CN111489351B - Video analysis method for measuring surface segregation degree based on stack image - Google Patents
Video analysis method for measuring surface segregation degree based on stack image Download PDFInfo
- Publication number
- CN111489351B CN111489351B CN202010312435.5A CN202010312435A CN111489351B CN 111489351 B CN111489351 B CN 111489351B CN 202010312435 A CN202010312435 A CN 202010312435A CN 111489351 B CN111489351 B CN 111489351B
- Authority
- CN
- China
- Prior art keywords
- image
- segregation
- particle
- particles
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N2017/008—Diagnosis, testing or measuring for television systems or their details for television teletext
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a video analysis method for measuring surface segregation degree based on stack images, which is based on the principles of image identification and image analysis, and is characterized in that a particle surface segregation picture sequence in an acquired video is stacked and regarded as a whole, required pixels are extracted by using color space conversion, and the surface particle concentration is converted into the pixel concentration, so that the surface segregation degree of particles is accurately evaluated. The method does not need to stop sampling or use an invasive probe, overcomes the defect that the traditional method needs to stop repeatedly to acquire pictures when used for experimental research, and the segregation state of the particles in the roller is easily influenced in the stopping process, so that the experimental process of quantifying the segregation degree of the surface layer of the particles can be realized at a higher speed, and the method is favorable for better analyzing and researching the mixing and segregation performance of a particle system and influencing factors thereof.
Description
Technical Field
The invention relates to the technical field of particle system mixing and segregation online measurement and evaluation, in particular to a video analysis method for measuring surface segregation degree based on stack images.
Background
The ball mill is widely applied to engineering machinery. The flow of particles in the rotating drum often presents a complex phenomenon, particularly during mixing and segregation, which has been the topic of interest to researchers. The mixing performance is also related to the mixing quality and production efficiency of the rotating drum. Therefore, the problems of particle movement and material mixing and segregation in the ball mill are further researched, and the industrial application of particle mixing equipment such as the ball mill is facilitated.
With the rapid development of computer technology, a numerical simulation technology related to particle motion is generated correspondingly, especially a Discrete Element Method (DEM) of a motion form in the aspect of particle dynamics is most attractive, such research can avoid interference of external conditions, can also acquire transient information of a particle system to the maximum extent, captures a dynamic flow transition process of particles, and effectively deepens people' understanding of a mechanism related to the particle system. Although the discrete element method can count the time and space information of each particle, it is limited to a simulation model and lacks effective experimental verification.
Currently, the commonly used Optical Image Processing (Optical Image Processing) method has the advantages of being relatively simple and low in cost, and is one of the most commonly used non-invasive methods in experimental research. However, the method has the defects that only image information collected in a fixed area is researched, and the collection process is easily influenced by external factors; when the roller is long and the number of particles is large, the segregation degree is quantified by the particle concentration, and the effect is not ideal; in addition, when the method is used for experimental research, repeated shutdown is needed to collect pictures, and the segregation state of the particles in the roller is easily influenced in the shutdown process. Therefore, if information such as a particle motion track, a velocity, an acceleration and the like is acquired in an experiment and the segregation degree of the surface layer of the particle is accurately quantified, it is necessary to research an image identification method and a segregation degree quantification method so that the method is better suitable for numerical simulation and experimental verification.
Therefore, the video analysis method for measuring the surface segregation degree based on the stack image is provided, and stopping sampling or using an invasive probe are not needed, so that the efficiency of quantifying the surface segregation degree is improved.
