Information processing method, device and equipment based on image analysis

Abstract

The present invention provides an information processing method, device and equipment based on image analysis, wherein the method comprises: obtaining real-time working condition image data of an aeration tank; analyzing and processing the real-time working condition image data to determine bubble data on the water surface in the real-time working condition image data; determining adjustment parameters of the aeration tank according to the bubble data; and adjusting the parameters of the aeration tank according to the adjustment parameters of the aeration tank. The scheme of the present invention can associate the aeration image with the process parameters of the aeration tank, thereby realizing the judgment of the aeration state and the inference of the process parameters through the video, judging the process parameters through the image and guiding the optimization of the adjustment of the aeration parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of computer information processing, and in particular to an information processing method, device and equipment based on image analysis. Background Art

[0002] In the prior art, the status of the aeration tank can only identify the start and end stages of aeration, and simultaneously upload video clips of the start and end stages of sewage tank aeration to the server for download and viewing by regulatory authorities, but it is unable to optimize the adjustment of aeration parameters. Summary of the invention

[0003] The present invention provides an information processing method, device and equipment based on image analysis, which solves the problem in the prior art that the adjustment of aeration parameters cannot be optimized.

[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0005] An information processing method based on image analysis, comprising:

[0006] Obtain real-time image data of the aeration tank;

[0007] Analyzing and processing the real-time working condition image data to determine bubble data on the water surface in the real-time working condition image data;

[0008] Determining adjustment parameters of the aeration tank according to the bubble data;

[0009] The parameters of the aeration tank are adjusted according to the adjustment parameters of the aeration tank.

[0010] Optionally, analyzing and processing the real-time working condition image data to determine bubble data on the water surface in the real-time working condition image data includes:

[0011] Extracting key frames from the real-time working condition image data to determine a target frame image containing bubbles;

[0012] Feature parameters are extracted from the target frame image to obtain bubble data on the water surface in the real-time working condition image data.

[0013] Optionally, extracting characteristic parameters of the target frame image to obtain bubble data on the water surface in the real-time working condition image data includes:

[0014] Feature parameters of at least one of the total foam area, the foam area variance, and the foam pixel position are extracted from the target frame image to obtain bubble data on the water surface in the real-time working condition image data, wherein the bubble data includes at least one of the following: bubble generation frequency, bubble size, bubble quantity, and bubble distribution.

[0015] Optionally, determining adjustment parameters of the aeration tank according to the bubble data includes:

[0016] The bubble data is input into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank. The aeration model is a relationship model between the bubble data and the aeration air volume.

[0017] Optionally, the aeration model is trained by the following process:

[0018] Acquire video image data of the aeration tank within a preset time period;

[0019] Analyzing and processing the video image data to obtain a plurality of key frame images containing bubbles in the video image data;

[0020] Analyzing and processing the multiple key frame images to determine the bubble data and sludge concentration in the multiple key frame images;

[0021] Establishing a correlation relationship between the bubble data, sludge concentration, and the blast air volume when collecting the video image data;

[0022] The aeration model is determined according to the association relationship.

[0023] Optionally, the bubble data is input into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank, including:

[0024] According to Y = f (q, w, g), the predicted adjustment parameters of the aeration tank are obtained;

[0025] Among them, Y is the predicted adjustment parameter, q is the bubble data, w is the sludge concentration, and g is the blast air volume.

[0026] Optionally, adjusting the parameters of the aeration tank according to the adjustment parameters of the aeration tank includes:

[0027] Obtain the difference between the predicted dissolved oxygen value and the dissolved oxygen set value to obtain the operating status of the aeration tank;

[0028] According to the operating status, if the oxygen is greater than a preset value, the blast air volume is reduced to a first target value. Otherwise, the blast air volume is increased to a second target value, and the second target value is greater than the first target value.

[0029] The present invention also provides an information processing device based on image analysis, comprising:

[0030] An acquisition module is used to obtain real-time working condition image data of the aeration tank;

[0031] The processing module is used to analyze and process the real-time working condition image data to determine the bubble data on the water surface in the real-time working condition image data; determine the adjustment parameters of the aeration tank according to the bubble data; and adjust the parameters of the aeration tank according to the adjustment parameters of the aeration tank.

[0032] The present invention also provides a computing device, comprising: a processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed.

[0033] The present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above.

