CN112652028B - Method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image - Google Patents
Method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image Download PDFInfo
- Publication number
- CN112652028B CN112652028B CN202110075606.1A CN202110075606A CN112652028B CN 112652028 B CN112652028 B CN 112652028B CN 202110075606 A CN202110075606 A CN 202110075606A CN 112652028 B CN112652028 B CN 112652028B
- Authority
- CN
- China
- Prior art keywords
- wood
- pine
- epidemic
- pixel
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 241000243771 Bursaphelenchus xylophilus Species 0.000 title claims abstract description 91
- 201000010099 disease Diseases 0.000 title claims abstract description 66
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 66
- 241000018646 Pinus brutia Species 0.000 title claims abstract description 48
- 235000008331 Pinus X rigitaeda Nutrition 0.000 title claims abstract description 47
- 235000011613 Pinus brutia Nutrition 0.000 title claims abstract description 47
- 241000196324 Embryophyta Species 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 37
- 239000002023 wood Substances 0.000 claims abstract description 170
- 238000000605 extraction Methods 0.000 claims abstract description 25
- 230000003595 spectral effect Effects 0.000 claims abstract description 22
- 238000001914 filtration Methods 0.000 claims abstract description 20
- 235000011609 Pinus massoniana Nutrition 0.000 claims description 20
- 241000018650 Pinus massoniana Species 0.000 claims description 20
- 235000011611 Pinus yunnanensis Nutrition 0.000 claims description 12
- 241000018652 Pinus yunnanensis Species 0.000 claims description 12
- 230000007797 corrosion Effects 0.000 claims description 10
- 238000005260 corrosion Methods 0.000 claims description 10
- 235000008585 Pinus thunbergii Nutrition 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 235000005205 Pinus Nutrition 0.000 claims description 5
- 241000218602 Pinus <genus> Species 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 5
- 238000003672 processing method Methods 0.000 claims description 4
- 206010027336 Menstruation delayed Diseases 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 241000218686 Pinus thunbergii Species 0.000 claims 1
- 241000607479 Yersinia pestis Species 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 10
- 230000009286 beneficial effect Effects 0.000 description 9
- 235000005638 Austrian pine Nutrition 0.000 description 8
- 241000238631 Hexapoda Species 0.000 description 8
- 235000008565 Pinus banksiana Nutrition 0.000 description 8
- 244000019397 Pinus jeffreyi Species 0.000 description 8
- 235000013264 Pinus jeffreyi Nutrition 0.000 description 8
- 235000008578 Pinus strobus Nutrition 0.000 description 8
- 235000014030 Podocarpus spicatus Nutrition 0.000 description 8
- 235000017985 rocky mountain lodgepole pine Nutrition 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 6
- 238000001228 spectrum Methods 0.000 description 6
- 238000011109 contamination Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 208000015181 infectious disease Diseases 0.000 description 3
- 206010035148 Plague Diseases 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000002310 reflectometry Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a method for extracting pine information of single plant infected with pine wood nematode disease based on RGB images, which comprises the following steps: s1: constructing a pine wood nematode epidemic wood remote sensing index based on R, G and a B wave band; s2: extracting a pine wood nematode epidemic wood pixel level sample according to a pine wood nematode epidemic wood remote sensing index; s3: filtering the pixel-level sample of the pine wood nematode disease epidemic; s4: and extracting single wood information of the pine wood nematode disease pixel level sample to finish the extraction of the single plant infected pine wood nematode disease pine wood information. According to the extraction method, only 3 most common visible light wave band images with minimum aerial remote sensing of the unmanned aerial vehicle are utilized, the extraction of vegetation indexes based on spectral features is carried out, and the spectral features and the high spatial resolution characteristics of the vegetation indexes are utilized to perform single plant epidemic wood identification and information extraction.
Description
Technical Field
The invention belongs to the technical field of pine wood with pine wood nematode disease, and particularly relates to a method for extracting pine wood information of single plant infected with pine wood nematode disease based on RGB images.
