pixel points under different gray levels specifically includes:

determining gray level by utilizing gradient distribution characteristic values of each highlight region, and for any highlight region, performing block processing on the highlight region to obtain a sub-region to be analyzed, and constructing a gray level co-occurrence matrix of each sub-region to be analyzed in each set direction based on the gray level of each pixel point in each sub-region to be analyzed;

For any sub-region to be analyzed, the entropy value of the gray level co-occurrence matrix in each set direction is formed into a gray level texture sequence of the sub-region to be analyzed;

taking each sub-region to be analyzed as an initial seed point, and carrying out region growth based on the difference condition of gray texture sequences among the sub-regions to be analyzed to obtain a characteristic sub-region; calculating the variance of the areas of all the characteristic subareas in the highlight area to obtain a first characteristic coefficient;

when the number of the characteristic subareas contained in the highlight area is smaller than a preset number threshold, setting the value of the second characteristic coefficient to be a preset numerical value; when the number of the feature subareas contained in the highlight area is larger than or equal to a preset number threshold, taking a normalized value of the number of the feature subareas contained in the highlight area as a second feature coefficient, wherein the second feature coefficient is larger than a preset value;

and obtaining a texture direction distribution value of the highlight region according to the first characteristic coefficient and the second characteristic coefficient, wherein the first characteristic coefficient and the second characteristic coefficient are in positive correlation with the texture direction distribution value.

Preferably, the region growing is performed based on a difference condition of gray texture sequences between the subregions to be analyzed, so as to obtain a characteristic subregion, which specifically includes:

For any two sub-areas to be analyzed, calculating a normalized value of cosine similarity between gray texture sequences of the any two sub-areas to be analyzed to obtain the similarity of the any two sub-areas to be analyzed;

if the similarity degree between the gray texture sequences of the sub-region to be analyzed and the sub-region to be analyzed in the neighborhood is larger than a preset similarity threshold, growing is carried out until the similarity threshold is not met, and the characteristic sub-region is obtained.

Preferably, the partitioning processing is performed on the highlight region to obtain a sub-region to be analyzed, which specifically includes:

and acquiring a minimum circumscribed rectangle of the highlight region, recording the minimum circumscribed rectangle as a rectangle of the highlight region, and dividing the rectangle of the highlight region into a preset number of sub-regions to be analyzed.

Preferably, the obtaining the gradient distribution characteristic value of each highlight region according to the gradient distribution condition of the edge of each highlight region and the gradient distribution condition of the pixel points in the region specifically includes:

for any highlight region, edge detection is carried out on the highlight region to obtain an edge of the highlight region, the edge is marked as an outer edge, the region except the outer edge in the highlight region is used as an inner region of the highlight region, and edge detection is carried out on the inner region of the highlight region to obtain an edge of the inner region, and the edge is marked as an inner edge;

Obtaining the range of the gradient value of each pixel point on the inner edge, and carrying out normalization processing on the range to obtain a first coefficient; taking the total number of all different values of the gradient values of all pixel points on the inner edge as a second coefficient; obtaining gradient distribution characteristic values of the highlight region according to the first coefficient and the second coefficient; the first coefficient and the second coefficient are in positive correlation with the gradient distribution characteristic value.

Preferably, the obtaining the gray level concentration degree of each highlight region according to the aggregation degree of the gray level distribution of the pixel points on the edge of each highlight region specifically includes:

for any highlight region, acquiring gradient value types of pixel points on the outer side edge of the highlight region, and forming an edge gray sequence by gray values of all the pixel points on the outer side edge of the highlight region;

marking pixel points on the outer edge of a highlight region corresponding to any gradient value as characteristic edge points, acquiring interpolation pixel values of other pixel points except the characteristic edge points on the outer edge of the highlight region by using a bilinear interpolation method based on pixel coordinates and gray values of each characteristic edge point, and forming interpolation edge sequences of the gradient values of the corresponding types of the characteristic edge points; the positions of the pixel points in the interpolation edge sequence and the edge gray sequence are in one-to-one correspondence;

Calculating the similarity between the interpolation edge sequence and the edge gray sequence to obtain a distribution evaluation coefficient of gradient values of the corresponding types of the characteristic edge points; and calculating the average value of the distribution evaluation coefficients of all kinds of gradient values on the outer side edge of the highlight region to obtain the gray level concentration degree of the highlight region.

Preferably, the screening the highlight area according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value to determine the non-water spray area specifically includes:

for any highlight region, obtaining the water spray characteristic degree of the highlight region according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value of the highlight region; the gradient distribution characteristic value and the texture direction distribution value are in positive correlation with the water spray characteristic degree, the gray concentration degree and the water spray characteristic degree are in negative correlation, and the water spray characteristic degree is a normalized numerical value;

and marking the highlight area corresponding to the water spray characteristic degree smaller than or equal to the preset characteristic threshold value as a non-water spray area.

