Search

CN-122023139-A - Rhizosphere microbial community distribution analysis method based on image enhancement

CN122023139ACN 122023139 ACN122023139 ACN 122023139ACN-122023139-A

Abstract

The invention relates to the technical field of image enhancement, in particular to a rhizosphere microbial community distribution analysis method based on image enhancement, which comprises the following steps of: and acquiring a continuous microscopic image sequence of the rhizosphere microenvironment, converting the continuous microscopic image sequence into a gray data space, and constructing a standard input frame set for optical flow calculation. According to the invention, the stability scores of all the features are calculated by utilizing the signal-to-noise ratio stability measurement function, and the long life cycle features are screened and reserved, so that random imaging noise in a short life cycle can be filtered fundamentally, a target community topological structure is obtained, the topological structure is mapped back to an original image space by utilizing an inverse persistence mapping algorithm and is fused with a continuous microscopic image sequence, and finally weak biological signals are obviously enhanced while the original form details are reserved, so that high signal-to-noise ratio analysis of space-time distribution of microbial communities in a complex rhizosphere environment is realized.

Inventors

  • SONG LIGUO
  • Liu Taikun
  • ZHANG RUI

Assignees

  • 临沂科技职业学院

Dates

Publication Date
20260512
Application Date
20260202

Claims (10)

  1. 1. The rhizosphere microbial community distribution analysis method based on image enhancement is characterized by comprising the following steps of: Acquiring a continuous microscopic image sequence of a rhizosphere microenvironment, converting the continuous microscopic image sequence into a gray data space, and constructing a standard input frame set for optical flow calculation; Based on an energy functional model introducing static constraint, calculating a dense optical flow field between every two adjacent pixel frames in the standard input frame set, solving a two-dimensional motion vector of each pixel point, and generating original optical flow field data containing full-view field motion information; Analyzing the original optical flow field data by using a vector field differential operator, extracting a divergence characteristic component and a rotation characteristic component in the original optical flow field data, generating a bioactive intensity map through weighted fusion calculation, and generating active microorganism mask data for removing a static soil background according to the bioactive intensity map; extracting an effective area of the bioactive intensity map by utilizing the active microorganism mask data, mapping the extracted data to a topological data space, constructing a cubic complex based on pixel point amplitude, executing multi-scale filtering operation on the cubic complex, and recording the generation time and the extinction time of topological features in the cubic complex; Generating a continuous coherent bar code map according to the generation moment and the extinction moment, calculating the stability score of each characteristic in the continuous coherent bar code map by utilizing a signal-to-noise ratio stability measurement function, and screening and retaining long life cycle characteristics representing a real biological aggregation structure according to the stability score to obtain a target community topological structure; And mapping the target community topological structure back to an original image space by using an inverse persistence mapping algorithm, and fusing the target community topological structure with the continuous microscopic image sequence to generate a rhizosphere microbial community space-time distribution analysis result.
  2. 2. The image enhancement-based rhizosphere microbial community distribution analysis method according to claim 1, wherein the step of calculating a dense optical flow field between every two adjacent pixel frames in the standard input frame set adopts a global energy function minimization strategy; The calculation formula of the global energy function is as follows: ; Wherein, the A value representing the global energy function, 、 、 Respectively representing the standard input frame set in space Direction, space Direction and time The gray partial derivative of the direction, Representing the two-dimensional motion vector in The component of the direction is used to determine, Representing the two-dimensional motion vector in The component of the direction is used to determine, Representing the smoothed weight coefficient(s), Representing the components Is used for the gradient vector of (a), Representing the components Is used for the gradient vector of (a), Representing the square of the vector modulus, Representing the weight coefficient of the static constraint, And the static probability map function of the root system is represented and is used for inhibiting misjudgment movement of the root system area at the optical flow calculation level.
  3. 3. The image-enhancement-based rhizosphere microbial community distribution analysis method according to claim 2, wherein in the calculation formula of the global energy function, the root system static probability map function The construction method of (1) comprises the following steps: calculating pixel gray level variance of the standard input frame set on a time axis, marking a region with the pixel gray level variance lower than a preset static threshold as a high-probability static region, and giving the region with the pixel gray level variance to the standard input frame set A maximum value; marking the area with the pixel gray variance higher than the preset static threshold value as a low-probability static area, and endowing the area with the pixel gray variance A minimum value; by a third term in the global energy function Zero-forcing penalty is applied to the two-dimensional motion vectors of pixel points located in the high probability static region.
  4. 4. The image-enhancement-based rhizosphere microbial community distribution analysis method according to claim 1, wherein the step of analyzing the original optical flow field data by using a vector field differential operator and extracting a divergence characteristic component and a rotation characteristic component in the original optical flow field data specifically comprises the steps of: Calculating the source-sink strength of the two-dimensional motion vector by using a divergence operator to obtain the divergence characteristic component; Calculating the rotation strength of the two-dimensional motion vector by using a rotation operator to obtain the rotation characteristic component; Fusing the divergence characteristic component and the rotation characteristic component by adopting a nonlinear biological dynamics index formula to obtain the biological activity intensity graph; the nonlinear biological dynamic index formula is: ; Wherein, the Representing the biological activity value of the current pixel point in the biological activity intensity graph, Representing the normalized gain factor of the gain signal, The two-dimensional motion vector is represented and, The characteristic component of the divergence is represented, Representing the characteristic component of the rotation, A modulus value representing the two-dimensional motion vector, A characteristic scale parameter representing the speed of movement.
  5. 5. The method for analyzing rhizosphere microbial community distribution based on image enhancement according to claim 4, wherein the step of generating active microbial mask data for removing static soil background according to the bioactive intensity map specifically comprises the following steps: performing histogram statistics on the bioactive intensity graph, and calculating an optimal segmentation threshold by adopting a self-adaptive maximum inter-class variance method; performing binarization processing on the bioactive intensity map by using the optimal segmentation threshold value to generate a preliminary binary mask; And performing morphological open operation on the preliminary binary mask to eliminate isolated noise points, and performing morphological closed operation on the preliminary binary mask to fill internal cavities to obtain the active microorganism mask data.
  6. 6. The image-enhancement-based rhizosphere microbial community distribution analysis method according to claim 1, wherein the step of constructing a cubic complex based on pixel point amplitude comprises the following steps: Defining each pixel point in the bioactive intensity map, which is positioned in the active microorganism mask data area, as a vertex, establishing a connecting edge between adjacent pixel points, and defining an area formed by four adjacent pixel points as a two-dimensional patch to form the cubic complex; The step of executing the multi-scale filtering operation comprises the steps of setting an increasing numerical threshold sequence, and sequentially activating elements in the cubic complex according to the numerical threshold sequence; Calculating the Betty number in the current activation state, wherein the Betty number comprises a zero-dimensional Betty number representing a communication component and a one-dimensional Betty number representing a hole structure; and recording a threshold value of each topological feature appearing for the first time as the generation moment, and recording a threshold value of merging or disappearing of the topological features as the extinction moment.
  7. 7. The image enhancement-based rhizosphere microbial community distribution analysis method according to claim 1, wherein in the step of calculating the stability score of each feature in the continuous coherent barcode map using a signal-to-noise ratio stability metric function, the formula of the signal-to-noise ratio stability metric function is: ; Wherein, the The stability score is indicated as a function of the stability score, Indicating the moment of said extinction of the blood, The time of the generation is indicated as such, The length of the lifecycle of the features is indicated, Representing a pre-statistically derived average life cycle of the ambient random noise, Representing the corresponding projected area of the topological feature in image space.
  8. 8. The method of claim 7, wherein the step of screening for retention of long life cycle characteristics representative of true bioaggregation structures based on the stability score comprises: setting a significance determination threshold, comparing the stability score with the significance determination threshold; If the stability score is lower than the significance judgment threshold, judging that the corresponding topological feature is short noise and eliminating the short noise; And if the stability score is higher than or equal to the significance judging threshold, judging that the corresponding topological feature is a strip signal, and reserving a topological generator corresponding to the strip signal as the target community topological structure.
  9. 9. The image-enhancement-based rhizosphere microbial community distribution analysis method of claim 1, wherein the step of mapping the target community topology back to the original image space using an inverse persistence mapping algorithm, and fusing with the continuous microscopic image sequence, comprises: positioning a coordinate index of a topology generating element corresponding to the target community topology structure in an original pixel space; generating a topology enhancement layer with the same resolution as the continuous microscopic image sequence, and highlighting a region corresponding to the coordinate index in the topology enhancement layer; and stacking the topological enhancement map to the continuous microscopic image sequence by utilizing Alpha mixing, generating a space-time distribution analysis result of the rhizosphere microbial community, and generating a thermodynamic diagram reflecting the space occupation density of microorganisms based on the coordinate index.
  10. 10. The image-enhanced rhizosphere microbial community distribution analysis method of claim 1, further comprising, prior to acquiring the sequence of successive microscopic images of the rhizosphere microenvironment: collecting dark noise images under the condition of no illumination, and calculating the fixed mode noise distribution of the sensor; When the continuous microscopic image sequence is acquired, dark field correction is carried out on each frame of image by utilizing the fixed pattern noise distribution; and carrying out histogram equalization enhancement on the corrected continuous microscopic image sequence, and expanding the dynamic range of the image.