Disclosure of Invention
To solve the above existing problems. The invention provides a video analysis method for measuring surface segregation degree based on stack images, which is used for accurately evaluating the surface segregation degree of particles in a mixing process on the basis of a traditional Lacey index method and an image recognition principle under the condition of considering the axial segregation phenomenon of the particles in a ball mill and is used for mutual verification with a simulation model. To achieve this object:
the invention provides a video analysis method for measuring surface segregation degree based on stack images, which comprises the following specific steps:
1) Building a ball mill experiment platform and a simulation model;
firstly, building a ball mill experimental platform, which comprises a ball mill rotary drum, granular substances, a torque sensor, a stepping motor and a high-speed camera, wherein the rotary drum and the granular substances are transparent in order to observe phenomena and collect videos conveniently, and particles with different particle sizes are added into the rotary drum and the rotary drum is rotated until the segregation state of the particles is stable;
then constructing a ball mill simulation model, generating a certain number of particles with two particle sizes in the simulation model according to the initial configuration with the same volume, wherein the simulation model parameters are consistent with the experiment parameters, then rotating the ball mill, wherein the rotation time of the ball mill is consistent with the experiment time, and then entering the next step;
2) Video acquisition and image digitization;
the image digitization mainly comprises three parts: firstly, shooting a video of a particle movement process by adopting a high-speed camera in the rotating process of a ball mill rotary drum; then, manually marking interested regions ROI for all videos; finally, cutting out an ROI area from each video frame to form an image belt;
3) Stacking the images and carrying out image processing;
next, in order to improve the quality of the stacked image, image processing is required, image bands are stacked according to a time sequence to form a stacked image, then a color image is converted from an RGB color space to HSV (Hue, saturation, value and color space), and then the image quality is improved through binary reconstruction of the image, so that the effect of noise reduction is achieved;
4) Processing data and quantifying the particle segregation degree;
carrying out grid division on the stack picture subjected to binarization reconstruction along the axial direction and the radial direction, counting the number of different types of particle pixels in each grid, replacing the particle concentration in a Lacey index method by using the pixel concentration in a segregation pattern picture, and finally calculating a segregation index according to a pixel concentration method, wherein particles with larger particle sizes are black pixels, so that the segregation degree is quantized, wherein the calculation formula is as follows:
wherein, the first and the second end of the pipe are connected with each other,the variance of the black pixel concentration when the binary mixture is completely segregated;the variance of the black pixel concentration when the binary mixture is completely mixed; sigma 2 The variance of the concentration of the black pixels in the stack image collected currently;
5) Analyzing data and verifying a model;
and quantifying the segregation degree of the surface layer of the particles by taking the segregation index and the time as indexes according to the information of the segregation index in the stack image, namely reflecting the particle segregation characteristics of the surface layer of the roller of the ball mill, drawing a picture, and comparing and analyzing the results of the experimental model and the simulation model so as to research the collision characteristics and the aggregation tendency of the particles in the mixing process.
As a further improvement of the invention, the high speed camera shoots at 25 frames/second during the video capture in step 2 and video capture is started before the drum starts to rotate until the particle segregation characteristics tend to stabilize in the drum and the ball mill stops rotating.
As a further improvement of the invention, the number of lines in the stack image in step 3 corresponds to the image index number in the time-sequential video frame, and the height of the stack image should be the length of one rotation of the drum, i.e. the outer circumference of the drum.
As a further improvement of the invention, the HSV color space in step 3 converts the three RGB color components into hue, saturation and brightness values, so that pixels of different types of particles can be conveniently extracted, the pixels are converted into the RGB color space again after being extracted, and the image is binarized and reconstructed to improve the image quality, thereby achieving the effect of noise reduction.
As a further improvement of the invention, the particle concentration is replaced by the pixel concentration in step 4, and the segregation index is calculated based on the conventional Lacey method, which considers a binary mixture of two different types of pixels a and b, and only a single pixel type, such as particles with larger particle size, is considered in the calculation process, and for any given total number of samples n s Weight k of sample i i The calculation formula is as follows:
wherein n is i The number of pixels of the type in the sample i is taken; n is t Is the sum of the number of pixels of the type;
variance of black pixel concentration at full segregation using pixel fractionVariance of black pixel density at full mixingVariance σ of black pixel density in current stack image 2 The calculation is shown as follows:
wherein n is s Is a sample assemblyCounting; a is a i The proportion of the type pixel in the sample i is shown; a is the pixel proportion of the type of pixels in the roller; p is the proportion of the type of pixels; (1-P) is the proportion of another pixel; n is the average pixel number within each sample, and thus, according to the above formula, the segregation index SI formula particle is expressed as:
as a further improvement of the invention, the simulation model parameters in the model comprise density, shear modulus and Poisson ratio, and the model parameters comprise roller length, diameter, rotating speed and particle size.
The invention discloses a video analysis method for measuring surface segregation degree based on stack images, which is characterized in that under the assumption that segregation phenomenon exists in particles in a ball mill during the movement process, videos and pictures can be acquired on line for analysis through the video analysis method for measuring the surface segregation degree based on the stack images, the defect of stopping sampling or using an invasive probe in the traditional method is eliminated, the surface segregation degree of the particles in the experimental mixing process is accurately evaluated and compared with a simulation model for analysis, and the experimental verification efficiency is greatly improved.
Drawings
FIG. 1 is a work flow diagram;
FIG. 2 is the steady segregation distribution of different particle size particles in the bowl of the ball mill;
FIG. 3 is a schematic view of video cropping;
FIG. 4 is a schematic diagram of stacked images;
FIG. 5 is a schematic diagram of mesh partitioning and binarization reconstruction;
FIG. 6 is a comparison of surface segregation indices in experimental and simulation models;
FIG. 7 is a flow chart of experimental and simulated measurement of the degree of segregation;
fig. 8 is an algorithm flow chart.