[0034] The above solution of the present invention includes at least the following beneficial effects:

[0035] The above scheme of the present invention obtains real-time working condition image data of the aeration tank; analyzes and processes the real-time working condition image data to determine the bubble data on the water surface in the real-time working condition image data; determines the adjustment parameters of the aeration tank according to the bubble data; and adjusts the parameters of the aeration tank according to the adjustment parameters of the aeration tank. By associating the aeration image with the process parameters of the aeration tank, it is possible to judge the aeration state and infer the process parameters through the video, judge the process parameters through the image and guide the optimization of the adjustment of the aeration parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flowchart of an information processing method based on image analysis according to an embodiment of the present invention;

[0037] Figure 2 It is a module block diagram of an information processing device based on image analysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0039] like Figure 1 As shown, an embodiment of the present invention provides an information processing method based on image analysis, comprising:

[0040] Step 11, obtaining real-time working condition image data of the aeration tank;

[0041] Step 12, analyzing and processing the real-time working condition image data to determine the bubble data on the water surface in the real-time working condition image data;

[0042] Step 13, determining the adjustment parameters of the aeration tank according to the bubble data;

[0043] Step 14: adjusting the parameters of the aeration tank according to the adjustment parameters of the aeration tank.

[0044] In this embodiment of the present invention, the obtained real-time working condition image data of the aeration tank is analyzed and processed to determine the bubble data on the water surface in the real-time working condition image data, and then the adjustment parameters of the aeration tank are determined according to the bubble data, and the parameters of the aeration tank are adjusted according to the adjustment parameters of the aeration tank. In this way, by associating the working condition image of the aeration tank with the process parameters of the aeration tank, it is possible to judge the aeration state and infer the process parameters through the video, judge the process parameters through the image and guide the optimization of the adjustment of the aeration parameters.

[0045] In an optional embodiment of the present invention, the step 12 may include:

[0046] Step 121, extracting key frames from the real-time working condition image data to determine a target frame image containing bubbles;

[0047] Step 122: extract characteristic parameters of the target frame image to obtain bubble data on the water surface in the real-time working condition image data.

[0048] In this embodiment, video data of the aeration tank is collected, images are obtained from the video by extracting key frames in the video, and a target frame image is obtained. Feature parameters of the target frame image are extracted to obtain bubble data on the water surface in the real-time working image data, so as to determine whether there are bubbles generated by aeration on the water surface in the real-time working image data. In this way, it can be determined whether aeration is being performed in the aeration tank at the current moment.

[0049] It should be noted that image processing technology or machine vision algorithm can be used to identify the foam in the target frame image. For example, OpenCV (cross-platform computer vision and machine learning software library) machine vision algorithm can be used to identify white foam on the water surface, thereby determining that the aeration tank is being aerated.

[0050] In another optional embodiment of the present invention, the step 122 may include:

[0051] Step 1221, extract feature parameters of at least one of the total foam area, foam area variance, and foam pixel position of the target frame image to obtain bubble data on the water surface in the real-time working condition image data, and the bubble data includes at least one of the following: bubble generation frequency, bubble size, bubble number, and bubble distribution.

[0052] In this embodiment, when aeration is being carried out in the aeration tank, a visual algorithm can be used to extract feature parameters of at least one of the total foam area, the foam area variance, and the foam pixel position in the target image, so as to identify the bubbles on the surface of the aeration tank and obtain the bubble data on the water surface in the real-time working condition image data, wherein the bubble data includes at least one of the bubble generation frequency, the bubble size, the bubble number, and the bubble distribution. In this way, relevant parameter information of the foam in the aeration tank can be accurately obtained, thereby realizing the association between the working condition image of the aeration tank and the process parameters of the aeration tank.

[0053] Specifically, the total foam area is the total area occupied by the foam in the target frame image; the foam area variance is used to characterize the foam spatial distribution. The smaller the foam area variance is, the more uniform the foam spatial distribution is, and the better the aeration state of the aeration tank is; the foam pixel position represents the position of the foam in the target frame image, for example: p(x, y), p is the pixel position of the foam in the target frame image, x is the horizontal coordinate of the foam pixel in the target frame image, and y is the vertical coordinate of the foam pixel in the target frame image;

[0054] The bubble generation frequency is the number of bubbles generated per second; the bubble size is the average size of all bubbles; the bubble number is the total number of bubbles in a target frame image; and the bubble distribution is the average spacing between all bubbles in a target frame image.