Background
The remote sensing technology has the observation characteristics of long distance, large area, no direct contact and the like, and is widely applied to the forest monitoring field. At present, a pine tree detection technology for pine wilting disease (pine wood nematode disease) based on a remote sensing technology is commonly used for automatically identifying infected pine trees by using multispectral and hyperspectral satellite images and combining typical spectral characteristics, and a proper algorithm or a manual spectral characteristic threshold setting technology is provided. However, such techniques suffer from the following drawbacks or limitations: (1) In the regions of China with cloudiness and raininess, because of the limitation of satellite orbit revisit period (which is different from a few days to a few tens of days), high-resolution and large-area satellite image data are difficult to acquire in time, and timeliness of the data cannot be ensured, so that the spread of epidemic diseases cannot be restrained in time. (2) The resolution of the satellite image is not high, the spatial resolution of the multi/hyperspectral image is different from 2 meters to hundreds of meters, and the positioning and qualitative requirements of the single plant of the affected pine can not be met, especially the single plant pine with the crown width of only sub-meter level can not be met.
Aiming at the defects and limitations, unmanned aerial vehicle aerial remote sensing technology can be adopted as a solution. The timeliness of pine tree detection of the pine wilt disease is effectively ensured by utilizing the capability of rapidly acquiring high-resolution (up to centimeter level) image data of a target area in real time. However, the unmanned aerial vehicle aerial remote sensing technology also has the self problems. When the unmanned aerial vehicle monitoring work is carried out by taking over the hyperspectral satellite image automatic identification technology (combining one or more typical spectral characteristics, providing a proper algorithm or manually setting a spectral characteristic threshold value and automatically identifying the affected pine), the problems of indefinite spectrum segments (from 700nm to thousands of nm, from 4 spectrum segments to hundreds of spectrum segments) and high instrument price of the multi/hyperspectral camera capable of being carried by the unmanned aerial vehicle are caused, so that the universal spectral characteristic extraction and the affected pine automatic identification method are difficult to provide.
Aiming at the problems, partial researchers propose to automatically identify the affected pine tree by using only 3 visible light wave band (R, G, B wave band) images which are most common in unmanned aerial vehicle aerial remote sensing. In the visible literature, there is an automatic recognition technology based on the concept of supervised classification, and the technology needs to collect a large amount of verified correct samples, divide the samples into at least two types, and set a rule set for recognizing features according to the statistical features of the extracted sample images, so as to realize automatic recognition. The problem of the technology is that firstly, the collection of the supervision sample is a complicated process requiring manual input, and the working efficiency is limited; secondly, the recognition accuracy depends on the quality of the supervision sample, and the quality of the sample cannot be generally ensured due to the complexity of the environment background and the limitation of sample representation, and the applicability of the supervision classification method is also affected.
Disclosure of Invention
The invention aims to solve the problem of extracting pine information of single plant infection pine wood nematode disease, and provides a method for extracting the pine information of single plant infection pine wood nematode disease based on RGB images.
The technical scheme of the invention is as follows: the method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image comprises the following steps:
s1: constructing a pine wood nematode epidemic wood remote sensing index based on R, G and a B wave band;
s2: extracting a pine wood nematode epidemic wood pixel level sample according to a pine wood nematode epidemic wood remote sensing index;
S3: filtering the pixel-level sample of the pine wood nematode disease epidemic;
s4: and extracting single wood information of the pixel-level sample of the pine wood nematode disease after the filtering treatment, and finishing the extraction of the single plant infected pine wood nematode disease pine wood information.
The beneficial effects of the invention are as follows:
(1) According to the extraction method, only 3 visible light wave band images with minimum aerial remote sensing of the unmanned aerial vehicle and most common are utilized, the vegetation index is extracted based on the spectral features of red light (R), green light (G) and blue light (B), and the single plant epidemic wood identification and information extraction are performed by utilizing the spectral features and the high spatial resolution characteristics of the vegetation index. On the premise of considering the geometric characteristics of the image of the crown, a complete technical scheme for automatically identifying the affected pine tree is provided by combining an image processing method. The method can provide more accurate information of the contamination or not and the spatial position for the identification of the contamination epidemic wood at the single plant level.