Detailed Description
The invention is described in further detail below with reference to the following figures and embodiments:
the invention provides a video analysis method for measuring surface segregation degree based on stack images, which is used for accurately evaluating the surface segregation degree of particles in a mixing process and mutually verifying the surface segregation degree with a simulation model on the basis of a traditional Lacey index method and an image identification principle under the condition of considering the axial segregation phenomenon of the particles in a ball mill.
As a specific embodiment of the present invention, the present invention provides a video analysis method for measuring a degree of surface segregation based on a stack image, a flowchart is shown in fig. 1, and specific steps are as follows;
the method comprises the following steps: experimental platform and simulation model for building ball mill
Firstly, a ball mill experiment platform is built and comprises a ball mill rotary drum, granular substances, a torque sensor, a stepping motor, a high-speed camera and the like, the rotary drum and the granular substances are transparent in order to observe phenomena and collect videos conveniently, granules with different particle sizes are added into the rotary drum, the granules are initially configured into a complete segregation state, the rotary drum is rotated until the granule segregation state is stable, and the videos need to be shot in the process.
And then constructing a ball mill simulation model, generating a certain number of particles with two particle sizes in the simulation model according to the initial configuration of the same volume, and generating parameters (density, shear modulus and Poisson ratio) of the simulation model, the length, the diameter and the rotating speed of the roller, wherein the particle size parameters are consistent with the experimental parameters. Then, the ball mill was rotated for a period of time consistent with the experimental time, and the particle distribution of the large and small particles was as shown in FIG. 2, it was found that the large particles appeared to be aggregated at both ends and in the middle of the distribution, and the small particles were aggregated at other positions. The next step can be entered at this point.
Step two: video acquisition and image digitization
Firstly, in the process of the rotation of the rotary drum of the ball mill in the previous step, a high-speed camera is adopted to shoot a video of the particle movement process, and a video acquisition area is shown in fig. 3; then, the region of interest (ROI) was manually marked for all simulated and experimentally acquired video, and the target region picture (ROI region) was cropped out by Adobe Premiere Pro software to form image strips.
Step three: stacking images and performing image processing
Image strips are first stacked in chronological order to form a stack image, as shown in fig. 4. The number of lines in the stack image corresponds to the image index of the time sequential video frames and the height of the stack image should be the length of one revolution of the drum, i.e. the outer circumference of the drum.
Next, in order to improve the quality of the stacked image, image processing is required. As shown in fig. 5, the color image is converted from the RGB color space to the HSV (Hue, saturation) color space, and after the pixel is extracted, the color image is converted into the RGB color space again, and the reconstructed image is binarized to improve the picture quality, thereby achieving the effect of noise reduction.
Step four: data processing and quantification of degree of particle segregation
As shown in fig. 5, the stack picture obtained by the above binarization reconstruction is subjected to grid division along the axial direction and the radial direction, the number of different types of particle pixels in each grid is counted, the pixel concentration in the segregation pattern photo is used instead of the particle concentration in the Lacey index method, two types of particles with different particle sizes, namely a and b, are considered here, and only a single type of particle (for example, a particle with a larger particle size) is considered in the calculation process. For any given total number of samples n s Weight k of sample i i The calculation formula is as follows:
wherein n is i The type pixel number in the sample i is taken; n is a radical of an alkyl radical t Is the sum of the pixel numbers of the type.
Variance of black pixel concentration at full segregation using pixel fractionVariance of black pixel density at full mixingVariance σ of black pixel density in current stack image 2 The calculation is shown as follows:
wherein n is s Is the total number of samples; a is i The proportion of the type pixel in the sample i is shown; a is the pixel proportion of the type of pixels in the roller; p is the proportion of the type of pixels; (1-P) is the proportion of another pixel; n is the average number of pixels in each sample. Therefore, according to the above calculation formula, the segregation index SI calculation formula is expressed as:
wherein the segregation index SI =1 indicates that the particles are in a completely segregated state, and the segregation index SI =0 indicates that the particles are in a completely mixed state. Thereby obtaining a particle segregation index for each time period.
