Chrominance and hough transform based algal cell counting method, system, and media

Abstract

The application discloses a kind of based on chroma and hough transform algal cell counting method, system and medium, the algal cell counting method based on chroma and hough transform of the present application includes the color removal impurities in water based on microscopic image, obtains the microscopic image after removal impurities in water;The microscopic image of removal impurities in water is grayed and binarized, and the binary image of microscopic image is obtained;Edge detection is carried out for the binary image of microscopic image;Class circle detection is carried out using hough transform for edge detection image, and based on the circle center counting of class circle detection detection result, the counting result of algal cell is obtained, by the color removal impurities in water based on microscopic image, and using hough transform for class circle detection, the present application can conveniently, quickly and accurately obtain the number of algal cell, provide data for water quality detection.

Description

Technical Field

[0001] This invention relates to computer image processing technology, specifically to a method, system, and medium for counting algal cells based on chromaticity and Hough transform. Background Technology

[0002] Eutrophication-induced algal blooms are becoming increasingly serious in my country. Microcystin is a common algal toxin in freshwater lakes, and its presence has been reported in major freshwater lakes and reservoirs in recent years. The occurrence of toxic cyanobacterial blooms not only seriously harms aquatic ecosystems but also poses direct or potential health risks to humans. Studies by scholars both domestically and internationally have found that microcystin is a significant contributing factor to the high incidence of primary liver cancer in China. Microcystin is not only hepatotoxic but also nephrotoxic, neurotoxic, and reproductively toxic, making it a serious environmental toxin that threatens human health. Similarly, in pond fish farming, when Microcystis, Oscillatoria, and Anabaena are dominant, the pond water exhibits a copper-rust-colored, flocculent bloom with a foul odor, harmful to fish. To monitor water quality changes and prevent excessive algal growth, it is necessary to statistically analyze the number of Microcystis cells in the water. Controlling algal blooms and eliminating the harm caused by toxins is now imperative.

[0003] Currently, algal cell counts primarily rely on professionals identifying and counting algae under a microscope using a field-of-view (Field-of-view) counting method. This method typically involves counting algae in five positions within the counting frame: upper left, upper right, lower left, lower right, and center. During the counting process, some algae are within the counting field of view, while others are outside. The upper and lower boundaries, as well as the left and right boundaries, can be defined. If more than half of an algae is within the counting field, it is counted; if half is within and the other half is outside, the upper boundary is recorded, but the lower boundary is ignored; the left boundary is recorded, but the right boundary is ignored. This method is labor-intensive, inefficient, and time-consuming. Therefore, a faster and more convenient method is needed to replace manual counting. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, system and medium for algal cell counting based on chromaticity and Hough transform, which can conveniently, quickly and accurately obtain the number of algal cells and provide data for water quality testing.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A method for counting algal cells based on chromaticity and Hough transform, comprising:

[0007] S101, remove impurities in water based on the colorimetry of the microscopic image to obtain a microscopic image after removing impurities in water;

[0008] S102, the microscopic image after removing impurities from the water is grayscaled and binarized to obtain a binary image of the microscopic image;

[0009] S103 performs edge detection on the binary map of the microscopic image;

[0010] S104. Hough transform is used to perform circle-like detection on the edge detection map, and the circle center is counted based on the circle-like detection results to obtain the algal cell count results.

[0011] Optionally, step S101, which involves removing impurities from water based on the colorimetric properties of a microscopic image, includes: determining a corresponding colorimetric threshold based on the colorimetric properties of the counted algal cell species, and setting other colors in the microscopic image that are less than the colorimetric threshold to white.

[0012] Optionally, step S104 includes:

[0013] S301, Hough transform is used to detect circles in the edge detection map. The range of radius r is determined based on the minimum and maximum radii of algal cells. Circles with radii within the range of radius r are selected as candidate circles.