Preferably, the determining the gray level by using the gradient distribution characteristic value of each highlight region specifically includes:

for any highlight region, carrying out negative correlation mapping on gradient distribution characteristic values of the highlight region to obtain a quantity coefficient; the number coefficients are rounded upwards to obtain the initial gray level number of the highlight region; calculating the average value of the initial gray level number of all the highlight areas to obtain the preferred gray level number; and dividing gray scales of gray values of all pixel points in the surface image by using the preferred gray scale number to obtain gray scales.

Preferably, the monitoring of the textile auxiliary production wastewater treatment result according to the non-water pattern area specifically comprises the following steps:

the image of the non-water mark area is a water surface gray image, the non-water mark area in the water surface gray image is enhanced, the enhanced water surface gray image is obtained, and the textile auxiliary production wastewater treatment result is monitored based on the enhanced water surface gray image; the method for acquiring the water surface gray level image comprises the step of carrying out gray level processing on the water surface image to obtain the water surface gray level image.

In a second aspect, the invention provides textile auxiliary production wastewater treatment equipment, which comprises a textile auxiliary production wastewater treatment device, wherein the textile auxiliary production wastewater treatment equipment further comprises a wastewater treatment result monitoring device, the wastewater treatment result monitoring device comprises an image collector and a data controller, the image collector is in signal connection with the data controller, and the image collector is used for collecting water surface images after wastewater treatment is completed;

the data controller is used for dividing the water surface image according to the brightness value of each pixel point in the received water surface image to obtain a highlight region;

obtaining a gradient distribution characteristic value of each highlight region according to the edge gradient distribution condition of each highlight region and the gradient distribution condition of pixel points in the region;

determining gray levels by utilizing the gradient distribution characteristic values, and obtaining a texture direction distribution value of each highlight region according to texture direction distribution in each highlight region and texture distribution conditions of pixel points under each gray level;

The embodiment of the invention has at least the following beneficial effects:

the data processing process realized by the device of the invention firstly divides the water surface image after wastewater treatment is completed to obtain a highlight region; the highlight region represents a region in the image where a water spray feature may exist, and may be a highlighted target region, or may be a highlight portion affected by illumination. And further analyzing the gradient distribution condition of the edge of each highlight region and the gradient distribution condition of the pixel points in the region to obtain the gradient distribution characteristic value of each highlight region, wherein the gradient distribution characteristic value reflects the complex condition of gradient distribution on the edge of the highlight region. Then, analyzing the aggregation degree of the gray level distribution of the pixel points on the edge of each highlight region to obtain the gray level concentration degree of each highlight region; the gray level concentration level of the highlight region reflects the concentration level of the gray value distribution of the pixel point on the edge of the highlight region. Further, analyzing the texture direction distribution in each highlight region and the texture distribution condition of the pixel points under different gray levels to obtain a texture direction distribution value of each highlight region; the texture direction distribution value reflects the complexity of the directional texture distribution in the highlight region. Finally, combining the texture distribution characteristics of the highlight areas in three aspects, screening each highlight area, removing the splash areas accurately, obtaining a non-splash area, avoiding the condition that the splash areas influence the evaluation of the wastewater treatment results in subsequent images, further monitoring the textile auxiliary production wastewater treatment results based on the non-splash areas, obtaining more accurate monitoring results, and further enabling related staff to take different measures according to different monitoring results, so that the wastewater treatment results of the textile auxiliary production wastewater treatment equipment are better.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a data processing process of a data controller in textile auxiliary production wastewater treatment equipment.

Detailed Description

In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the textile auxiliary production wastewater treatment equipment according to the invention with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

The following specifically describes a specific scheme of textile auxiliary production wastewater treatment equipment provided by the invention with reference to the accompanying drawings.

An embodiment one of textile auxiliary production wastewater treatment equipment:

this embodiment provides a textile auxiliary waste water treatment facility, including textile auxiliary waste water treatment facilities, it can be understood that textile auxiliary waste water treatment facilities is common waste water treatment facilities for handle textile auxiliary waste water, for example, application number 202121190799.7, the patent of a "textile auxiliary waste water zero release treatment facility" can regard as textile auxiliary waste water treatment facilities.

The textile auxiliary production wastewater treatment equipment further comprises a wastewater treatment result monitoring device, wherein the wastewater treatment result monitoring device comprises an image collector and a data controller. The image collector can be a conventional industrial camera and is used for shooting the water surface processed by the wastewater treatment device to obtain a water surface image.

The data controller may be a data processing chip such as CPU, MCU, FPGA or a data processing device such as a host computer. The image collector is in signal connection with the data controller, and the image collector and the data controller can be connected in a wired mode through a data transmission line, and can also be connected in a wireless mode through Bluetooth, wiFi and other wireless communication modes. In addition, the image collector and the controller can be integrated to form a device integrating the functions of image collection and image processing.