Step five: data analysis and verification model
And (4) according to the information of the segregation indexes in the stack image, quantifying the segregation degree of the surface layer of the particles by taking the segregation indexes and time as indexes, namely reflecting the particle segregation characteristics of the surface layer of the roller of the ball mill. Drawing a picture as shown in fig. 6, comparing and analyzing results of the experimental model and the simulation model, and finding that in the simulation and the experiment, the experimental segregation index in the initial mixing stage is slightly lower than that in the simulation, and it is possible that the initial rotation speed of the roller has an acceleration stage and is unstable in the experimental process, which affects the mixing efficiency of the particles. However, the final segregation indexes calculated by the two models are basically consistent, which shows that the proposed video analysis method for measuring the surface segregation degree based on the stack image can accurately quantize the particle surface segregation indexes. Fig. 7 is a flow chart of experimental and simulated measurement of the degree of segregation, and the algorithm may be represented by the flow chart of fig. 8.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and any modifications or equivalent variations made in accordance with the technical spirit of the present invention may fall within the scope of the present invention as claimed.
Claims (6)
1. A video analysis method for measuring surface segregation degree based on stack image comprises the following specific steps:
1) Building a ball mill experiment platform and a simulation model;
firstly, building an experimental platform of the ball mill, wherein the experimental platform comprises a rotary drum of the ball mill, granular substances, a torque sensor, a stepping motor and a high-speed camera, the rotary drum and the granular substances are transparent in order to observe phenomena and collect videos conveniently, and the rotary drum is added with granules with different grain diameters and rotated until the segregation state of the granules is stable;
then constructing a ball mill simulation model, generating a certain number of particles with two particle sizes in the simulation model according to the initial configuration with the same volume, wherein the simulation model parameters are consistent with the experiment parameters, then rotating the ball mill, wherein the rotation time of the ball mill is consistent with the experiment time, and then entering the next step;
2) Video acquisition and image digitization;
the image digitization mainly comprises three parts: firstly, shooting a video of a particle movement process by adopting a high-speed camera in the rotating process of a rotary drum of the ball mill; then, manually marking interested regions ROI for all videos; finally, cutting out an ROI area from each video frame to form an image belt;
3) Stacking the images and performing image processing;
next, in order to improve the quality of the stacked image, image processing is required, image bands are stacked according to a time sequence to form a stacked image, then a color image is converted from an RGB color space to HSV (Hue, saturation, value and color space), and then the image quality is improved through binary reconstruction of the image, so that the effect of noise reduction is achieved;
4) Processing data and quantifying the particle segregation degree;
carrying out grid division on the stack picture subjected to binarization reconstruction along the axial direction and the radial direction, counting the number of different types of particle pixels in each grid, replacing the particle concentration in a Lacey index method by using the pixel concentration in a segregation pattern picture, and finally calculating a segregation index according to a pixel concentration method, wherein particles with larger particle sizes are black pixels, so that the segregation degree is quantized, wherein the calculation formula is as follows:
wherein, the first and the second end of the pipe are connected with each other,the variance of the black pixel concentration when the binary mixture is completely segregated;the variance of the black pixel concentration when the binary mixture is completely mixed; sigma 2 The variance of the concentration of the black pixels in the currently collected stack image is obtained;
5) Analyzing data and verifying a model;
according to the information of the segregation indexes in the stack image, the segregation indexes and the time are used as indexes to quantify the degree of segregation on the surface layer of the particles, namely, the segregation characteristics of the particles on the surface layer of the roller of the ball mill are reflected, pictures are drawn, the results of the experimental model and the simulation model are compared and analyzed, and then the collision characteristics and the aggregation tendency of the particles in the mixing process are researched.
2. The video analysis method of claim 1, wherein the video analysis method comprises: in the step 2, the high-speed camera shoots at 25 frames/second in the video acquisition process, and video capture is started before the drum starts to rotate until the particle segregation characteristics in the drum tend to be stable and the ball mill stops rotating.
3. The video analysis method of claim 1, wherein the video analysis method comprises: the number of lines in the stack image in step 3 corresponds to the image sequence number in the time-sequential video frame, and the height of the stack image should be the length of one rotation of the drum, i.e. the outer circumference of the drum.
4. The video analysis method of claim 1, wherein the video analysis method comprises: in the step 3, the HSV color space converts the three RGB color components into hue, saturation and brightness values, so that pixels of different types of particles can be conveniently extracted, the extracted pixels are converted into the RGB color space again, and the image is binarized and reconstructed to improve the quality of the image, so that the noise reduction effect is achieved.
