Graphene oxide evaluation method and system based on image features

Abstract

The invention discloses a graphene oxide evaluation method and system based on image features, and the method comprises the steps: firstly, collecting optical microscopic images of different regions of a plurality of graphene oxide samples through a camera, and calculating the areas of graphene oxide sheets with different sizes in each microscopic image; then, carrying out statistical analysis on the area values of the graphene oxide sheets in all microscopic images, and drawing an area cumulative frequency curve corresponding to the graphene oxide sample; and finally, accurately evaluating the morphological characteristics and dispersion quality of the sample through the area cumulative frequency curve. Compared with an electron microscope and an atomic force microscope, the method does not need complex operation and is low in cost, so that the graphene oxide sample can be evaluated efficiently and economically, and the evaluation requirements of a large number of samples are met.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a graphene oxide evaluation method and system based on image features. Background Art

[0002] Graphene oxide (GO) is a single carbon atom layer with a regular hexagonal crystalline structure and oxygen-containing functional groups such as hydroxyl and carboxyl. Graphene oxide can extend to several microns to tens of microns in lateral dimensions. Its structure is close to a plane and presents a two-dimensional network. It has excellent conductivity, chemical stability and large specific surface area. When graphene oxide is added to concrete, it will affect the hydration reaction of cement, increase the output of hydrated calcium silicate in concrete, promote the formation and arrangement of cement hydration crystals, improve the structure of the interface between cement stone and aggregate, reduce porosity and increase density. While improving the microstructure, graphene oxide can also improve the durability of concrete, such as impermeability, resistance to chloride ion erosion, resistance to carbonization, and resistance to freeze-thaw.

[0003] In practical applications, the performance and application effects of graphene oxide are significantly affected by its morphological characteristics and dispersion quality. Traditional detection methods, such as electron microscopy and atomic force microscopy, can provide high-resolution images, but are usually complicated to operate and costly. They cannot meet the performance evaluation needs of a large number of graphene oxides. Therefore, there is an urgent need for an efficient and economical graphene oxide performance evaluation technology. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention proposes a graphene oxide evaluation method and system based on image features, which can efficiently and economically evaluate the performance quality of graphene oxide. The specific technical solution is as follows:

[0005] In a first aspect, a graphene oxide evaluation method based on image features is provided. In a first implementable manner of the first aspect, the method comprises:

[0006] Calculate the area of ​​each graphene oxide sheet in each microscopic image of the graphene oxide sample;

[0007] Performing statistical analysis on the area of ​​the graphene oxide sheet, and drawing an area cumulative frequency curve corresponding to the graphene oxide sample;

[0008] The performance quality of the graphene oxide sample is evaluated by the area cumulative frequency curve.

[0009] In combination with the first implementable manner of the first aspect, in a second implementable manner of the first aspect, calculating the area of ​​the graphene oxide sheet in the microscopic image of the graphene oxide sample includes:

[0010] The microscopic image of the graphene oxide sample is homogenized.

[0011] In combination with the second implementable manner of the first aspect, in a third implementable manner of the first aspect, performing homogenization processing on the microscopic image includes:

[0012] Acquire a background image of the microscopic image, and perform background correction on the microscopic image using the background image;

[0013] adjusting the histogram distribution of the corrected microscopic image, and performing equalization processing on a part of the adjusted microscopic image;

[0014] The microscopic images after equalization were normalized.

[0015] In combination with the first implementable manner of the first aspect, in a fourth implementable manner of the first aspect, calculating the area of ​​each graphene oxide sheet in the microscopic image of the graphene oxide sample includes:

[0016] Performing edge detection on the microscopic image to determine the edge profile of each graphene oxide sheet;

[0017] The area of ​​each graphene oxide sheet was calculated using a polygonal area algorithm based on the corresponding edge contours.

[0018] In combination with the first implementable manner of the first aspect, in a fifth implementable manner of the first aspect, statistical analysis is performed on the area of ​​the graphene oxide sheet, including:

[0019] Counting the frequencies corresponding to graphene oxide sheets of different areas in each of the microscopic images respectively;

[0020] Accumulating the frequencies corresponding to the graphene oxide sheets of the same area in all the microscopic images to obtain accumulated frequencies corresponding to different areas;

[0021] The area cumulative frequency curve is drawn based on the accumulated frequencies corresponding to all areas.

[0022] In combination with the fifth implementable manner of the first aspect, in a sixth implementable manner of the first aspect, a semi-logarithmic coordinate system is used to draw the area cumulative frequency curve.