[0014] S302, traverse through all candidate class circles and select a current candidate class circle;

[0015] S303, using the center coordinates of the current candidate circle as the starting value (m0, n0), the radius of the current candidate circle plus the minimum radius of the algal cell minus 1 is used as the distance l, as shown in the following constraint:

[0016] (m k -m0) 2 +(n k -n0) 2 >l 2 ,

[0017] For the center coordinates (m) of other candidate circles k ,n k The process iterates through the data to determine the number (num) of other candidate circles that do not meet the constraints, and the coordinates (m) of the centers of these other candidate circles. kno ,n kno ) and radius r kno ;

[0018] S304, merge the current candidate circle and its corresponding number of other candidate circles num and other candidate circles into a new circle, and take the average of the center and radius of the merged current candidate circle to obtain the center coordinates (m,n) and radius r of the new circle;

[0019] S305, determine whether the candidate circles have been traversed. If not, jump to step S303; otherwise, count the centers of all unmerged candidate circles and the new circles obtained by merging candidate circles to obtain the algal cell count.

[0020] Optionally, in step S303, when selecting circles whose radii fall within the range of radius r as candidate circles, the function expression for circles whose radii fall within the range of radius r is:

[0021] |(x0-x i ) 2 +(y0-y i ) 2 -r 2 |<ε,

[0022] In the above formula, x0 and y0 are the coordinates of the center of the circle-like circle, x i and y i Let r be the coordinates of a point on the circle-like structure, r be the radius of the circle-like structure, and ε be the difference between the minimum and maximum radii of the algal cell.

[0023] Optionally, in step S305, the function expression for averaging the center and radius of the merged current candidate circles to obtain the new circle's center coordinates (m,n) and radius r is as follows:

[0024]

[0025] In the above formula, round is the floor function, (m kno ,n kno Let r be the coordinates of the center of any other candidate circle that does not meet the constraints. kno Let r be the radius of the other candidate circles that do not satisfy the constraints, num be the number of other candidate circles that do not satisfy the constraints, and r be the radius of the other candidate circles that do not satisfy the constraints. min This is the minimum radius of an algal cell.

[0026] Optionally, before performing edge detection on the binary image of the microscopic image in step S103, morphological filtering is performed on the binary image of the microscopic image to filter out noisy regions, and an opening operation is performed to segment algal cells that are adhered to impurities.

[0027] Optionally, when performing morphological filtering, the value of the central circular region of the kernel is 1, and the value of the remaining regions is 0.

[0028] Optionally, the opening operation includes performing an erosion operation first, and then dilating the microscopic image after the erosion operation.

[0029] Furthermore, the present invention also provides an algal cell counting system based on chromaticity and Hough transform, comprising a microprocessor and a memory interconnected thereto, wherein the microprocessor is programmed or configured to execute the aforementioned algal cell counting method based on chromaticity and Hough transform.

[0030] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program that is programmed or configured by a microprocessor to perform the aforementioned algal cell counting method based on chromaticity and Hough transform.

[0031] Compared with existing technologies, the present invention has the following main advantages: The algal cell counting method based on chromaticity and Hough transform of the present invention includes removing impurities in water based on the chromaticity of a microscopic image to obtain a microscopic image after removing impurities; converting and binarizing the microscopic image after removing impurities to grayscale to obtain a binary image of the microscopic image; performing edge detection on the binary image of the microscopic image; performing circle-like detection using Hough transform on the edge detection image, and counting the center of the circle based on the circle-like detection result to obtain the algal cell counting result. By removing impurities in water based on the chromaticity of a microscopic image and performing circle-like detection using Hough transform, the present invention can conveniently, quickly and accurately obtain the number of algal cells, providing data for water quality testing. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the basic process of the method in an embodiment of the present invention.

[0033] Figure 2 This is a microscopic image input in an embodiment of the present invention.

[0034] Figure 3 This is a microscopic image after removing impurities from the water in an embodiment of the present invention.

[0035] Figure 4 This is a binary image of a microscopic image in an embodiment of the present invention.