The image collector is used for collecting the water surface image after wastewater treatment is completed, and it can be understood that the image collected by the image collector is an RGB image, that is, the water surface image is an RGB image, and the image collector outputs the collected sleep image to the data controller, as shown in fig. 1, which is a schematic flow chart of the data processing process of the data controller in the textile auxiliary production wastewater treatment equipment.

Dividing the water surface image according to the brightness value of each pixel point in the water surface image to obtain a highlight region; and obtaining the gradient distribution characteristic value of each highlight region according to the edge gradient distribution condition of each highlight region and the gradient distribution condition of the pixel points in the region.

In practice, the conventional method for monitoring whether the wastewater treatment is qualified is to process the water surface image through threshold segmentation, screen out the area with texture part in the water surface image, further identify the characteristics of the area such as color tone, texture and the like, and judge whether the wastewater treatment is qualified. However, due to the existence of illumination or water bloom, the threshold segmentation result may be affected, and the accuracy of the judgment result is lower. The illumination part belongs to the highlight part, so that the illumination part is easily confused with the target highlight part in the water surface image, and accurate segmentation is difficult. And because of the white area part where the water bloom appears, the characteristic information of the area is not considered when the wastewater treatment result is monitored, and the interference of misleading characteristics can be reduced. Because the water spray is not the characteristic of the water quality, but is a dynamic phenomenon body generated by the flow and mutual collision of the water body, the area part where the water spray is positioned covers the inherent attribute of the wastewater, such as color, suspended matters, grease layers and the like, so that the accuracy of monitoring the wastewater treatment result can be effectively improved by neglecting the characteristic of the area part when the wastewater treatment result is monitored.

Considering that the acquired water surface image may be affected by illumination and water bloom, the accuracy of the subsequent judgment result is not high, so that the water bloom region and the illumination region in the water surface image can be distinguished, further, the feature analysis of the water bloom region can be avoided, the enhancement processing is carried out on the illumination region, and the judgment result is more accurate.

Firstly, dividing a water surface image according to the brightness value of each pixel point in the water surface image to obtain a highlight region, wherein the water surface image is an RGB image, converting the water surface image into an HSV image, obtaining an S channel image in the HSV image, processing the S channel image by using an Ojin threshold segmentation method to obtain the highlight pixel point, analyzing the connected domain of the highlight pixel point to obtain a plurality of highlight connected domains in the S channel image, carrying out gray-scale treatment on the water surface image to obtain a water surface gray-scale image, and further marking the corresponding region of the highlight connected domains in the water surface gray-scale image as the highlight region.

And acquiring an optimal brightness threshold for dividing all pixel points in the S channel image by using an Ojin threshold dividing method, and marking the pixel points with the pixel values larger than the optimal brightness threshold in the S channel image as highlight pixel points. The highlight region thus obtained may be a highlight region of a portion subjected to light or a highlight region of a portion where the water bloom is located.

It should be noted that, since the water spray usually presents a dynamic, irregular shape, and the boundary between the edge of the area where the water spray is located and the surrounding water body is less clear, the area affected by the light usually has a certain regularity. For example, sunlight reflects from the water surface to form bright spots resembling specular reflection, and the edges of the area where the bright spots are located are more regular and more clearly defined with the surrounding water body. In contrast, the smooth transition of the edge of the area where the water spray is located is stronger, and the edge limit is less obvious. The edge transitivity refers to whether the gray value of the pixel point has a mutation phenomenon, and the more the edge pixel point with mutation, the weaker the transitivity of the edge part is.

Then, the edge distribution condition of each highlight area to be distinguished is analyzed respectively, the gray distribution of the pixel points on the edge of the area where the water spray is located is complex, the gray distribution of the pixel points on the edge of the illumination partial area is simple and single, and the gradient value reflects the gray change condition of the position where the pixel points are located. Based on the above, the gradient distribution characteristic value of each highlight region is obtained according to the gradient distribution condition of the edge of each highlight region and the gradient distribution condition of the pixel points in the region.

Specifically, for any highlight region, edge detection is performed on the highlight region to obtain an edge of the highlight region, the edge is marked as an outer edge, a region except for an outer edge in the highlight region is used as an inner region of the highlight region, and edge detection is performed on the inner region of the highlight region to obtain an edge of the inner region, and the edge is marked as an inner edge; obtaining the range of the gradient value of each pixel point on the inner edge, and carrying out normalization processing on the range to obtain a first coefficient; taking the total number of all different values of the gradient values of all pixel points on the inner edge as a second coefficient; obtaining gradient distribution characteristic values of the highlight region according to the first coefficient and the second coefficient; the first coefficient and the second coefficient are in positive correlation with the gradient distribution characteristic value.

It should be noted that, the process of edge detection on an image is a well-known technique, and will not be described herein too much, and meanwhile, a gradient value of each pixel point on the edge of each highlight region may be obtained. The inner region and the outer edge communicate to form a highlight region, the outer edge can represent the outermost edge part of the highlight region, the inner region can represent the region part except the outermost edge, and the inner edge and the outer edge can be regarded as progressive distribution.