(2) In the research of the remote sensing monitoring field of forestry diseases and insect pests, the extraction method can simply, rapidly and accurately acquire the information of the individual plant epidemic wood, can effectively reduce the consumption and waste of time, manpower and economic cost in the traditional monitoring of the forestry diseases and insect pests, effectively suppress the spread of the forestry diseases and insect pests and the ecological damage and property loss caused by the spread of the forestry diseases and insect pests in time, and promote the ecological civilization construction.
Further, step S1 comprises the sub-steps of:
S11: for pine trees in different growth stages, measuring R, G and B wave band spectral emissivity corresponding to the pine tree in different growth stages by using a spectrometer;
s12: and determining a late index T 1 and an early index T 2 of the pine wood nematode disease wood according to R, G and B wave band spectral emissivity, and completing construction of a pine wood nematode disease wood remote sensing index based on R, G and B wave bands.
The beneficial effects of the above-mentioned further scheme are: in the invention, for different growth state stages (healthy, early, medium and late) of various pine (masson pine, yunnan pine and black pine), the corresponding R, G, B-band spectral reflectivity is measured by utilizing a spectrometer, and the statistical analysis of a large number of samples is carried out. According to the statistical result, 2 pine wood nematode epidemic remote sensing indexes based on R, G, B wave bands are established for distinguishing affected pine trees from healthy pine trees.
Further, in step S11, the growth stage of pine tree includes healthy, early, medium and late stages.
Further, in step S12, the calculation formula of the late index T 1 in pine wood nematode disease wood is:
T1＝DNR/DNB
wherein DN R represents an R-band image pixel value, and DN B represents a B-band image pixel value;
The calculation formula of the early index T 2 of pine wood nematode disease epidemic is as follows:
T2＝DNG/DNR
Wherein DN G represents a G-band image pixel value, and DN R represents an R-band image pixel value.
T 1 is the ratio of R band image pixel value to B band image pixel value, and T 2 is the ratio of G band image pixel value to R band image pixel value.
Further, step S2 comprises the sub-steps of:
S21: extracting a suspected epidemic wood sample and a suspected non-epidemic wood sample according to the late index T 1 in the pine wood nematode epidemic wood;
S22: and performing secondary judgment on the suspected non-epidemic wood sample according to the early index T 2 of the pine wood nematode epidemic wood to obtain a pine wood nematode epidemic wood pixel-level sample.
The beneficial effects of the above-mentioned further scheme are: in the invention, the middle and late stage epidemic woods of various pine trees (masson pine, yunnan pine and black pine) can be judged according to T 1, and the pixel information of the epidemic woods is extracted. The middle and late stages herein are those including just dying, one month of death and half year of death. After middle and late stage epidemic wood extraction is carried out by utilizing middle and late stage indexes of the pine wood nematode disease epidemic wood, judging the rest early stage epidemic wood and healthy multiple pine trees according to T 2, and extracting pixel level information of the epidemic wood.
Further, in step S21, the method for extracting the suspected epidemic wood sample and the suspected non-epidemic wood sample includes: if the intermediate and late period index T 1 of pine wood nematode epidemic wood of each pixel is greater than 2, the pixel is a suspected epidemic wood sample of Pinus yunnanensis, a suspected epidemic wood sample of Pinus kusnezoffii or a suspected epidemic wood sample of Pinus massoniana in an intermediate and late death state, otherwise, the pixel is a suspected non-epidemic wood sample;
In step S22, in the suspected non-epidemic wood sample, if the early index T 2 of pine wood nematode disease wood of each pixel is less than 1.1, the pixel is the suspected epidemic wood sample of masson pine in early half-dead state, otherwise, the pixel is the suspected non-epidemic wood sample;
And taking the suspected epidemic wood sample of the Yunnan pine, the suspected epidemic wood sample of the black pine, the suspected epidemic wood sample of the masson pine in a middle and late dead state and the suspected epidemic wood sample of the masson pine in an early half-dead state as pine wood nematode disease epidemic wood pixel-level samples.