[0036] Figure 5 This refers to the kernel used for morphological filtering in this embodiment of the invention.

[0037] Figure 6 The image shown is a microscopic image obtained after opening operations in an embodiment of the present invention.

[0038] Figure 7 This is the edge detection map obtained in an embodiment of the present invention.

[0039] Figure 8 This is a diagram showing the counting effect obtained in an embodiment of the present invention. Detailed Implementation

[0040] The following section will use Microcystis aeruginosa as an example of algal cells to further illustrate the algal cell counting method, system, and medium based on chromaticity and Hough transform of the present invention.

[0041] like Figure 1 As shown, the algal cell counting method based on chromaticity and Hough transform in this embodiment includes:

[0042] S101, remove impurities in water based on the colorimetry of the microscopic image to obtain a microscopic image after removing impurities in water;

[0043] S102, the microscopic image after removing impurities from the water is grayscaled and binarized to obtain a binary image of the microscopic image;

[0044] S103 performs edge detection on the binary map of the microscopic image;

[0045] S104. Hough transform is used to perform circle-like detection on the edge detection map, and the circle center is counted based on the circle-like detection results to obtain the algal cell count results.

[0046] In this embodiment, step S101, which involves removing impurities from water based on the colorimetry of a microscopic image, includes: determining a corresponding colorimetry threshold based on the colorimetry of the counted algal cell species, and setting other colors in the microscopic image that are less than the colorimetry threshold to white. Figure 2 The input microscopic image in this embodiment shows that the pigments of Microcystis are mostly green. The pigments of Microcystis can be extracted for colorimetry. Then, a margin is set as a colorimetric threshold, and other colors in the microscopic image below this threshold are set to white. This yields the microscopic image after removing impurities from the water, as shown below. Figure 3 As shown.

[0047] In step S102, the Ostu algorithm is used to binarize the grayscale microscopic image. Other binarization algorithms can also be used as needed. Specifically, this embodiment... Figure 3 After grayscale transformation, the Ostu algorithm is used, with class variance as the discrimination criterion. The threshold that maximizes the inter-class variance and minimizes the intra-class variance is selected as the optimal threshold. Then, the grayscale image is binarized to obtain the binary image of the microscopic image, as shown below. Figure 4 As shown.

[0048] As an optional implementation, step S103 of this embodiment, before edge detection on the binary image of the microscopic image, further includes morphological filtering of the binary image of the microscopic image to filter out noise regions, and performing an opening operation to segment algal cells adhering to impurities. Based on the morphological characteristics analysis of algal cells, their shape is very similar to a circle. In this embodiment, when performing morphological filtering, the value of the central circular region of the nucleus is 1, and the value of the remaining regions is 0. Specifically, as shown... Figure 5As shown, morphological filtering using a kernel similar to a circle can improve the accuracy in noisy regions. In this embodiment, the opening operation includes first performing an erosion operation, and then dilating the eroded microscopic image. This effectively segments algal cells adhering to impurities. The microscopic image obtained after the opening operation is shown below. Figure 6 As shown.

[0049] In this embodiment, step S103, edge detection of the binary image of the microscopic image, is implemented using an edge detection algorithm. The edge detection algorithm is an existing image algorithm; for example, as an optional implementation, this embodiment uses the Canny algorithm, which makes edges more complete, accurate, and refined. The process of edge detection of the binary image of the microscopic image using the Canny algorithm includes: performing Gaussian filtering on the image; calculating the gradient magnitude and direction using the Sobel operator template; performing non-maximum suppression on the gradient magnitude based on the gradient direction angle; detecting and connecting edges using a double threshold algorithm. After non-maximum suppression, many discrete points are obtained. Points that are truly edges are connected, while isolated noise is removed. Using a double threshold, edges greater than the higher threshold are considered strong edges, those less than the lower threshold are not edges, and those in between are weak edges. If a strong edge exists within the 8-neighborhood of a weak edge, the weak edge becomes a strong edge; otherwise, it is not an edge. Finally, the edge detection image obtained by edge detection of the binary image of the microscopic image using the Canny algorithm is shown below. Figure 7 As shown.