5. The video analysis method of claim 1, wherein the video analysis method comprises: in the step 4, the particle concentration is replaced by the pixel concentration, the segregation index is calculated based on the traditional Lacey method, the binary mixture of two different types of pixels a and b is considered by the method, only a single pixel type, such as particles with larger particle size, is considered in the calculation process, and the total number n of any given sample is s Weight k of sample i i The calculation formula is as follows:
wherein n is i The number of pixels of the type in the sample i is taken; n is t Is the sum of the number of pixels of the type;
using pixel fractions, full biasVariance of black pixel density during analysisVariance of black pixel density at full blendVariance σ of black pixel density in current stack image 2 The calculation is shown as follows:
wherein n is s Is the total number of samples; a is i The proportion of the type of pixel in the sample i is calculated; a is the pixel proportion of the type of pixels in the roller; p is the proportion of the type of pixels; (1-P) is the proportion of another pixel; n is the average pixel number within each sample, and thus, according to the above formula, the segregation index SI formula particle is expressed as:
6. the method of claim 1, wherein the method comprises: the simulation model parameters in the model comprise density, shear modulus and Poisson ratio, and the model parameters comprise roller length, diameter, rotating speed and particle size.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010312435.5A CN111489351B (en) | 2020-04-20 | 2020-04-20 | Video analysis method for measuring surface segregation degree based on stack image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010312435.5A CN111489351B (en) | 2020-04-20 | 2020-04-20 | Video analysis method for measuring surface segregation degree based on stack image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111489351A CN111489351A (en) | 2020-08-04 |
CN111489351B true CN111489351B (en) | 2022-11-18 |
Family
ID=71798574
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010312435.5A Active CN111489351B (en) | 2020-04-20 | 2020-04-20 | Video analysis method for measuring surface segregation degree based on stack image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111489351B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115615990B (en) * | 2022-12-16 | 2023-04-07 | 北京科技大学 | Research method for carbon segregation of tire cord steel casting blank |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106157335A (en) * | 2016-06-15 | 2016-11-23 | 长江水利委员会长江科学院 | Bed material top layer based on digital picture grating observation and analysis method and system |
CN108074235A (en) * | 2017-12-26 | 2018-05-25 | 太原理工大学 | Carbon fiber surface defect degree method of estimation based on region growing algorithm |
-
2020
- 2020-04-20 CN CN202010312435.5A patent/CN111489351B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106157335A (en) * | 2016-06-15 | 2016-11-23 | 长江水利委员会长江科学院 | Bed material top layer based on digital picture grating observation and analysis method and system |
CN108074235A (en) * | 2017-12-26 | 2018-05-25 | 太原理工大学 | Carbon fiber surface defect degree method of estimation based on region growing algorithm |
Non-Patent Citations (1)
Title |
---|
回转装置内三组元颗粒径向混合评价方法分析;李少华等;《化工进展》;20130605(第06期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111489351A (en) | 2020-08-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA1161271A (en) | Method and apparatus for measuring mean cell volume of red blood cells | |
Friedland et al. | SIOX: Simple interactive object extraction in still images | |
CN112668594B (en) | Unsupervised image target detection method based on antagonism domain adaptation | |
CN103530638B (en) | Method for pedestrian matching under multi-cam | |
CN101059425A (en) | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis | |
CN107358141B (en) | Data identification method and device | |
CN101061477A (en) | Variance-based event clustering | |
CN111489351B (en) | Video analysis method for measuring surface segregation degree based on stack image | |
CN110738128A (en) | repeated video detection method based on deep learning | |
CN113624645A (en) | Device and method for detecting particle flow distribution | |
CN110097537A (en) | A kind of meat quantitative analysis evaluation method based on three-D grain feature | |
CN108765405A (en) | A kind of image authenticating method and system | |
CN111340019A (en) | Grain bin pest detection method based on Faster R-CNN | |
CN114596316A (en) | Road image detail capturing method based on semantic segmentation | |
CN114037693A (en) | Rock hole-fracture and impurity characteristic evaluation method based on deep learning | |
CN114078127B (en) | Object defect detection and counting method, device, equipment and storage medium | |
Fan et al. | Image blur classification and blur usefulness assessment | |
CN112183640B (en) | Detection and classification method based on irregular object | |
CN109241932A (en) | A kind of thermal infrared human motion recognition method based on movement variogram phase property | |
CN1885345A (en) | Analysis method of digital image color analysis system | |
CN117237736A (en) | Daqu quality detection method based on machine vision and deep learning | |
CN115598017B (en) | Method for identifying morphology and evaluating skeleton of coarse aggregate for pervious concrete | |
CN1508755A (en) | Sensitive video-frequency detecting method | |
CN115456957B (en) | Method for detecting change of remote sensing image by full-scale feature aggregation | |
CN116342505A (en) | Detection method and detection system for granulating degree of aerobic granular sludge |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |