Bacteria identification and counting method based on pulse coupling neural network

Abstract

The invention provides a bacterium identification and counting method based on a pulse coupling neural network. According to the method, firstly, an image is preprocessed through Gamma transformation and spatial domain filtering so as to enhance the characteristics of bacteria; and then, using a pulse coupling neural network to identify the central brightness features of the bacteria in the image. And finally, counting the bacteria by adopting a contour detection technology. According to the method, Gamma transformation, spatial domain filtering, a pulse coupling neural network, morphological operation, morphological reconstruction and contour detection technologies are integrated, and accurate identification and automatic counting of bacteria can be effectively realized. The method has the advantages of being high in accuracy and efficiency, remarkably reducing dependence of manual operation, reducing operation cost and time consumption, and ensuring objectivity and repeatability of results.

Description

Technical Field

[0001] The invention relates to the fields of microbiology and image recognition, and in particular to a bacteria recognition and counting method based on a pulse coupled neural network. Background Art

[0002] Bacteria counting is crucial in microbiology research. Traditional methods rely on manual operation and face challenges such as high labor consumption, slow speed and subjective errors. In recent years, computer vision technology has been widely used in the field of bacteria identification. The pulse coupled neural network algorithm can accurately identify the central brightness characteristics of bacteria and count bacteria through subsequent processing. This method significantly reduces the reliance on manual operation, thereby reducing operating costs and time consumption, while ensuring the objectivity and repeatability of the counting results. Summary of the invention

[0003] In view of this, the present invention proposes a highly efficient bacteria identification and counting method.

[0004] The technical solution of the present invention is implemented as follows: The present invention provides a method for identifying and counting bacteria based on a pulse coupled neural network, which comprises the following steps:

[0005] (1) Taking photos to collect bacterial image samples;

[0006] (2) Upload images to the system;

[0007] (3) Load the image in grayscale mode;

[0008] (4) Performing image enhancement processing on the image;

[0009] (5) Perform contrast enhancement on the image;

[0010] (6) Pulse coupled neural network is used to identify the central brightness characteristics of bacteria;

[0011] (7) Perform morphological operations on the image;

[0012] (8) Count bacteria using contour detection method;

[0013] On the basis of the above counting scheme, preferably, in the step (3), the pixel values ​​of the color channels (R, G, B) are converted into gray values ​​using a weighted average method, so that the color image is loaded in the mode of a gray image, that is, when Gray = 0.299 × R + 0.587 × G + 0.114 × B, the most reasonable gray image is obtained.

[0014] Based on the above counting scheme, preferably, in step (4), the image after grayscale loading is enhanced using Gamma transformation, and the Gamma value is set to 1.5 to darken the image and enhance the contrast.

[0015] Based on the above counting scheme, preferably, in step (5), a spatial domain filtering method is used to perform contrast enhancement processing on the image after Gamma transformation processing, the size of the two-dimensional array is 3x3, and the values ​​of all elements are 1, and all elements are divided by 9 to form a mean filter, thereby smoothing the image, reducing details and noise, and making the image more blurred; the mean filter averages each pixel of the image and the pixels in its neighborhood, thereby reducing local contrast differences.

[0016] Based on the above counting scheme, preferably, in step (6), a pulse coupled neural network is used to perform feature recognition on the image after spatial domain filtering to obtain a segmented image; the activation threshold is 0.1, and the maximum number of iterations is 10 times.

[0017] Based on the above counting scheme, preferably, in step (7), the image is further processed using morphological processing and morphological reconstruction to make the image area more coherent.

[0018] Based on the above counting scheme, preferably, in step (8), a contour detection method is used to display the final counting result.

[0019] The bacteria identification and counting method based on pulse coupled neural network of the present invention has the following beneficial effects compared with the prior art:

[0020] (1) The image is processed by gamma transform and spatial domain filtering, so that the pulse coupled neural network can well identify the central brightness of bacteria.

[0021] (2) The use of morphological reconstruction makes the image area more coherent and effectively removes noise.

[0022] (3) Gamma transform method, spatial domain filtering method, pulse coupled neural network, morphological operation, morphological reconstruction and contour detection method are used to realize the recognition and automatic counting of bacteria with good accuracy and high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0024] Figure 1 The present invention is a flow chart of a method for identifying and counting bacteria based on a pulse coupled neural network. DETAILED DESCRIPTION

[0025] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] like Figure 1 As shown, a method for identifying and counting bacteria based on a pulse coupled neural network of the present invention comprises the following steps:

[0027] (1) Taking photos to collect bacterial image samples;

[0028] (2) Upload images to the system;

[0029] (3) Load the image in grayscale mode;

[0030] (4) Performing image enhancement processing on the image;

[0031] (5) Perform contrast enhancement on the image;

[0032] (6) Pulse coupled neural network is used to identify the central brightness characteristics of bacteria;

[0033] (7) Perform morphological operations on the image;

[0034] (8) Count bacteria using contour detection method;

[0035] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying and counting bacteria based on pulse coupled neural networks, characterized in that: The steps include: (1) Taking photos to collect bacterial image samples; (2) Upload images to the system; (3) Load the image in grayscale mode; (4) Performing image enhancement processing on the image; (5) Perform contrast enhancement on the image; (6) Pulse coupled neural network is used to identify the central brightness characteristics of bacteria; (7) Perform morphological operations on the image; (8) Bacteria were counted using the contour detection method.

2. The method according to claim 1, characterized in that In the step (3), the pixel values ​​of the color channels (R, G, B) are converted into grayscale values ​​using a weighted average method, so that the color image is loaded in the mode of a grayscale image, that is, when Gray = 0.299×R+0.587×G+0.114×B, the most reasonable grayscale image is obtained.

3. The method according to claim 1, characterized in that In the step (4), the image after grayscale loading is enhanced by using Gamma transformation, and the Gamma value is set to 1.5 to darken the image and enhance the contrast.

4. The method according to claim 1, characterized in that: In the step (5), the spatial domain filtering method is used to perform contrast enhancement processing on the image after the Gamma transform processing. The size of the two-dimensional array is 3x3, and the values ​​of all elements are 1. All elements are divided by 9 to form a mean filter, thereby smoothing the image, reducing details and noise, and making the image more blurred; the mean filter averages each pixel of the image and the pixels in its neighborhood, thereby reducing local contrast differences.

5. The method according to claim 1, characterized in that In the step (6), a pulse coupled neural network is used to perform feature recognition on the image after spatial domain filtering to obtain a segmented image; the activation threshold is 0.1, and the maximum number of iterations is 10.

6. The method according to claim 1, characterized in that In the step (7), the image is further processed using morphological processing and morphological reconstruction to make the image area more coherent.

7. The method according to claim 1, characterized in that In the step (8), a contour detection method is used to display the final counting result.

Images