[0050] When analyzing the morphological characteristics of algal cells, the algal cells are approximated as circular. Therefore, in this embodiment, Hough transform is used for circle-like detection. The principle of circle-like detection using Hough transform is to use the duality of points and lines to map the edge pixels of the image space to the parameter space. Then, all point elements in the parameter space are projected into an accumulator for accumulation and evaluation. The center position and radius of the candidate circle are determined based on the calculated accumulated value. In this embodiment, step S104 includes:

[0051] S301, Hough transform is used to detect circles in the edge detection map. The range of radius r is determined based on the minimum and maximum radii of algal cells. Circles with radii within the range of radius r are selected as candidate circles.

[0052] S302, traverse through all candidate class circles and select a current candidate class circle;

[0053] S303, using the center coordinates of the current candidate circle as the starting value (m0, n0), the radius of the current candidate circle plus the minimum radius of the algal cell minus 1 is used as the distance l, as shown in the following constraint:

[0054] (m k -m0) 2 +(n k-n0) 2 >l 2 ,

[0055] For the center coordinates (m) of other candidate circles k ,n k The process iterates through the data to determine the number (num) of other candidate circles that do not meet the constraints, and the coordinates (m) of the centers of these other candidate circles. kno ,n kno ) and radius r kno ;

[0056] S304, merge the current candidate circle and its corresponding number of other candidate circles num and other candidate circles into a new circle, and take the average of the center and radius of the merged current candidate circle to obtain the center coordinates (m,n) and radius r of the new circle;

[0057] S305, determine whether the candidate circles have been traversed. If not, jump to step S303; otherwise, count the centers of all unmerged candidate circles and the new circles obtained by merging candidate circles to obtain the algal cell count.

[0058] In the Cartesian coordinate system, the mathematical expression for a circle is as follows:

[0059] (x i -x0) 2 +(y i -y0) 2 =r 2 , i = (1,2,3,...,n),

[0060] Where (x0, y0) are the coordinates of the center of the circle, and r is the radius of the circle. i ,y i Let be the coordinates on the circumference of the circle. The Hough transform converts the circle in the image space to the parameter space, and the mathematical expression for the circle in the parameter space is as follows:

[0061] (x0-x i ) 2 +(y0-y i ) 2 =r 2 ,

[0062] When determining the center coordinates and radius of a circle in the original image using the Hough transform, a three-dimensional accumulator A(x0, y0, r) needs to be established in the parameter space, where r is a variable. When analyzing the morphology of algal cells, the minimum radius is determined, and adding the appropriate maximum radius gives the range of the variable r. Each pixel on the circumference of the source image is accumulated according to the formula in the parameter space. By calculating the maximum value A max (x0, y0, r) in the accumulator A(x0, y0, r), and selecting an appropriate threshold value T (0 < T < 1), the center coordinates and radius in the accumulator A(x0, y0, r) that exceed ε * A max (x0, y0, r) are obtained. Therefore, when considering the circle-like shapes with radii within the range of the radius r as candidate circle-like shapes in step S303 of this embodiment, the functional expression of the circle-like shapes with radii within the range of the radius r is:

[0063] |(x0 - x i ) 2 + (y0 - y i ) 2 - r 2 | < ε,

[0064] In the above formula, x0 and y0 are the center coordinates of the circle-like shape, x i and y i are the coordinates of a point on the circle-like shape, r is the radius of the circle-like shape, and ε is the difference between the minimum radius and the maximum radius of the algal cells.

[0065] In step S303, taking the center coordinates of the current candidate circle-like shape as the starting values (m0, n0), adding the minimum radius value of the algal cells minus 1 to the radius of the current candidate circle-like shape as the distance l, with the following constraint condition:

[0066] (m k - m0) 2 + (n k - n0) 2 > l 2 ,

[0067] For the center coordinates (m k , n k ) of other candidate circle-like shapes, traverse to determine the number num of other candidate circle-like shapes that do not satisfy the constraint condition, as well as the center coordinates (m kno , n kno ) and radius r kno of the other candidate circle-like shapes that do not satisfy the constraint condition; if the center coordinates do not satisfy the above formula, then accumulate the initial center coordinates and the center coordinates of the unsatisfied conditions correspondingly, and accumulate the radii correspondingly. For the circles (m kno , n kno , r knoThe system calculates the number of unsatisfied conditions (num) and then iterates through all candidate circle center coordinates. After traversing all candidate circle center coordinates, the final accumulated value is divided by the number of unsatisfied conditions to obtain the average center coordinates. The radius is also calculated by dividing the average and adding the minimum radius value minus 1. In this embodiment, the function expression for averaging the center coordinates (m,n) and radius r of the merged current candidate circles in step S305 is:

[0068]

[0069] In the above formula, round is the floor function, (m kno ,m kno Let r be the coordinates of the center of any other candidate circle that does not meet the constraints. kno Let r be the radius of the other candidate circles that do not satisfy the constraints, num be the number of other candidate circles that do not satisfy the constraints, and r be the radius of the other candidate circles that do not satisfy the constraints. min This is the minimum radius of an algal cell.

[0070] Finally, after traversing all candidate circles, the position, size, and number of algal cells in the entire constrained image can be obtained to produce a counting result image (the obtained circles are then appended to the original image). Figure 8 As shown in Table 1, the results obtained by the traditional manual counting method (hereinafter referred to as manual counting) and the results obtained by the algal cell counting method based on chromaticity and Hough transform in this embodiment (hereinafter referred to as the counting by this method) are compared for multiple microscopic images.

[0071] Table 1: Results of manual counting and counting using this method.

[0072] Serial Number Manual counting This method counts. accuracy 1 29 29 100％ 2 34 33 97％ 3 35 33 94％ 4 37 37 100％ 5 40 40 100％ 6 66 63 95％ 7 118 116 98％ 8 162 164 99％ 9 188 186 99％ 10 202 200 99％

[0073] Referring to Table 1, the function expression for calculating accuracy is:

[0074] 1-(|AB|) / A,

[0075] In the above formula, A represents the result obtained by the traditional manual counting method (hereinafter referred to as manual counting), and B represents the result obtained by the algal cell counting method based on chromaticity and Hough transform in this embodiment (hereinafter referred to as the counting by this method). As shown in Table 1, the algal cell counting method based on chromaticity and Hough transform in this embodiment has a high counting accuracy. Furthermore, the algal cell counting method based on chromaticity and Hough transform in this embodiment is implemented using a computer program, thus offering advantages such as fast counting speed and high counting efficiency. In summary, the algal cell counting method based on chromaticity and Hough transform in this embodiment can conveniently, quickly, and accurately obtain the number of algal cells, thereby providing data for water quality testing. Undoubtedly, the algal cell counting method based on chromaticity and Hough transform in this embodiment is not only applicable to the cell counting of Microcystis aeruginosa, but also to the cell counting of other similar round or near-round algae.

[0076] Furthermore, this embodiment also provides an algal cell counting system based on chromaticity and Hough transform, including a microprocessor and a memory interconnected thereto. The microprocessor is programmed or configured to execute the aforementioned algal cell counting method based on chromaticity and Hough transform. Additionally, this embodiment also provides a computer-readable storage medium storing a computer program for being programmed or configured by the microprocessor to execute the aforementioned algal cell counting method based on chromaticity and Hough transform.

[0077] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0078] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for counting algal cells based on chrominance and Hough transform, characterized by, The method comprises: S101, removing impurities in water based on chrominance of the microscopic image to obtain a microscopic image after removing impurities in water; S102, gray-scale and binarization of the microscopic image after removing impurities in water to obtain a binary image of the microscopic image; S103, edge detection on the binary image of the microscopic image; S104, circle detection on the edge detection image by using Hough transformation, and circle center counting based on the circle detection result to obtain a counting result of the algae cells; Step S104 comprises: S301, adopt Hough transform for edge detection graph to perform circle-like detection, determine radius according to minimum radius and maximum radius of algal cells , take the circle-like with the radius in the value range of the radius as a candidate circle-like. S302, selecting a current candidate circle from all candidate circles; S303, taking the center coordinate of the current candidate class circle as a starting value , the radius of the current candidate class circle plus the minimum radius value of the algal cells minus 1 as the distance , the constraint condition shown in the following formula: ， center coordinates of the other candidate class circles that do not satisfy the constraint condition the number of other candidate class circles that do not satisfy the constraint condition , and the center coordinates of the other candidate class circles that do not satisfy the constraint condition and the radius ; S304, for the current candidate class circle and its corresponding other candidate class circle number And other candidate class circle into a new circle, and the average value of the merging of the current candidate class circle and the radius of the circle center coordinate of the new circle And radius ; S305, determining whether the candidate circles are traversed, if not, jumping to step S303; otherwise, counting the circle centers for all candidate circles not merged and new circles obtained by merging the candidate circles to obtain the counting result of the algae cells.

2. The colorimetric and Hough transform based algal cell counting method according to claim 1, wherein, In step S101, removing impurities in water based on chrominance of the microscopic image comprises: determining a corresponding chrominance threshold according to the chrominance of the algae cell species to be counted, and setting other chrominances in the microscopic image smaller than the chrominance threshold as white.

3. The colorimetric and Hough transform based algal cell counting method according to claim 1, wherein, When the circle-like shape in step S303 is taken as a candidate circle-like shape, the function expression of the circle-like shape with a radius in the value range of the radius is as follows: When the circle-like shape in step S303 is taken as a candidate circle-like shape, the function expression of the circle-like shape with a radius in the value range of the radius is as follows: ， In the above formula, and is the center coordinate of the circle, and is the coordinate of a point on the circle, is the radius of the circle, is the difference between the minimum radius and the maximum radius of the algal cell.

4. The colorimetric and Hough transform based algal cell counting method according to claim 1, wherein, In step S305, the center and the radius of the merged current candidate class circle are averaged to obtain the center coordinate of the new circle and the radius of the function expression are as follows: ， In the above formula, round is a rounding function, the center coordinates of the other candidate circles that do not satisfy the constraint condition, the radii of the other candidate circles that do not satisfy the constraint condition, the number of the other candidate circles that do not satisfy the constraint condition, the minimum radius of the algal cells.

5. The method for counting algal cells based on the colorimetric and Hough transform according to any one of claims 1 to 4, characterized in that, Before the edge detection on the binary image of the microscopic image in step S103, the method further comprises morphological filtering on the binary image of the microscopic image to filter noise regions, and open operation to separate the algae cells adhered to the impurities.

6. The colorimetric and Hough transform based algal cell counting method according to claim 5, wherein, In the morphological filtering, the value of the central circular region of the kernel is 1, and the values of the other regions are 0.

7. The colorimetric and Hough transform based algal cell counting method as claimed in claim 5, wherein, The open operation comprises: first, erosion operation, and then, inflation on the microscopic image after the erosion operation.

8. A colorimetric and Hough transform based algal cell counting system comprising a microprocessor and a memory interconnected, characterized in that, The microprocessor is programmed or configured to execute the algae cell counting method based on chrominance and Hough transformation in any one of claims 1-7.

9. A computer-readable storage medium having stored therein a computer program, characterized in that, The computer program is used for programming or configuring the microprocessor to execute the algae cell counting method based on chrominance and Hough transformation in any one of claims 1-7.

Images