CN-122023452-A - Spore capturing and identifying method and spore capturing analyzer

CN122023452ACN 122023452 ACN122023452 ACN 122023452ACN-122023452-A

Abstract

The embodiment of the invention provides a spore capturing and identifying method and a spore capturing and analyzing instrument thereof, and relates to the technical field of spore capturing and identifying technology. The method comprises the steps of obtaining an original spore image, preprocessing the original spore image to obtain a first image, carrying out self-adaptive threshold segmentation processing on the first image to obtain a regional mask, carrying out first processing on the basis of the regional mask, carrying out multi-feature verification on a first separation result, and determining spore information according to the multi-feature verification result. The invention solves the problem of low image segmentation precision, thereby achieving the effect of improving spore segmentation recognition precision.

Inventors

QIU HUIXUE
SONG YUEJIAO
WANG LI
Gu Dongjiao
CAO KANGQI
PANG XILONG
ZHANG YUXI
HE SHILEI
YUAN JUNCHAO
SHAO GUANGYU
LIU HAITAO
WANG RUIJIN
Gu Yingchen

Assignees

山东中天宇信信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20251005

Claims (10)

1. A spore capture and identification method, comprising: Acquiring an original image of spores; Preprocessing the original spore image to obtain a first image; Performing self-adaptive threshold segmentation processing on the first image to obtain a region mask; Performing a first process based on the region mask, the first process including performing a connected region analysis process on the region mask to determine overlapping clusters; And performing multi-feature verification on the first separation result, and determining spore information according to the multi-feature verification result.
2. The method of claim 1, wherein said subjecting the overlapped mass to a first separation process by a preset first algorithm comprises: carrying out Euclidean distance transformation on the overlapped block area to calculate the shortest distance from each pixel point in the overlapped block to a background pixel point, wherein the distance value is in direct proportion to the probability value of the core area of the spores; Threshold processing is carried out on Euclidean distance transformation results to obtain a plurality of core marks, wherein each mark corresponds to the center of one spore; morphological dilation of the core markers to ensure that each of the core markers covers a spore core region; watershed segmentation is carried out on the overlapped lumps based on the core mark so as to obtain a plurality of monospore areas; and carrying out area verification on the monospore area to obtain a screened target area meeting the condition, and taking the target area as the first separation processing result.
3. The method of claim 1, wherein the multi-feature verification of the first separation result comprises: performing circularity verification calculation on the first separation result, and determining that the first separation result is effective monospore under the condition that the calculation result meets the circularity condition; And/or the number of the groups of groups, And extracting spore edges in the first separation result by adopting a second edge detection algorithm, calculating the integrity of the spore edges, removing the first separation result with the integrity which does not meet the integrity condition, and obtaining the remaining first separation result as effective monospores.
4. The method of claim 1, wherein determining spore information from the multi-feature verification result comprises: Counting the number of effective monospores in the multi-feature verification result, and recording the center coordinate of each monospore; generating a monospore localization marker map based on the center coordinates, and taking the monospore localization marker map and the number of effective monospores as the spore information.
5. The method of claim 1, wherein the adaptively thresholding the first image comprises: constructing a color mask according to a preset spore color range, and screening to obtain a suspected spore area; Calculating a segmentation threshold value of a first image containing a suspected spore area through a preset self-adaptive algorithm to obtain an optimal segmentation threshold value, and converting the first image containing the suspected spore area into a binary image; and carrying out morphological open operation on the binary image based on the optimal segmentation threshold value to obtain the regional mask.
6. A spore capture analyzer, comprising: the image acquisition module is used for acquiring an original image of the spore; the pretreatment module is used for carrying out pretreatment on the original spore image so as to obtain a first image; The self-threshold segmentation module is used for carrying out self-adaptive threshold segmentation processing on the first image so as to obtain a region mask; The first processing module is used for carrying out first processing based on the region mask, wherein the first processing comprises the steps of carrying out communication region analysis processing on the region mask to determine overlapped lumps; And the verification module is used for carrying out multi-feature verification on the first separation result and determining spore information according to the multi-feature verification result.
7. The spore capture analyzer of claim 6, wherein the first separation process of the overlapping bolus by a predetermined first algorithm includes: carrying out Euclidean distance transformation on the overlapped block area to calculate the shortest distance from each pixel point in the overlapped block to a background pixel point, wherein the distance value is in direct proportion to the probability value of the core area of the spores; Threshold processing is carried out on Euclidean distance transformation results to obtain a plurality of core marks, wherein each mark corresponds to the center of one spore; morphological dilation of the core markers to ensure that each of the core markers covers a spore core region; watershed segmentation is carried out on the overlapped lumps based on the core mark so as to obtain a plurality of monospore areas; and carrying out area verification on the monospore area to obtain a screened target area meeting the condition, and taking the target area as the first separation processing result.
8. The spore capture analyzer of claim 6, wherein the multi-feature validation of the first separation result includes: performing circularity verification calculation on the first separation result, and determining that the first separation result is effective monospore under the condition that the calculation result meets the circularity condition; And/or the number of the groups of groups, And extracting spore edges in the first separation result by adopting a second edge detection algorithm, calculating the integrity of the spore edges, removing the first separation result with the integrity which does not meet the integrity condition, and obtaining the remaining first separation result as effective monospores.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program, wherein the computer program is arranged to execute the method of any of the claims 1 to 5 when run.
10. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of any of the claims 1 to 5.