[0055] Bubble generation frequency: The rate of bubble generation is obtained based on the change of the total foam area over time.

[0056] Bubble size: Use image recognition algorithm to calculate the average bubble size.

[0057] Number of bubbles: The number of bubbles is obtained based on the total foam area and bubble size data.

[0058] Bubble distribution: refers to the distribution of bubbles in the image, which represents the actual distribution of bubbles in the aeration tank.

[0059] It should be noted that the bubble data can be classified according to its distribution, for example, the bubble data can be graded into three categories: "excellent", "medium" and "poor". Combining the classification results and manual experience, the target frame images of the training data set are annotated.

[0060] The image classification model is trained with the labeled training data set. The actual acquired images are processed and analyzed by the image classification model to obtain the current bubble and aeration status: "Excellent" means that the current aeration condition is good and the process parameters are appropriate; "Medium" means that the process parameters need to be slightly adjusted to increase or decrease the aeration volume; "Poor" means that there are certain problems in the operation of the biochemical pool, and further judgment is needed based on parameters such as sludge concentration to make larger adjustments.

[0061] According to the three types of bubbles, three aeration models are established, namely the first characteristic model, the second characteristic model and the third characteristic model, to more accurately reflect the reaction process of the biochemical pool. When the real-time image is determined to be at a certain level by the image classification model, the aeration model of the corresponding level is used for calculation.

[0062] In another optional embodiment of the present invention, the step 13 may include:

[0063] Step 131, inputting the bubble data into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank, wherein the aeration model is a relationship model between the bubble data and the aeration air volume.

[0064] In this embodiment, the bubble data is combined with the external aeration air volume (which can be updated every hour) and the real-time sludge concentration value to carry out training and learning to realize aeration bubble recognition, and establish the relationship between bubble generation frequency, bubble size, bubble number, bubble distribution and aeration air volume, so that the normal aeration recognition model has the ability to predict and estimate the global aeration volume value.

[0065] In another optional embodiment of the present invention, in step 131, the aeration model may be trained by the following process:

[0066] Step 131-1, obtaining video image data of the aeration tank within a preset time period;

[0067] Step 131-2, analyzing and processing the video image data to obtain a plurality of key frame images containing bubbles in the video image data;

[0068] Step 131-3, analyzing and processing the multiple key frame images to determine the bubble data and sludge concentration in the multiple key frame images;

[0069] Step 131-4, establishing a correlation relationship between the bubble data, sludge concentration, and the blast air volume when collecting the video image data;

[0070] Step 131 - 5 , determining the aeration model according to the association relationship.

[0071] In a specific implementation, this embodiment may predict the adjustment parameters using the first feature model, the second feature model, or the third feature model according to different model input data, which may specifically include:

[0072] The texture and contour features of the individual foam state are identified through the foam feature end, and the identification data is directly transmitted to the first feature model training set for training, and training learning is carried out to establish the corresponding identification of image features and foam features, with the proportion value of the foam area and the global area;

[0073] Specifically, the first characteristic model may be: y=b1q+b2w+b3g, wherein y is the predicted adjustment parameter, q is the bubble data, w is the sludge concentration, g is the blast volume, and b1, b2 and b3 are all coefficients.

[0074] The floating mud feature terminal identifies the texture and contour features of the individual floating mud state, and transmits the identified features to the second feature model training set, externally accesses the sludge concentration value (which can be updated per hour per group of data), conducts training and learning, realizes the floating mud state recognition, and establishes the hidden danger relationship between the floating mud and the sludge concentration value, so that the floating mud recognition model has the characteristic memory of the floating mud state in the image, and at the same time, through contour recognition, has the proportion value of the floating mud area and the global area;

[0075] Specifically, the second characteristic model may be: y=α(w*w x ), where y is the predicted adjustment parameter, α is the coefficient, w is the sludge concentration, and w x is the characteristic value of floating mud.

[0076] The water surface feature end is based on the image features of the water surface, and converts the RGB (RedGreenBlue) color space into the HIS (HueIntensitySaturation) color space, that is, the three basic feature quantities of hue, saturation and brightness. The obtained feature values ​​are sent to the third feature model training set for training, and training learning is carried out to establish the corresponding relationship between the water body image features and the dissolved oxygen parameters, so that the dissolved oxygen concentration of the current aeration tank can be calculated through the parameters of the water surface image.