If the highlight region is the region where the water spray is located, the gray value distribution of the pixel points on the outer edge and the inner edge of the highlight region is complex, so that the gray value of the pixel points has more different change conditions, and the gradient value distribution of the pixel points is complex. If the highlight region is the region where the illumination part is located, the outer edge of the highlight region belongs to a more regular edge, the inner edge is the highlight point of the illumination part, the gray values of the pixel points on the inner edge are more similar, so that the gray values of the pixel points do not have more change conditions, and the gradient values of the pixel points are simpler and more single in distribution.

In this embodiment, taking any highlight region as an example, the calculation formula of the gradient distribution characteristic value of the highlight region can be expressed as:

wherein,gradient distribution characteristic value representing the ith highlight region,/->Maximum value of gradient value representing pixel point on inner edge of ith highlight region,/and>minimum value representing gradient value of pixel point on inner edge of ith highlight region,/->Representing the total number of all different values of the gradient values of all pixel points on the inner edge of the ith highlight region.

The range of gradient distribution on the inner edge of the highlight region is reflected by the extremely poor gradient value of each pixel point on the inner edge, and the larger the extremely poor gradient value is, the first coefficient +.>The larger the value of (a) is, the larger the distribution range of the gradient value of the pixel point on the inner edge of the highlight region is, and the larger the richness of the gradient value distribution of the pixel point on the inner edge is, so that the inside is further illustratedThe gray distribution of the pixel points on the edge is complex, and the larger the corresponding gradient distribution characteristic value is, the larger the probability that the corresponding highlight region is the region where the water bloom is located is.

The second coefficient indicates the number of kinds of gradient values of the pixel points on the inner edge of the highlight region, for example, if the gradient values of the four pixel points on the inner edge are 10,10,15,20, the corresponding number of kinds of gradient values is 3, that is, the total number of all different gradient values of the four pixel points on the inner edge is 3./>The complex condition of gradient value distribution of the pixel points of the inner edge row is reflected, and the larger the value is, the more complex the gradient value distribution of the pixel points on the inner edge is, and the larger the probability that the corresponding highlight area is the area where the water bloom is located is further.

The gradient distribution characteristic value reflects the complexity of gradient value distribution on the edge of the highlight region and the complexity of gray scale distribution of the pixel point. The larger the gradient distribution characteristic value is, the more complex the gray distribution of the edge characteristics of the highlight region is, and the more complex the gradient distribution is, the greater the possibility that the corresponding highlight region is the region where the water spray is located is.

And step two, obtaining the gray level concentration degree of each highlight region according to the concentration degree of the gray level distribution of the pixel points on the edge of each highlight region.

Because more gradient values are distributed in a staggered manner at the edge of the area where the water bloom is located, the edge of the area where the water bloom is located is less obvious in an image, the continuity of gray values of pixel points on the edge of the area where the water bloom is located is poor, the distribution concentration of the pixel points with the same gray level is small, and the aggregation degree of the gray distribution of the pixel points on the edge of each highlight area is analyzed to obtain the gray concentration degree of each highlight area.

Considering that the running of the calculated pixel point can reflect the gray continuity of the pixel point, but ignoring the condition that the gray values are relatively close, a certain deviation exists in the analysis result. Therefore, the present embodiment adopts bilinear interpolation to process the pixel points with the same gray value. Specifically, the pixel points on the outer side edge of the highlight region corresponding to any one gradient value are marked as characteristic edge points, interpolation pixel values of other pixel points on the outer side edge of the highlight region except the characteristic edge points are obtained by utilizing a bilinear interpolation method based on the pixel coordinates and the gray value of each characteristic edge point, and the gray value of the characteristic edge point on the outer side edge of the highlight region and the interpolation pixel values of other pixel points form an interpolation edge sequence of the gradient values of the corresponding type of the characteristic edge points; and the positions of the pixel points in the interpolation edge sequence and the edge gray sequence are in one-to-one correspondence.

Wherein the object of analyzing the degree of aggregation of the gray scale distribution on the edge of the highlight region is the outer edge. On the outer edge of the highlight region, assuming that a pixel point with a gradient value of 15 is marked as a characteristic edge point, and any one pixel point on the outer edge of the highlight region except the characteristic edge point is marked as a pixel point to be processed, the main idea of obtaining the interpolation pixel value of the pixel point to be processed by using the bilinear interpolation method is to obtain a plurality of characteristic edge points closest to the pixel point to be processed on the outer edge for linear interpolation calculation, and the calculation process is a known technology and is not described in more detail herein.

Then, on the outer edge of the highlight region, the gray value of the feature edge point and the interpolation pixel value of the other pixel point form an interpolation edge when the gradient value is 15, the interpolation edge reflects the continuous gray distribution condition of the corresponding pixel point when the gradient value is 15, and the gray value of the feature edge point and the interpolation pixel value of the other pixel point on the interpolation edge form an interpolation edge sequence when the gradient value is 15.