Description

Spore capturing and identifying method and spore capturing analyzer Technical Field The embodiment of the invention relates to the field of spore capture and identification, in particular to a spore capture and identification method and a spore capture analyzer thereof. Background In the field of agricultural disease monitoring and early warning, spore capturing technology is an important pathogenic bacteria monitoring means, and early detection, prevention and control of crop diseases are realized by collecting and analyzing pathogenic spores in air. The traditional spore capturing method mainly relies on manual microscopic examination, has low efficiency and strong subjectivity, and is difficult to meet the requirements of large-scale and real-time monitoring. The existing spore identification method based on image analysis is low in image segmentation precision, and the problem of overlapping, blurring or uneven distribution of spores in the acquired images often occurs, so that the accuracy of subsequent identification is reduced. Disclosure of Invention The embodiment of the invention provides a spore capturing and identifying method and a spore capturing analyzer thereof, which are used for at least solving the problem of low image segmentation quality in the related technology. According to an embodiment of the present invention, there is provided a spore capturing and identifying method including: Acquiring an original image of spores; Preprocessing the original spore image to obtain a first image; Performing self-adaptive threshold segmentation processing on the first image to obtain a region mask; Performing a first process based on the region mask, the first process including performing a connected region analysis process on the region mask to determine overlapping clusters; And performing multi-feature verification on the first separation result, and determining spore information according to the multi-feature verification result. In an exemplary embodiment, the performing the first separation process on the overlapped block by a preset first algorithm includes: carrying out Euclidean distance transformation on the overlapped block area to calculate the shortest distance from each pixel point in the overlapped block to a background pixel point, wherein the distance value is in direct proportion to the probability value of the core area of the spores; Threshold processing is carried out on Euclidean distance transformation results to obtain a plurality of core marks, wherein each mark corresponds to the center of one spore; morphological dilation of the core markers to ensure that each of the core markers covers a spore core region; watershed segmentation is carried out on the overlapped lumps based on the core mark so as to obtain a plurality of monospore areas; and carrying out area verification on the monospore area to obtain a screened target area meeting the condition, and taking the target area as the first separation processing result. In an exemplary embodiment, the multi-feature verification of the first separation result includes: performing circularity verification calculation on the first separation result, and determining that the first separation result is effective monospore under the condition that the calculation result meets the circularity condition; And/or the number of the groups of groups, And extracting spore edges in the first separation result by adopting a second edge detection algorithm, calculating the integrity of the spore edges, removing the first separation result with the integrity which does not meet the integrity condition, and obtaining the remaining first separation result as effective monospores. In an exemplary embodiment, the determining spore information from the multi-feature verification result includes: Counting the number of effective monospores in the multi-feature verification result, and recording the center coordinate of each monospore; generating a monospore localization marker map based on the center coordinates, and taking the monospore localization marker map and the number of effective monospores as the spore information. In an exemplary embodiment, the performing the adaptive thresholding process on the first image includes: constructing a color mask according to a preset spore color range, and screening to obtain a suspected spore area; Calculating a segmentation threshold value of a first image containing a suspected spore area through a preset self-adaptive algorithm to obtain an optimal segmentation threshold value, and converting the first image containing the suspected spore area into a binary image; and carrying out morphological open operation on the binary image based on the optimal segmentation threshold value to obtain the regional mask. According to another embodiment of the present invention, there is provided a spore capture analyzer including: the image acquisition module is used for acquiring an original image of the spore; th