[0077] Specifically, the third characteristic model may be: Wherein, y is the predicted adjustment parameter, R is the red value, G is the green value, and B is the blue value.

[0078] It should be noted that the first characteristic model, the second characteristic model and the third characteristic model can be used individually, or in pairs, or all three can be used simultaneously. The specific usage can be determined according to the actual needs on site.

[0079] In another optional embodiment of the present invention, the step 131 may include:

[0080] Step 1311, obtaining the predicted adjustment parameters of the aeration tank according to Y=f(q,w,g);

[0081] Among them, Y is the predicted adjustment parameter, q is the bubble data, w is the sludge concentration, and g is the blast air volume.

[0082] In this embodiment, during model training, the actual aeration volume and sludge concentration are incorporated into the data model to greatly improve the accuracy of the model. In this embodiment, the bubble data is combined with the external blast air volume (which can be updated per hour per group of data) and the real-time sludge concentration value to carry out training and learning, realize aeration bubble recognition, and establish the relationship between bubble generation frequency, bubble size, bubble number, bubble distribution and blast air volume and sludge concentration. Through the aeration tank video, combined with the influencing factors such as blast volume and sludge concentration, the aeration model is established to determine whether the aeration and other operating conditions of each area in the aeration tank are normal, and whether there are abnormal conditions such as sludge expansion and floating, foam accumulation, and garbage floating. There is no need for personnel to enter the covered pool for inspection. The aeration tank image can intuitively reflect the situation in the aeration tank. According to parameters such as bubble size, bubble distribution, and bubble number, the aeration model can continuously analyze and give the aeration value of each area. If the aeration value exceeds the normal range, different levels of alarms will be issued according to the numerical value. If sludge floating occurs in a certain area and the bubble shape cannot be detected, the type of abnormal situation in the area is judged according to the page texture and contour features, and classified alarms are issued. The precise aeration system adjusts the aeration volume according to the algorithm analysis results calculated from process data such as video, sludge concentration, and aeration volume, automatically controls the PLC (logic control unit) system, and issues tasks to the blower and electric valve to adjust the aeration volume in the area to the preset value to achieve precise aeration. The entire system is a measure to serve energy conservation and emission reduction in the sewage treatment system, and can adjust the activated sludge through aeration control to achieve its optimal growth state.

[0083] It should be noted that the trained aeration model can be used in the actual production environment after passing the accuracy test of the test machine. Through the processing of the aeration model, the calculated values ​​of sludge concentration and dissolved oxygen can be obtained based on the target frame image in the video of the aeration tank.

[0084] In another optional embodiment of the present invention, the step 14 may include:

[0085] Step 141, obtaining the difference between the predicted dissolved oxygen value and the dissolved oxygen set value, and obtaining the operating status of the aeration tank;

[0086] Step 142, according to the operating status indicating that the oxygen is greater than a preset value, the blast air volume is reduced to a first target value, otherwise, the blast air volume is increased to a second target value, and the second target value is greater than the first target value.

[0087] In this embodiment, the operating state of the aeration tank is obtained according to the difference between the predicted dissolved oxygen value and the dissolved oxygen set value. If the operating state indicates that the oxygen is equal to a preset value, it means that the oxygen in the aeration tank is moderate. If the operating state indicates that the oxygen is greater than a preset value, it means that the oxygen in the aeration tank is excessive, and it is necessary to appropriately reduce the blower air volume to the first target value. If the operating state indicates that the oxygen is less than a preset value, it means that the oxygen in the aeration tank is insufficient, and it is necessary to increase the blower air volume to the second target value, wherein the second target value is greater than the first target value. In this way, the process parameters of the aeration tank can be fed back and optimized, and the function of obtaining parameters such as dissolved oxygen from the aeration video can be realized.

[0088] In the above embodiment of the present invention, by collecting video data of the aeration tank, obtaining image information by extracting key frames in the video, analyzing bubbles in the aeration tank using a visual algorithm, and associating with process monitoring data, the function of judging the operation of the aeration tank through the aeration video and image and obtaining the aeration process state parameters is realized. The association between the aeration state and the aeration process parameters can be established through the aeration tank video image and the algorithm, so that the aeration state can be analyzed and predicted through the aeration tank video, and the aeration process parameters can be feedback optimized to improve the aeration efficiency, reduce energy consumption, and improve the pollutant removal rate.