In this embodiment, the pixel point corresponding to the maximum value of the gradient values on the interpolation edge is taken as a starting point, the gray value or the interpolation pixel value of each pixel point is sequentially obtained according to the anticlockwise direction, and finally a difference edge sequence is formed, based on the difference edge sequence, the sequence formed by the gray values of all the pixel points on the outer edge of the highlight region, namely, the edge gray sequence, is in one-to-one correspondence with the positions of the pixel points in the difference edge sequence, so when the distribution evaluation coefficient corresponding to the gradient value of 15 is calculated, the arrangement sequence of the gray values of the pixel points in the edge gray sequence needs to be according to the position arrangement sequence of the pixel points in the interpolation edge sequence when the gradient value of 15 is calculated.

In other embodiments, on the outer edge of any highlight region, gray values of each pixel point can be sequentially acquired with any pixel point as a starting point according to a clockwise or anticlockwise direction to form an edge gray sequence, and then the edge gray sequences are arranged according to the same sequence when the interpolation edge sequence of the interpolation edge is acquired.

Further, calculating the similarity between the interpolation edge sequence and the edge gray sequence to obtain a distribution evaluation coefficient of the gradient value of the corresponding type of the characteristic edge point; and calculating the average value of the distribution evaluation coefficients of all kinds of gradient values on the outer side edge of the highlight region to obtain the gray level concentration degree of the highlight region.

In this embodiment, the cosine similarity between the interpolation edge sequence and the edge gray sequence is used as the similarity between the interpolation edge sequence and the edge gray sequence, and in other embodiments, the pearson correlation coefficient between the interpolation edge sequence and the edge gray sequence may be used as the similarity between the interpolation edge sequence and the edge gray sequence, so that the implementer may select according to the specific implementation scenario.

The interpolation edge sequence reflects the continuous gray distribution condition of the pixel points on the outer edge of the highlight region under the corresponding gradient value, the edge gray sequence reflects the gray distribution condition of the pixel points on the outer edge of the highlight region, the greater the similarity between the two sequences is, the greater the continuity of the original gray distribution condition on the outer edge of the highlight region is, the closer the gray distribution on the outer edge is to the condition of continuous distribution, and further the greater the gray aggregation degree of the pixel points on the outer edge of the highlight region is, the greater the value of the gray aggregation degree of the corresponding highlight region is, and the lower the possibility that the corresponding highlight region belongs to the region where the water bloom is.

It should be noted that, under each gradient value of all pixel points on the outer edge of the highlight region, a corresponding distribution evaluation coefficient can be calculated, so as to reflect the continuity degree and aggregation degree of gray distribution of the outer edge of the highlight region under different kinds of gradient values. Finally, the gray level concentration degree of the highlight region reflects the aggregation degree and the continuity degree of the gray value distribution of the pixel points on the outermost edge of the highlight region, and the larger the value of the gray level concentration degree is, the smaller the possibility that the corresponding highlight region belongs to the region where the water bloom is located is.

And thirdly, obtaining the texture direction distribution value of each highlight region according to the texture direction distribution in each highlight region and the texture distribution condition of the pixel points under different gray levels.

It should be noted that, the texture features of the area where the water spray is located are complex, various irregular water waves are included, and the texture features of the area belonging to the area with strong illumination are simple and uniform. Because of the complex gradient distribution, the corresponding texture direction distribution is also complex, in order to avoid the interference of the complexity of the gradient distribution on the feature analysis of the texture distribution direction distribution, the index reflecting the complexity of the gradient distribution is used as the condition of conditional probability, and the directionality of the texture under the condition can only more accurately represent the directionality of the water bloom texture. The larger the gradient distribution characteristic value is, the more complex the gradient distribution is on the edge of the highlight region, the wider the gradient distribution range is, and the influence of the gradient distribution on the direction texture analysis process can be avoided by setting a smaller number of gray scales.

Firstly, dividing gray scales of gray values of pixel points of each highlight region according to gradient distribution characteristic values of each highlight region, specifically, carrying out negative correlation mapping on the gradient distribution characteristic values of the highlight region for any highlight region to obtain a quantity coefficient; the number coefficients are rounded upwards to obtain the initial gray level number of the highlight region; calculating the average value of the initial gray level number of all the highlight areas to obtain the preferred gray level number; and dividing gray scales of gray values of all pixel points in the surface image by using the preferred gray scale number to obtain gray scales.

In this embodiment, the gradient distribution feature values of the highlight region are normalized, and then the inverse of the normalized gradient distribution feature values is calculated to perform negative correlation mapping on the gradient distribution feature values of the highlight region, and meanwhile, functions and methods of the negative correlation mapping are various, such as logarithmic functions and exponential functions, but the negative correlation mapping needs to consider the value range of the number coefficient after mapping, and the value range of the number coefficient should be greater than 1.

Meanwhile, it can be understood that the gradient distribution characteristic value and the number of gray levels are in a negative correlation, the larger the gradient distribution characteristic value is, the smaller the number of corresponding gray levels is, namely the smaller the number of gray levels is, and the characteristic analysis process of influencing the directional texture widely in the gradient distribution range is avoided by dividing the gray values of the pixel points in the highlight region into the smaller gray levels.

In other embodiments, the number of gray levels of the gray value of the pixel point in each highlight region may be directly set to 8, that is, the number of gray levels is directly set to be smaller, so that other interference factors are eliminated.

Then, for any highlight region, partitioning the highlight region to obtain sub-regions to be analyzed, and constructing a gray level co-occurrence matrix of each sub-region to be analyzed in each set direction based on gray level of each pixel point in each sub-region to be analyzed. Specifically, the minimum circumscribed rectangle of the highlight region is acquired and recorded as a rectangular region of the highlight region, and the rectangular region of the highlight region is divided into a preset number of sub-regions to be analyzed. In this embodiment, the size of the sub-area to be analyzed is set to define the preset number of sub-areas to be analyzed, that is, the rectangular area of the highlight area is divided into a plurality of window areas with the size of 21×21, and the window areas are recorded as sub-areas to be analyzed. In other embodiments, the plurality of sub-regions to be analyzed may also be obtained by uniformly dividing a rectangular region of the highlight region.

In this embodiment, the setting directions include 16 directions, which are 0 °, 22.5 °, 45 °, 67.5 °, 90 °, 112.5 °, 135 °, 157.5 °,180 °,202.5 °,225 °,247.5 °,270 °,292.5 °,315 °,337.5 °, and an operator may set according to a specific implementation scenario, and each sub-area to be analyzed has a corresponding gray level co-occurrence matrix in each direction. The construction method of the gray level co-occurrence matrix is a well-known technique, and will not be described here too much.

For any one sub-region to be analyzed, the entropy value of the gray level co-occurrence matrix in each set direction is formed into a gray level texture sequence of the sub-region to be analyzed, namely the gray level texture sequence contains entropy values corresponding to the sub-region to be analyzed in 16 directions.

Further, each sub-region to be analyzed is used as an initial seed point, region growth is carried out based on the difference condition of gray texture sequences among the sub-regions to be analyzed, and a characteristic sub-region is obtained; and calculating the variance of the areas of all the characteristic subareas in the highlight area to obtain a first characteristic coefficient. The process of region growth specifically comprises the following steps: for any two sub-areas to be analyzed, calculating a normalized value of cosine similarity between gray texture sequences of the any two sub-areas to be analyzed to obtain the similarity of the any two sub-areas to be analyzed; if the similarity degree between the gray texture sequences of the sub-region to be analyzed and the sub-region to be analyzed in the neighborhood is larger than a preset similarity threshold, growing is carried out until the similarity threshold is not met, and the characteristic sub-region is obtained.

In this embodiment, each sub-region to be analyzed is regarded as each initial seed point, that is, each window region is regarded as an initial seed point, and region growing operation is performed on the sub-region to be analyzed according to a set region growing rule. In this embodiment, the value of the similarity threshold is 0.7, and the implementer can set according to the specific implementation scenario.

The degree of similarity of any two sub-regions to be analyzed reflects the two sub-regions to be analyzed. When the similarity degree between the gray texture sequences of the sub-region to be analyzed and the sub-region to be analyzed in the adjacent region is larger than 0.7, the gray direction texture distribution conditions of the two sub-regions to be analyzed are similar, and then the two sub-regions with molecules can be combined, namely region growth is carried out. When the sub-region to be analyzed with the similarity degree larger than the similarity threshold value does not exist in the neighborhood of the sub-region to be analyzed, the growth can be stopped, and then the result of the growth of the region can be obtained and is recorded as a characteristic sub-region.

It will be appreciated that the gray-scale directional texture features of the sub-regions to be analyzed contained within each feature sub-region are relatively similar, and that each feature sub-region within the highlight region characterizes a region having different gray-scale directional texture features. The variance of the areas of all the characteristic subareas in the highlight region is calculated, the larger the variance is, the larger the different areas among the characteristic subareas in the highlight region are, the larger the corresponding first characteristic coefficient value is, and the more different gray scale direction textures are contained in the highlight region.

When the number of the characteristic subareas contained in the highlight area is smaller than a preset number threshold, setting the value of the second characteristic coefficient to be a preset numerical value; and when the number of the characteristic subareas contained in the highlight region is larger than or equal to a preset number threshold, taking the normalized value of the number of the characteristic subareas contained in the highlight region as a second characteristic coefficient, wherein the second characteristic coefficient is larger than a preset value.

In this embodiment, the average value of the number of feature sub-areas included in all the highlight areas is calculated and used as a preset number threshold, when the number of feature sub-areas included in the highlight areas is smaller than the number threshold, it is indicated that the number of feature sub-areas included in the highlight areas is smaller at this time, and further, that different gray-scale directional textures included in the highlight areas are smaller, directional texture distribution is simpler and single, the probability that the highlight areas belong to the area where the water spray is smaller, further, in this embodiment, the second feature coefficient of the highlight areas is set to a fixed value, that is, the value of the preset value is 0.1, and the value of the preset value can also be set to a minimum positive number such as 0.01 according to a specific implementation scenario.

When the number of the feature subareas contained in the highlight area is larger than or equal to the number threshold value, the number of the feature subareas contained in the highlight area is larger, and further, the fact that the number of the feature subareas contained in the highlight area is larger is indicated, namely, the direction texture distribution is more complex, the probability that the highlight area belongs to the area where the water spray is located is indicated, the normalized value of the number of the feature subareas contained in the highlight area is taken as the second feature coefficient, and the number of the feature subareas contained in the highlight area is larger, and the value of the corresponding second feature coefficient is larger.

And finally, obtaining a texture direction distribution value of the highlight region according to the first characteristic coefficient and the second characteristic coefficient, wherein the first characteristic coefficient and the second characteristic coefficient are in positive correlation with the texture direction distribution value. In this embodiment, for any one highlight region, the product of the first characteristic coefficient and the second characteristic coefficient corresponding to the highlight region is taken as the texture direction distribution value of the highlight region. The texture direction distribution value reflects the complexity of the direction texture distribution in the highlight region, and the larger the value is, the more complex the direction texture distribution in the highlight region is, and the smaller the value is, the simpler and the single the direction texture distribution in the highlight region is.

Step four, screening each highlight region according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value, and determining a non-water-bloom region; and monitoring the textile auxiliary production wastewater treatment result according to the non-water-bloom area.

The gradient distribution characteristic value reflects the gradient distribution condition of the inner edge of the highlight region, the gray level concentration degree reflects the gray level distribution aggregation condition of the outer edge of the highlight region, and the texture direction distribution value reflects the complexity condition of the directional texture distribution in the highlight region. The gradient distribution of the inner edge of the highlight region is more complex, the gray level distribution of the outer edge is less dense, and when the texture distribution in the inner direction is more complex, the probability that the corresponding highlight region is the region where the water spray is located is higher.

Based on the above, the texture distribution characteristics of the highlight regions in three aspects are combined, and each highlight region is screened according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value, so that the non-splash region is determined.

Specifically, for any highlight region, obtaining the water spray characteristic degree of the highlight region according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value of the highlight region; the gradient distribution characteristic value and the texture direction distribution value are in positive correlation with the water spray characteristic degree, the gray concentration degree and the water spray characteristic degree are in negative correlation, and the water spray characteristic degree is a normalized numerical value; and marking the highlight area corresponding to the water spray characteristic degree smaller than or equal to the preset characteristic threshold value as a non-water spray area.

In the embodiment, taking any highlight region as an example for explanation, a calculation formula of the water spray characteristic degree of the highlight region may be expressed as:

wherein,water spray feature degree indicating the ith highlight region, +.>Gradient distribution characteristic value representing the ith highlight region,/->Gray concentration degree indicating i-th highlight region,/, and >Representing the texture direction distribution value of the i-th highlight region, exp () represents an exponential function based on a natural constant e, and Norm () represents a linear normalization function.

The smaller the gradient distribution characteristic value of the highlight region is, the simpler and single gray distribution of the edge characteristic of the highlight region is, the smaller the gradient distribution is, and the smaller the value of the water spray characteristic degree is, the smaller the possibility that the corresponding highlight region belongs to the region where the water spray is. The smaller the value of the texture direction distribution value of the highlight region, the simpler and more single the direction texture distribution in the highlight region is, the smaller the value of the water spray characteristic degree is, and the lower the probability that the corresponding highlight region belongs to the region where the water spray is.

The larger the value of the gray level concentration degree of the highlight region, the larger the gray level concentration degree of the pixel points on the outer side edge of the highlight region is, the negative correlation mapping is carried out on the gray level concentration degree by utilizing an exponential function, and the smaller the value of the water spray characteristic degree is, the smaller the possibility that the corresponding highlight region belongs to the region where the water spray is located is. The water spray characteristic degree of the highlight region characterizes the probability that the highlight region belongs to the water spray region, and the smaller the value is, the smaller the probability that the highlight region belongs to the water spray region is.

And on the basis of the characteristic degree of the water spray, the highlight area corresponding to the characteristic threshold value with the characteristic degree smaller than or equal to the preset characteristic threshold value is recorded as a non-water spray area. In this embodiment, the value of the feature threshold is 0.8, and the implementer can set according to the specific implementation scenario. When the water spray characteristic degree of the highlight region is smaller than or equal to 0.8, the probability that the highlight region belongs to the region where the water spray is located is smaller, the highlight region possibly belongs to the part affected by illumination, and the part can be preprocessed, so that the texture characteristics of the highlight region are more obvious, and the wastewater treatment result is analyzed based on the image.