[0023] In combination with the first implementable manner of the first aspect, in a seventh implementable manner of the first aspect, evaluating the performance quality of the graphene oxide sample by the area cumulative frequency curve includes:

[0024] The area cumulative frequency curve within the set area interval is intercepted to evaluate the performance quality of the graphene oxide sample.

[0025] In combination with the first implementable manner of the first aspect, in an eighth implementable manner of the first aspect, evaluating the performance quality of the graphene oxide sample by the area cumulative frequency curve includes:

[0026] The performance quality of the graphene oxide sample is evaluated based on the steepness, smoothness, data coverage and tail trend of the area cumulative frequency curve.

[0027] In a second aspect, a graphene oxide evaluation system based on image features is provided. In a first implementable manner of the second aspect, the system comprises:

[0028] A calculation module configured to calculate the area of ​​each graphene oxide sheet in each microscopic image of the graphene oxide sample;

[0029] A statistical module, configured to perform statistical analysis on the area of ​​the graphene oxide sheet and draw an area cumulative frequency curve corresponding to the graphene oxide sample;

[0030] An evaluation module is configured to evaluate the performance quality of the graphene oxide sample through the area cumulative frequency curve.

[0031] In combination with the first implementable manner of the second aspect, in a second implementable manner of the second aspect, the calculation module includes:

[0032] an edge detection unit configured to perform edge detection on the microscopic image to determine the edge profile of each graphene oxide sheet;

[0033] The area calculation unit is configured to calculate the area of ​​each graphene oxide sheet using a polygonal area algorithm according to the corresponding edge contour.

[0034] Beneficial effects: By using the graphene oxide evaluation method and system based on image features of the present invention, the morphological features of graphene oxide in each microscopic image can be determined by calculating the areas of graphene oxide sheets of different sizes in each microscopic image. By statistically analyzing the areas of graphene oxide sheets in all microscopic images, an area cumulative frequency curve reflecting the performance of graphene oxide samples can be obtained, and the morphological features and dispersion quality of the samples can be accurately evaluated based on the area cumulative frequency curve. Compared with electron microscopes and atomic force microscopes, there is no need for complex operations and the cost is low, so that graphene oxide samples can be evaluated efficiently and economically to meet the evaluation needs of a large number of samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific implementation of the present invention, the following will briefly introduce the drawings required for use in the specific implementation. In all the drawings, each element or part is not necessarily drawn according to the actual scale.

[0036] Figure 1 A flowchart of a graphene oxide evaluation method based on image features provided by an embodiment of the present invention;

[0037] Figure 2 A system block diagram of a graphene oxide evaluation system based on image features provided by an embodiment of the present invention;

[0038] Figure 3 is an optical microscopic image of a sample;

[0039] Figure 4 This is the microscopic image after background correction;

[0040] Figure 5 This is the microscopic image after homogenization;

[0041] Figure 6 is the microscopic image obtained after binarization processing;

[0042] Figure 7 is the edge detection result;

[0043] Figure 8 The area cumulative frequency curves corresponding to the A-type graphene oxide sample and the B-type graphene oxide sample, wherein the upper one is the B-type graphene oxide sample;

[0044] Fig. 9 The area cumulative frequency curves corresponding to five brands of graphene oxide. DETAILED DESCRIPTION

[0045] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0046] It should be understood that the Cumulative Frequency Curve is a statistical chart used to display the cumulative frequency distribution of a data set. The cumulative frequency refers to the sum of the frequencies of occurrence of data values ​​at a certain value or less, and the gradual accumulation process forms a curve. Usually in statistics, the cumulative frequency curve can accurately reflect the distribution characteristics of data, especially when comparing large samples, it can intuitively display the accumulation of data in different intervals.

[0047] like Figure 1 The flowchart of the graphene oxide evaluation method based on image features shown in FIG. 1 includes:

[0048] Step 1, respectively calculating the area of ​​each graphene oxide sheet in each microscopic image of the graphene oxide sample;

[0049] Step 2, performing statistical analysis on the area of ​​the graphene oxide sheet, and drawing an area cumulative frequency curve corresponding to the graphene oxide sample;

[0050] Step 3: Evaluate the performance quality of the graphene oxide sample through the area cumulative frequency curve.

[0051] Specifically, first, a plurality of optical microscopic images of different areas of graphene oxide samples can be collected by a camera, and the areas of graphene oxide sheets of different sizes in each microscopic image can be calculated respectively. Then, the area values ​​of graphene oxide sheets in all microscopic images can be statistically analyzed, and the area cumulative frequency curve corresponding to the graphene oxide sample can be drawn. Finally, the morphological characteristics and dispersion quality of the sample can be accurately evaluated by the area cumulative frequency curve. Compared with electron microscopy and atomic force microscopy, there is no need for complex operations and the cost is low, so that graphene oxide samples can be evaluated efficiently and economically to meet the evaluation needs of a large number of samples.

[0052] In this embodiment, optionally, in step 1, calculating the area of ​​the graphene oxide sheet in the microscopic image of the graphene oxide sample includes:

[0053] The microscopic image of the graphene oxide sample is homogenized.

[0054] Specifically, if Figure 3 As shown in the figure, due to the concentrated light source of the optical microscope, the brightness of graphene oxide directly photographed by a high-resolution camera shows the characteristics of "bright in the middle and dark around". For this reason, before calculating the area of ​​the graphene oxide sheet, the acquired microscopic image can be homogenized to improve the image contrast and visual effect, enhance the image details, make the image clearer and easier to analyze, and improve the efficiency and accuracy of performance evaluation.

[0055] In this embodiment, optionally, the microscopic image is subjected to homogenization processing, including:

[0056] Acquire a background image of the microscopic image, and perform background correction on the microscopic image using the background image;

[0057] adjusting the histogram distribution of the corrected microscopic image, and performing equalization processing on a part of the adjusted microscopic image;

[0058] The microscopic images after equalization were normalized.

[0059] Specifically, when collecting a microscopic image, a background image corresponding to the background in the microscopic image can be collected at the same time. The background image can be a pure background image or a homogenized reference image. By performing background correction on the microscopic image through the background image, the background effect in the microscopic image can be removed, making the details of the target image more prominent, such as Figure 4 Specifically, the background effect can be eliminated by performing differential operations on the three channels of the microscopic image and the background image. The specific calculation formula is as follows:

[0060] I diff =I target -I background ;

[0061] Among them, I diff is the corrected microscopic image, I target is the microscopic image before correction, I background For the background image.

[0062] After background correction of the microscopic image, the histogram distribution of the microscopic image can also be adjusted to make the brightness value distribution of the microscopic image more uniform, thereby improving the contrast, such as Figure 5 At the same time, equalization can be performed in different areas of the microscopic image to avoid artifacts or unnatural effects that may be caused by global equalization. Finally, the processed microscopic image is standardized to a fixed range to ensure the consistency of image display and storage.

[0063] In this embodiment, optionally, in step 1, calculating the area of ​​each graphene oxide sheet in the microscopic image of the graphene oxide sample includes:

[0064] Performing edge detection on the microscopic image to determine the edge profile of each graphene oxide sheet;

[0065] The area of ​​each graphene oxide sheet was calculated using a polygonal area algorithm based on the corresponding edge contours.

[0066] Specifically, when calculating the area of ​​the graphene oxide sheet, first, the threshold method can be used to binarize the microscopic image to highlight the outline of the object in the image, which is convenient for further analysis and processing. Specifically, the global threshold method or the adaptive threshold method can be used for binarization, such as Figure 6 As shown in Figure 2, the global threshold method refers to using a fixed threshold for the entire image. This method is suitable for images with uniform illumination, where the contrast between the object and the background is more obvious.

[0067] The adaptive threshold method binarizes images with uneven illumination. Unlike the global threshold method which uses a fixed threshold, the adaptive threshold method dynamically calculates the threshold based on the local information of the neighborhood of each pixel in the image, making it perform better when processing images with varying illumination.

[0068] Then, the existing edge detection algorithm can be used to extract the edge contour of each graphene oxide sheet in the microscopic image, such as Figure 7 As shown. Edge detection algorithms can include Canny edge detection, Sobel operator, etc. Among them, Canny Edge Detection is a multi-stage edge detection algorithm. First, Gaussian smoothing is performed, and then the edge is found by calculating the image gradient, and non-maximum suppression and double threshold processing are combined to retain strong edges. The Sobel operator detects the edge in the image by calculating the gradient of the image. It identifies the edge by the gradient change of the image in the horizontal and vertical directions.

[0069] Finally, the polygonal area algorithm can be used to calculate the contour area of ​​each graphene oxide sheet in the microscopic image. The contour area reflects the size and shape of the graphene oxide sheet. By statistically analyzing the area of ​​the graphene oxide sheet in the microscopic image, the characteristics of the graphene oxide sheet in the microscopic image can be quantified and summarized. The specific calculation formula of the polygonal area algorithm is as follows:

[0070]

[0071] Among them, x i ,y i is the coordinate of the ith vertex in the edge profile of the graphene oxide sheet, and n is the total number of vertices in the edge profile of the graphene oxide sheet. Vertex i=n+1 is connected to vertex i=1 to ensure that the edge profile of the graphene oxide sheet is closed.

[0072] Since the area is calculated using the above algorithm, the length unit is pixel. Therefore, a rough estimation method can be used to calibrate the image length unit. Specifically, a 1:1 image ratio is used and a square cover glass is used for calibration. Assuming that the side length of the cover glass is amm, the pixel of the square image is b, and the image is taken along the side of the cover glass. The number of images taken from the starting point to the end point is recorded, and the average value is taken after multiple repeated operations. It is easy to know that the actual side length of each image is That is, b pixels are equivalent to the actual length Then we know that 1 pixel is Substituting the calibrated length unit into the area formula will give the actual area size.

[0073] In this embodiment, optionally, in step 2, statistical analysis is performed on the area of ​​the graphene oxide sheet, including:

[0074] Counting the frequencies corresponding to graphene oxide sheets of different areas in each of the microscopic images respectively;

[0075] Accumulating the frequencies corresponding to the graphene oxide sheets of the same area in all the microscopic images to obtain accumulated frequencies corresponding to different areas;

[0076] The area cumulative frequency curve is drawn based on the accumulated frequencies corresponding to all areas.

[0077] Specifically, first, the area values ​​of the graphene oxide sheets in each microscopic image can be arranged in order from small to large. The frequencies of different area values ​​or area intervals in each microscopic image are counted, that is, the number of graphene oxide sheets with the same area value in the microscopic image. Combined with the total number of graphene oxide sheets in the microscopic image, the frequency corresponding to the graphene oxide sheets of each area value or area interval can be calculated.

[0078] Then, the frequencies corresponding to the graphene oxide sheets with the same area value or area range in all the collected microscopic images are accumulated to obtain the area accumulation frequency corresponding to the area value or area range. Finally, the area accumulation frequency curve corresponding to the graphene oxide sample can be drawn according to the area accumulation frequencies corresponding to all area values ​​or area ranges.

[0079] The area data for graphene oxide can vary greatly, with some particles having smaller areas and others having larger areas. If plotted using a linear coordinate system, the smaller area data points will be compressed to one side and difficult to see clearly. Semi-logarithmic coordinate systems are particularly advantageous when the data spans multiple orders of magnitude. They can improve the visualization of the data, show exponential or power law relationships, and provide clear graphical representation when dealing with nonlinear growth data. For the cumulative frequency curve of the graphene oxide area data, the use of a semi-logarithmic coordinate system allows for better visualization of the area distribution of different orders of magnitude, avoiding the data being compressed to one side or appearing unbalanced.

[0080] In this embodiment, optionally, in step 3, evaluating the performance quality of the graphene oxide sample by the area cumulative frequency curve includes:

[0081] The area cumulative frequency curve within the set area interval is intercepted to evaluate the performance quality of the graphene oxide sample.

[0082] Specifically, since graphene oxide mainly contains epoxy and hydroxyl functional groups on the basal plane and carboxyl functional groups on the edges, the research value of graphene oxide on smaller sheets is limited. For larger sheets, since graphene oxide is prone to agglomeration during synthesis and processing, inaccurate results may appear in its performance and properties. Therefore, in order to obtain more reliable research data, the area cumulative frequency curve within the set area interval can be intercepted to evaluate the performance quality of the graphene oxide sample, so as to avoid interference caused by agglomeration and more accurately evaluate its characteristics. In this embodiment, the area interval range can be set to 8μm 2 To 40μm 2 between.

[0083] In this embodiment, optionally, in step 3, evaluating the performance quality of the graphene oxide sample by the area cumulative frequency curve includes:

[0084] The performance quality of the graphene oxide sample is evaluated based on the steepness, smoothness, data coverage and tail trend of the area cumulative frequency curve.

[0085] Specifically, the steepness of the curve indicates the distribution of the data. The faster the curve rises, the more concentrated the data is within this area. In other words, if the area cumulative frequency curve is steep within the interval, it means that most of the graphene oxide area is concentrated in this interval. Figure 8 As shown, at 10 μm 2 To 30μm 2 The area cumulative frequency curve rises faster, indicating that most of the graphene oxide particle areas in the sample are concentrated in this interval.

[0086] The smoothness of the curve reflects the continuity of the data distribution. If the area cumulative frequency curve is relatively smooth, it means that the size distribution of graphene oxide in the graphene oxide sample is relatively uniform, without significant discrete points or abnormal data.

[0087] The data coverage indicates the concentration range of graphene oxide sheets in the sample. Figure 8 As shown, the cumulative frequency of most of the data is concentrated at 8 μm 2 To 40μm 2 The accumulation is close to 90%. This shows that the graphene oxide particle area of ​​the sample is concentrated in a smaller range, and the area of ​​most particles is relatively small.

[0088] The tail trend of the area cumulative frequency curve reflects the distribution of large particle flakes in the sample. Figure 8 As shown, from 30μm 2 To 40μm 2In the area of ​​, the curve gradually becomes stable, indicating that there are relatively few data in this area. There are not many maximum data, which means that there are no particularly large particles in the area distribution of graphene oxide.

[0089] In order to verify the technical effect of this embodiment, the evaluation method provided in this embodiment is used to obtain the area cumulative frequency curves of 5 different brands of graphene oxide materials, such as Fig. 9 shown.

[0090] (1) Regarding the rate of increase of cumulative frequency and particle distribution

[0091] Brand 1: The cumulative frequency curve is at the top. Within the same graphene oxide area, the cumulative frequency rises fastest. For example, when the graphene oxide area reaches 20 μm 2 When , the cumulative frequency of brand 1 has reached about 75%, which shows that the particles of brand 1 have occupied a large proportion in a smaller area, indicating that the particle area distribution of brand 1 is concentrated, the particles are small and evenly distributed.

[0092] Brand 2, Brand 3, Brand 5: The curves of these brands rise slightly slower in cumulative frequency, especially in the middle area (20-30μm 2 ), which indicates that the particle distribution of these brands is more dispersed and not concentrated in a smaller area segment like Brand 1. Although Brand 2 and Brand 5 performed slightly better, they still did not reach the concentration of Brand 1.

[0093] Brand 4: The curve of Brand 4 rises the slowest among all brands, indicating that its particle area distribution is more dispersed and the particle size inconsistency is greater.

[0094] (2) Differences when the cumulative frequency approaches 100%

[0095] Brand 1: Graphene oxide area 30-40μm 2 The cumulative frequency is close to 90% or more within the range, indicating that most particles are concentrated in this smaller area, and there are fewer particles in the largest area. This shows that the particle distribution of brand 1 is very uniform, without many abnormally large particles.

[0096] Brand 2 and Brand 5: Although their performance is inferior to that of Brand 1, they can also achieve high cumulative frequencies in a smaller area, indicating that most of the particles are distributed in a relatively small area.

[0097] Brand 3 and Brand 4: In the same area segment (such as 35μm 2 The cumulative frequency increased significantly more slowly, indicating that these brands had more large particles distributed, especially brand 4, which had the most uneven particle distribution and contained more abnormally large particles.

[0098] In summary, the cumulative frequency curve of brand 1 is between 8 and 40 μm graphene oxide sheets. 2 The area range shows a higher cumulative frequency, indicating that the size distribution of its graphene oxide particles is more concentrated, and the proportion of smaller particles is larger, which is better in terms of material uniformity and consistency. Therefore, it can be considered that brand 1 is the best among these brands.

[0099] like Figure 2 The system block diagram of the graphene oxide evaluation system based on image features shown in FIG. 1 includes:

[0100] A calculation module configured to calculate the area of ​​each graphene oxide sheet in each microscopic image of the graphene oxide sample;

[0101] A statistical module, configured to perform statistical analysis on the area of ​​the graphene oxide sheet and draw an area cumulative frequency curve corresponding to the graphene oxide sample;

[0102] An evaluation module is configured to evaluate the performance quality of the graphene oxide sample through the area cumulative frequency curve.

[0103] Specifically, the evaluation system includes a calculation module, a statistics module and an evaluation module. Among them, the calculation module can collect optical microscopic images of different areas of multiple graphene oxide samples through a camera, and calculate the area of ​​graphene oxide sheets of different sizes in each microscopic image respectively. The statistics module can perform statistical analysis on the area values ​​of graphene oxide sheets in all microscopic images, and draw the area cumulative frequency curve corresponding to the graphene oxide sample. The evaluation module can accurately evaluate the morphological characteristics and dispersion quality of the sample through the area cumulative frequency curve. Compared with electron microscopes and atomic force microscopes, there is no need to perform complex operations and the cost is low, so that graphene oxide samples can be evaluated efficiently and economically to meet the evaluation needs of a large number of samples.

[0104] In this embodiment, optionally, the calculation module includes:

[0105] an edge detection unit configured to perform edge detection on the microscopic image to determine the edge profile of each graphene oxide sheet;

[0106] The area calculation unit is configured to calculate the area of ​​each graphene oxide sheet using a polygonal area algorithm according to the corresponding edge contour.

[0107] Specifically, the calculation module includes an edge detection unit and an area calculation unit. The edge detection unit can use a threshold method to binarize the microscopic image, and can use an existing edge detection algorithm to extract the edge contour of each graphene oxide sheet in the binarized microscopic image. The area calculation unit can use a polygonal area algorithm to calculate the contour area of ​​each graphene oxide sheet in the microscopic image. The contour area reflects the size and shape of the graphene oxide sheet. By performing statistical analysis on the area of ​​the graphene oxide sheet in the microscopic image, the characteristics of the graphene oxide sheet in the microscopic image can be quantified and summarized.

[0108] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A graphene oxide evaluation method based on image features, characterized in that: include: Calculate the area of ​​each graphene oxide sheet in each microscopic image of the graphene oxide sample; Performing statistical analysis on the area of ​​the graphene oxide sheet, and drawing an area cumulative frequency curve corresponding to the graphene oxide sample; The performance quality of the graphene oxide sample is evaluated by the area cumulative frequency curve.

2. The graphene oxide evaluation method based on image features according to claim 1, characterized in that: Calculate the area of ​​graphene oxide sheets in a microscopic image of a graphene oxide sample, including: The microscopic image of the graphene oxide sample is homogenized.

3. The graphene oxide evaluation method based on image features according to claim 2, characterized in that: The microscopic image is subjected to homogenization processing, comprising: Acquire a background image of the microscopic image, and perform background correction on the microscopic image using the background image; adjusting the histogram distribution of the corrected microscopic image, and performing equalization processing on a part of the adjusted microscopic image; The microscopic images after equalization were normalized.

4. The graphene oxide evaluation method based on image features according to claim 1, characterized in that: Calculate the area of ​​each graphene oxide sheet in the microscopic image of the graphene oxide sample, including: Performing edge detection on the microscopic image to determine the edge profile of each graphene oxide sheet; The area of ​​each graphene oxide sheet was calculated using a polygonal area algorithm based on the corresponding edge contours.

5. The graphene oxide evaluation method based on image features according to claim 1, characterized in that: Performing statistical analysis on the area of ​​the graphene oxide sheet, including: Counting the frequencies corresponding to graphene oxide sheets of different areas in each of the microscopic images respectively; Accumulating the frequencies corresponding to the graphene oxide sheets of the same area in all the microscopic images to obtain accumulated frequencies corresponding to different areas; The area cumulative frequency curve is drawn based on the accumulated frequencies corresponding to all areas.

6. The graphene oxide evaluation method based on image features according to claim 5, characterized in that: The area cumulative frequency curve is plotted using a semi-logarithmic coordinate system.

7. The graphene oxide evaluation method based on image features according to claim 1, characterized in that: The performance quality of the graphene oxide sample is evaluated by the area cumulative frequency curve, including: The area cumulative frequency curve within the set area interval is intercepted to evaluate the performance quality of the graphene oxide sample.

8. The graphene oxide evaluation method based on image features according to claim 1, characterized in that: The performance quality of the graphene oxide sample is evaluated by the area cumulative frequency curve, including: The performance quality of the graphene oxide sample is evaluated based on the steepness, smoothness, data coverage and tail trend of the area cumulative frequency curve.

9. A graphene oxide evaluation system based on image features, characterized in that: include: A calculation module configured to calculate the area of ​​each graphene oxide sheet in each microscopic image of the graphene oxide sample; A statistical module, configured to perform statistical analysis on the area of ​​the graphene oxide sheet and draw an area cumulative frequency curve corresponding to the graphene oxide sample; An evaluation module is configured to evaluate the performance quality of the graphene oxide sample through the area cumulative frequency curve.

10. The graphene oxide evaluation system based on image features according to claim 9, characterized in that: The calculation module comprises: an edge detection unit configured to perform edge detection on the microscopic image to determine the edge profile of each graphene oxide sheet; The area calculation unit is configured to calculate the area of ​​each graphene oxide sheet using a polygonal area algorithm according to the corresponding edge contour.

Images