The beneficial effects of the above-mentioned further scheme are: in the invention, in order to solve the problem of epidemic wood omission caused by confusing half-dead color-changing epidemic wood of pinus massoniana and healthy pinus wood, a suspected non-epidemic wood sample which does not meet the first judgment condition is subjected to second judgment.
Further, in step S3, the method for filtering the pixel-level sample of pine wood nematode disease is as follows: filtering false epidemic wood targets with abnormal pixels by using a sliding filter window with pixel resolution and epidemic wood crown size matched, wherein the calculation formula of the size W of the sliding filter window is as follows:
W＝n×n＝ρ2/0.25m2
where n represents the window length and ρ represents the resolution size.
The beneficial effects of the above-mentioned further scheme are: in the invention, the formed pine wood nematode suspected epidemic wood pixel level sample can introduce false epidemic wood targets caused by local noise and abnormal pixels. Aiming at the phenomenon, a sliding filter window matched with the empirical size of the epidemic wood crown according to the image resolution is provided to filter false epidemic wood targets with abnormal pixels. The resolution rho can be obtained from a header file or auxiliary information of the image, and the unit is m 2;0.25m2 which is the area formed by the conventional experience size of the epidemic wood crown; the window length n is in units of the number of pixels. The sliding step size is n/2 pixels. After filtering, two indices of outliers smaller than the sliding filter window size W are attenuated and two indices of outliers larger than the sliding filter window size are enhanced. The result of the operation is an integer.
Further, step S4 comprises the sub-steps of:
s41: constructing a circular structure with a pixel size r=n/2 as a radius, wherein n represents a window length;
s42: sequentially performing expansion treatment and corrosion treatment on the filtered pine wood nematode epidemic wood pixel level sample according to a graph closure operation by utilizing a circular structure;
s43: and discarding the false target after the graphic closed operation processing to finish the extraction of the single wood information.
The beneficial effects of the above-mentioned further scheme are: in the present invention, in the retained suspected epidemic target, there is a phenomenon that an individual plant of epidemic wood is regarded as a plurality of plants due to discontinuous distribution of pixels of the suspected epidemic wood. In addition, the form formed by the suspected individual plant epidemic wood pixels which are distributed in part continuously is obviously contrary to the actual form of the approximate circle of the individual plant epidemic wood crown. Aiming at the two phenomena, the technical scheme constructs a single wood information extraction method based on image morphology. A circular structure is constructed using r=n/2 pixels as radius. The single epidemic wood extracted based on the method can avoid false epidemic wood of other shapes from being extracted erroneously. And extracting the positions of the individual plant plague wood by using a closed operation. The extracted epidemic wood can be divided into a plurality of epidemic wood plants and is a single plant.
Further, in step S42, the expansion processing method includes: each pixel of the pine wood nematode disease pixel level sample is scanned and filtered by using a circular structure, structural elements of the circular structure and a binary image covered by the structural elements are processed and operated, if the result of the operation is 0, the pixel of the result image is 0, and otherwise, the pixel of the result image is 1;
the method for carrying out corrosion treatment comprises the following steps: and carrying out operation on structural elements of the circular structure and the covered binary image by using each pixel of the pine wood nematode disease pixel level sample after the circular structure scanning and filtering treatment, wherein if the operation result is 1, the pixel of the result image is 1, otherwise, the pixel of the result image is 0, and storing the result image with the pixel result of 1 as a closed operation result.