And finally, monitoring the textile auxiliary production wastewater treatment result according to the non-splash area. Specifically, the image in which the non-water mark region is located is a water surface gray level image, the non-water mark region in the water surface gray level image is enhanced, and the enhanced water surface gray level image is obtained.

And monitoring the textile auxiliary production wastewater treatment result based on the enhanced water surface gray level image. In the embodiment, a convolutional neural network is adopted to process the enhanced water surface gray level image, and an evaluation result of textile auxiliary production wastewater treatment results is obtained. The method for evaluating the wastewater treatment result based on the gray level image of the wastewater surface by using the neural network is a common technology in the field, for example, the invention patent with the application number of 202310283757.5, named as a sewage treatment plant working condition evaluation method and system based on the self-organizing neural network, discloses a method for evaluating the wastewater treatment result by using the neural network.

Meanwhile, after the sewage evaluation result is obtained, related staff can take different measures according to different sewage treatment evaluation results, so that the effect of the wastewater treatment result of the textile auxiliary production wastewater treatment equipment is better. For example, the evaluation result of sewage treatment is that the sewage treatment result is not qualified, and the related staff needs to reprocess the wastewater, i.e. needs to continue purifying. The sewage treatment evaluation result is that the sewage treatment result is qualified, and related staff can discharge the wastewater qualified in treatment.

A textile auxiliary production wastewater treatment device embodiment II:

the embodiment provides textile auxiliary production wastewater treatment equipment, which comprises a processor and a memory, wherein the processor is used for processing instructions stored in the memory to realize the following monitoring process:

acquiring a water surface image after wastewater treatment is completed, and dividing the water surface image according to the brightness value of each pixel point in the water surface image to obtain a highlight region;

Therefore, the textile auxiliary production wastewater treatment apparatus provided in this embodiment is essentially a processor apparatus, and is implemented by an internal data processing process, and since the data processing process has been described in detail in the above-mentioned first embodiment of the textile auxiliary production wastewater treatment apparatus, the details are not repeated here.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the scope of the embodiments of the present application, and are intended to be included within the scope of the present application.

Claims

1. A textile auxiliary waste water treatment apparatus comprising a processor and a memory, the processor being adapted to process instructions stored in the memory to effect a monitoring process of:

screening each highlight region according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value to determine a non-water spray region, wherein the method comprises the following steps:

for any highlight region, obtaining the water spray characteristic degree of the highlight region according to the gradient distribution characteristic value, the gray level concentration degree and the texture direction distribution value of the highlight region; the gradient distribution characteristic value and the texture direction distribution value are in positive correlation with the water spray characteristic degree, the gray concentration degree and the water spray characteristic degree are in negative correlation, and the water spray characteristic degree is a normalized numerical value; marking a highlight region corresponding to the water spray characteristic degree smaller than or equal to a preset characteristic threshold value as a non-water spray region;

And monitoring the textile auxiliary production wastewater treatment result according to the non-water-bloom area.

2. The textile auxiliary production wastewater treatment device according to claim 1, wherein the obtaining the texture direction distribution value of each highlight region according to the texture direction distribution in each highlight region and the texture distribution condition of the pixel points under different gray levels specifically comprises:

3. The textile auxiliary production wastewater treatment device according to claim 2, wherein the region growing is performed based on the difference condition of gray texture sequences among the subregions to be analyzed, and the characteristic subregions are obtained, specifically comprising:

4. The textile auxiliary production wastewater treatment device according to claim 2, wherein the partitioning treatment is performed on the highlight region to obtain a sub-region to be analyzed, and the method specifically comprises the following steps:

5. The textile auxiliary production wastewater treatment device according to claim 1, wherein the gradient distribution characteristic value of each highlight region is obtained according to the gradient distribution condition of the edge of each highlight region and the gradient distribution condition of the pixel points in the region, and specifically comprises the following steps:

6. The textile auxiliary production wastewater treatment apparatus according to claim 5, wherein the obtaining the gray level concentration degree of each highlight region according to the concentration degree of the gray level distribution of the pixel points on the edge of each highlight region specifically comprises:

7. A textile auxiliary waste water treatment apparatus according to claim 2, wherein the gray level determination using the gradient distribution characteristic value of each highlight region comprises:

8. The textile auxiliary production wastewater treatment device according to claim 1, wherein the monitoring of the textile auxiliary production wastewater treatment result according to the non-splash area specifically comprises:

9. The textile auxiliary production wastewater treatment equipment comprises a textile auxiliary production wastewater treatment device, and is characterized by further comprising a wastewater treatment result monitoring device, wherein the wastewater treatment result monitoring device comprises an image collector and a data controller, the image collector is in signal connection with the data controller, and the image collector is used for collecting water surface images after wastewater treatment is completed;