[0089] In addition, given the video image of the aeration tank, the artificial intelligence algorithm is used to associate the aeration image with the process parameters, so that the process parameters can be judged and the adjustment of the aeration parameters can be optimized without relying on various sensors.

[0090] like Figure 2 As shown, an embodiment of the present invention further provides an information processing device 20 based on image analysis, the device 20 comprising:

[0091] An acquisition module 21 is used to acquire real-time working condition image data of the aeration tank;

[0092] The processing module 22 is used to analyze and process the real-time working condition image data, determine the bubble data on the water surface in the real-time working condition image data; determine the adjustment parameters of the aeration tank according to the bubble data; and adjust the parameters of the aeration tank according to the adjustment parameters of the aeration tank.

[0093] Optionally, analyzing and processing the real-time working condition image data to determine bubble data on the water surface in the real-time working condition image data includes:

[0094] Extract key frames from the real-time working condition data image data to determine a target frame image containing bubbles;

[0095] Feature parameters are extracted from the target frame image to obtain bubble data on the water surface in the real-time working condition image data.

[0096] Optionally, extracting characteristic parameters of the target frame image to obtain bubble data on the water surface in the real-time working condition image data includes:

[0097] Feature parameters of at least one of the total foam area, the foam area variance, and the foam pixel position are extracted from the target frame image to obtain bubble data on the water surface in the real-time working condition image data, wherein the bubble data includes at least one of the following: bubble generation frequency, bubble size, bubble quantity, and bubble distribution.

[0098] Optionally, determining adjustment parameters of the aeration tank according to the bubble data includes:

[0099] The bubble data is input into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank. The aeration model is a relationship model between the bubble data and the aeration air volume.

[0100] Optionally, the aeration model is trained by the following process:

[0101] Acquire video image data of the aeration tank within a preset time period;

[0102] Analyzing and processing the video image data to obtain a plurality of key frame images containing bubbles in the video image data;

[0103] Analyzing and processing the multiple key frame images to determine the bubble data and sludge concentration in the multiple key frame images;

[0104] Establishing a correlation relationship between the bubble data, sludge concentration, and the blast air volume when collecting the video image data;

[0105] The aeration model is determined according to the association relationship.

[0106] Optionally, the bubble data is input into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank, including:

[0107] According to Y = f (q, w, g), the predicted adjustment parameters of the aeration tank are obtained;

[0108] Among them, Y is the predicted adjustment parameter, q is the bubble data, w is the sludge concentration, and g is the blast air volume.

[0109] Optionally, adjusting the parameters of the aeration tank according to the adjustment parameters of the aeration tank includes:

[0110] Obtain the difference between the predicted dissolved oxygen value and the dissolved oxygen set value to obtain the operating status of the aeration tank;

[0111] According to the operating status, if the oxygen is greater than a preset value, the blast air volume is reduced to a first target value. Otherwise, the blast air volume is increased to a second target value, and the second target value is greater than the first target value.

[0112] It should be noted that the device is a device corresponding to the above method, and all implementation methods in the above method embodiments are applicable to the embodiments of the device and can achieve the same technical effects.

[0113] The embodiment of the present invention further provides a processing device, comprising: a processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0114] The embodiment of the present invention further provides a computer-readable storage medium, including instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0115] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0116] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0118] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0119] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0120] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0121] In addition, it should be noted that in the apparatus and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. Moreover, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it is understood that all or any steps or components of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in hardware, firmware, software or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0122] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or device. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