The beneficial effects of the above-mentioned further scheme are: in the invention, the image closing operation is a process of sequentially performing expansion and corrosion treatment on the image. The image expands and erodes, which helps to fill small voids in the object, connect adjacent objects, smooth their boundaries, and does not significantly change their area. Thus, the mistakes can be divided into epidemic woods of a plurality of plants and are single plants. Inflation is the process of incorporating all background points in contact with an object into the object, expanding the boundary outward. Can be used to fill voids in objects. Corrosion is a process of eliminating boundary points and causing the boundary to shrink inward. Can be used to eliminate small and meaningless objects.
Further, in step S43, pixels with a size smaller than (pi×r 2)/2 are discarded to complete extraction of the log information, where r represents a radius of the graphic structure.
The beneficial effects of the above-mentioned further scheme are: in the invention, because the closing operation brings about a few false targets with small size, the false targets with the size smaller than (pi multiplied by r 2)/2 pixels are finally discarded, and the false small targets are eliminated.
Drawings
FIG. 1 is a flow chart of a method of extracting pine information of individual plant infection pine wood nematode disease;
FIG. 2 is a diagram showing spectral characteristics of masson pine;
FIG. 3 is a spectral signature of Yunnan pine;
Fig. 4 is a graph of the spectral characteristics of black pine.
Detailed Description
Embodiments of the present invention are further described below with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image, comprising the following steps:
s1: constructing a pine wood nematode epidemic wood remote sensing index based on R, G and a B wave band;
s2: extracting a pine wood nematode epidemic wood pixel level sample according to a pine wood nematode epidemic wood remote sensing index;
S3: filtering the pixel-level sample of the pine wood nematode disease epidemic;
s4: and extracting single wood information of the pixel-level sample of the pine wood nematode disease after the filtering treatment, and finishing the extraction of the single plant infected pine wood nematode disease pine wood information.
In the embodiment of the present invention, as shown in fig. 1, step S1 includes the following sub-steps:
S11: for pine trees in different growth stages, measuring R, G and B wave band spectral emissivity corresponding to the pine tree in different growth stages by using a spectrometer;
s12: and determining a late index T 1 and an early index T 2 of the pine wood nematode disease wood according to R, G and B wave band spectral emissivity, and completing construction of a pine wood nematode disease wood remote sensing index based on R, G and B wave bands.
In the embodiment of the invention, as shown in fig. 2, a masson pine spectrum characteristic diagram is shown; as shown in fig. 3, a spectrum characteristic diagram of Yunnan pine is shown; as shown in fig. 4, a black pine spectrum characteristic diagram is shown.
In the invention, for different growth state stages (healthy, early, medium and late) of various pine (masson pine, yunnan pine and black pine), the corresponding R, G, B-band spectral reflectivity is measured by utilizing a spectrometer, and the statistical analysis of a large number of samples is carried out. According to the statistical result, 2 pine wood nematode epidemic remote sensing indexes based on R, G, B wave bands are established for distinguishing affected pine trees from healthy pine trees.
In the embodiment of the present invention, as shown in fig. 1, in step S11, the growth stage of pine includes healthy, early, medium and late stages.
In the embodiment of the present invention, as shown in fig. 1, in step S12, the calculation formula of the late index T 1 in pine wood nematode epidemic wood is as follows:
T1＝DNR/DNB
wherein DN R represents an R-band image pixel value, and DN B represents a B-band image pixel value;
The calculation formula of the early index T 2 of pine wood nematode disease epidemic is as follows:
T2＝DNG/DNR
Wherein DN G represents a G-band image pixel value, and DN R represents an R-band image pixel value.
In the present invention, T 1 is the ratio of the R-band image pixel value to the B-band image pixel value, and T 2 is the ratio of the G-band image pixel value to the R-band image pixel value.
In the embodiment of the present invention, as shown in fig. 1, step S2 includes the following sub-steps:
S21: extracting a suspected epidemic wood sample and a suspected non-epidemic wood sample according to the late index T 1 in the pine wood nematode epidemic wood;
S22: and performing secondary judgment on the suspected non-epidemic wood sample according to the early index T 2 of the pine wood nematode epidemic wood to obtain a pine wood nematode epidemic wood pixel-level sample.