Description

Rhizosphere microbial community distribution analysis method based on image enhancement Technical Field The invention relates to the technical field of image enhancement, in particular to a rhizosphere microbial community distribution analysis method based on image enhancement. Background Image enhancement techniques transform and process digital image pixels through specific mathematical models and algorithms to improve the visual perception quality of the image or highlight feature information in the image that is critical to machine vision analysis. The existing image enhancement technology mainly relies on pixel level transformation or spatial filtering processing of a single frame image, focuses on improving overall visual contrast or smoothing statistical noise, often faces the dilemma that signals and backgrounds are difficult to separate when microscopic observation data containing complex solid-liquid-gas three-phase media are processed, and is difficult to distinguish static soil impurities and dynamic active targets with similar gray features only by gray histogram adjustment or edge sharpening operation due to the lack of effective utilization of time dimension motion information, so that background noise points are synchronously amplified while target textures are enhanced, and the false detection rate in subsequent analysis is increased. Therefore, improvements are needed. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides an image enhancement-based rhizosphere microbial community distribution analysis method. In order to achieve the above purpose, the invention adopts the following technical scheme that the rhizosphere microbial community distribution analysis method based on image enhancement comprises the following steps: Acquiring a continuous microscopic image sequence of a rhizosphere microenvironment, converting the continuous microscopic image sequence into a gray data space, and constructing a standard input frame set for optical flow calculation; Based on an energy functional model introducing static constraint, calculating a dense optical flow field between every two adjacent pixel frames in the standard input frame set, solving a two-dimensional motion vector of each pixel point, and generating original optical flow field data containing full-view field motion information; Analyzing the original optical flow field data by using a vector field differential operator, extracting a divergence characteristic component and a rotation characteristic component in the original optical flow field data, generating a bioactive intensity map through weighted fusion calculation, and generating active microorganism mask data for removing a static soil background according to the bioactive intensity map; extracting an effective area of the bioactive intensity map by utilizing the active microorganism mask data, mapping the extracted data to a topological data space, constructing a cubic complex based on pixel point amplitude, executing multi-scale filtering operation on the cubic complex, and recording the generation time and the extinction time of topological features in the cubic complex; Generating a continuous coherent bar code map according to the generation moment and the extinction moment, calculating the stability score of each characteristic in the continuous coherent bar code map by utilizing a signal-to-noise ratio stability measurement function, and screening and retaining long life cycle characteristics representing a real biological aggregation structure according to the stability score to obtain a target community topological structure; And mapping the target community topological structure back to an original image space by using an inverse persistence mapping algorithm, and fusing the target community topological structure with the continuous microscopic image sequence to generate a rhizosphere microbial community space-time distribution analysis result. Preferably, the step of calculating a dense optical flow field between every two adjacent pixel frames in the standard input frame set adopts a global energy function minimization strategy; The calculation formula of the global energy function is as follows: ; Wherein, the A value representing the global energy function,、、Respectively representing the standard input frame set in spaceDirection, spaceDirection and timeThe gray partial derivative of the direction,Representing the two-dimensional motion vector inThe component of the direction is used to determine,Representing the two-dimensional motion vector inThe component of the direction is used to determine,Representing the smoothed weight coefficient(s),Representing the componentsIs used for the gradient vector of (a),Representing the componentsIs used for the gradient vector of (a),Representing the square of the vector modulus,Representing the weight coefficient of the static constraint,And the static probability map function of the