[0123] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An information processing method based on image analysis, characterized in that: include: Obtain real-time image data of the aeration tank; Analyzing and processing the real-time working condition image data to determine bubble data on the water surface in the real-time working condition image data; Determining adjustment parameters of the aeration tank according to the bubble data; Adjusting the parameters of the aeration tank according to the adjustment parameters of the aeration tank; Wherein, determining the adjustment parameters of the aeration tank according to the bubble data includes: Inputting the bubble data into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank, wherein the aeration model is a relationship model between the bubble data and the aeration volume; The aeration model is trained through the following process: Acquire video image data of the aeration tank within a preset time period; Analyzing and processing the video image data to obtain a plurality of key frame images containing bubbles in the video image data; Analyzing and processing the multiple key frame images to determine the bubble data and sludge concentration in the multiple key frame images; Establishing a correlation relationship between the bubble data, sludge concentration, and the blast air volume when collecting the video image data; Determine the aeration model according to the association relationship; According to different model input data, the adjustment parameters are predicted by the first characteristic model, the second characteristic model or the third characteristic model respectively; The texture and contour features in the single foam state are identified through the foam feature end, and the identification data is input into the first feature model; the first feature model is y=b1q+b2w+b3g, wherein y is the predicted adjustment parameter, q is the bubble data, w is the sludge concentration, g is the blast volume, and b1, b2 and b3 are all coefficients; The floating mud feature end identifies the texture and contour features in the single floating mud state and inputs the identified features into the second feature model; the second feature model is y=α(w*w x ), where y is the predicted adjustment parameter, α is the coefficient, w is the sludge concentration, and w x is the characteristic value of floating mud; The water surface feature end converts the RGB red, green and blue color space into the HIS hue, brightness and saturation color space based on the image features of the water surface, and transmits the obtained feature values ​​to the third feature model; the third feature model is Wherein, y is the predicted adjustment parameter, R is the red value, G is the green value, and B is the blue value.

2. The information processing method based on image analysis according to claim 1, characterized in that: Analyzing and processing the real-time working condition image data to determine bubble data on the water surface in the real-time working condition image data includes: Extracting key frames from the real-time working condition image data to determine a target frame image containing bubbles; Feature parameters are extracted from the target frame image to obtain bubble data on the water surface in the real-time working condition image data.

3. The information processing method based on image analysis according to claim 2, characterized in that: Extracting characteristic parameters of the target frame image to obtain bubble data on the water surface in the real-time working condition image data includes: Feature parameters are extracted for at least one of the total foam area, the foam area variance, and the foam pixel position of the target frame image to obtain bubble data on the water surface in the real-time working condition image data, wherein the bubble data includes at least one of the following: bubble generation frequency, bubble size, bubble quantity, and bubble distribution.

4. The information processing method based on image analysis according to claim 1, characterized in that: According to the adjustment parameters of the aeration tank, the parameters of the aeration tank are adjusted, including: Obtain the difference between the predicted dissolved oxygen value and the dissolved oxygen set value to obtain the operating status of the aeration tank; According to the operating status, if the oxygen is greater than a preset value, the blast air volume is reduced to a first target value. Otherwise, the blast air volume is increased to a second target value, and the second target value is greater than the first target value.

5. An information processing device based on image analysis, characterized in that: include: An acquisition module is used to obtain real-time working condition image data of the aeration tank; A processing module, used for analyzing and processing the real-time working condition image data, and determining the bubble data on the water surface in the real-time working condition image data; Determining adjustment parameters of the aeration tank according to the bubble data; Adjusting the parameters of the aeration tank according to the adjustment parameters of the aeration tank; Wherein, determining the adjustment parameters of the aeration tank according to the bubble data includes: Inputting the bubble data into a preset aeration model for processing to obtain predicted adjustment parameters of the aeration tank, wherein the aeration model is a relationship model between the bubble data and the aeration volume; The aeration model is trained through the following process: Acquire video image data of the aeration tank within a preset time period; Analyzing and processing the video image data to obtain a plurality of key frame images containing bubbles in the video image data; Analyzing and processing the multiple key frame images to determine the bubble data and sludge concentration in the multiple key frame images; Establishing a correlation relationship between the bubble data, sludge concentration, and the blast air volume when collecting the video image data; Determine the aeration model according to the association relationship; According to different model input data, the adjustment parameters are predicted by the first characteristic model, the second characteristic model or the third characteristic model respectively; The texture and contour features in the single foam state are identified through the foam feature end, and the identification data is input into the first feature model; the first feature model is y=b1q+b2w+b3g, wherein y is the predicted adjustment parameter, q is the bubble data, w is the sludge concentration, g is the blast volume, and b1, b2 and b3 are all coefficients; The floating mud feature end identifies the texture and contour features in the single floating mud state and inputs the identified features into the second feature model; the second feature model is y=α(w*w x ), where y is the predicted adjustment parameter, α is the coefficient, w is the sludge concentration, and w x is the characteristic value of floating mud; The water surface feature end converts the RGB red, green and blue color space into the HIS hue, brightness and saturation color space based on the image features of the water surface, and transmits the obtained feature values ​​to the third feature model; the third feature model is Wherein, y is the predicted adjustment parameter, R is the red value, G is the green value, and B is the blue value.

6. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 4 is performed.

7. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 4.

Images