In the invention, the middle and late stage epidemic woods of various pine trees (masson pine, yunnan pine and black pine) can be judged according to T 1, and the pixel information of the epidemic woods is extracted. The middle and late stages herein are those including just dying, one month of death and half year of death. After middle and late stage epidemic wood extraction is carried out by utilizing middle and late stage indexes of the pine wood nematode disease epidemic wood, judging the rest early stage epidemic wood and healthy multiple pine trees according to T 2, and extracting pixel level information of the epidemic wood.
In the embodiment of the present invention, as shown in fig. 1, in step S21, the method for extracting the suspected epidemic wood sample and the suspected non-epidemic wood sample includes: if the intermediate and late period index T 1 of pine wood nematode epidemic wood of each pixel is greater than 2, the pixel is a suspected epidemic wood sample of Pinus yunnanensis, a suspected epidemic wood sample of Pinus kusnezoffii or a suspected epidemic wood sample of Pinus massoniana in an intermediate and late death state, otherwise, the pixel is a suspected non-epidemic wood sample;
In step S22, in the suspected non-epidemic wood sample, if the early index T 2 of pine wood nematode disease wood of each pixel is less than 1.1, the pixel is the suspected epidemic wood sample of masson pine in early half-dead state, otherwise, the pixel is the suspected non-epidemic wood sample;
And taking the suspected epidemic wood sample of the Yunnan pine, the suspected epidemic wood sample of the black pine, the suspected epidemic wood sample of the masson pine in a middle and late dead state and the suspected epidemic wood sample of the masson pine in an early half-dead state as pine wood nematode disease epidemic wood pixel-level samples.
In the invention, in order to solve the problem of epidemic wood omission caused by confusing half-dead color-changing epidemic wood of pinus massoniana and healthy pinus wood, a suspected non-epidemic wood sample which does not meet the first judgment condition is subjected to second judgment.
In the embodiment of the present invention, as shown in fig. 1, in step S3, a method for filtering a pixel-level sample of pine wood nematode disease is as follows: filtering false epidemic wood targets with abnormal pixels by using a sliding filter window with pixel resolution and epidemic wood crown size matched, wherein the calculation formula of the size W of the sliding filter window is as follows:
W＝n×n＝ρ2/0.25m2
where n represents the window length and ρ represents the resolution size.
In the invention, the formed pine wood nematode suspected epidemic wood pixel level sample can introduce false epidemic wood targets caused by local noise and abnormal pixels. Aiming at the phenomenon, a sliding filter window matched with the empirical size of the epidemic wood crown according to the image resolution is provided to filter false epidemic wood targets with abnormal pixels. The resolution rho can be obtained from a header file or auxiliary information of the image, and the unit is m 2;0.25m2 which is the area formed by the conventional experience size of the epidemic wood crown; the window length n is in units of the number of pixels. The sliding step size is n/2 pixels. After filtering, two indices of outliers smaller than the sliding filter window size W are attenuated and two indices of outliers larger than the sliding filter window size are enhanced. The result of the operation is an integer.
In the embodiment of the present invention, as shown in fig. 1, step S4 includes the following substeps:
s41: constructing a circular structure with a pixel size r=n/2 as a radius, wherein n represents a window length;
s42: sequentially performing expansion treatment and corrosion treatment on the filtered pine wood nematode epidemic wood pixel level sample according to a graph closure operation by utilizing a circular structure;
s43: and discarding the false target after the graphic closed operation processing to finish the extraction of the single wood information.
In the present invention, in the retained suspected epidemic target, there is a phenomenon that an individual plant of epidemic wood is regarded as a plurality of plants due to discontinuous distribution of pixels of the suspected epidemic wood. In addition, the form formed by the suspected individual plant epidemic wood pixels which are distributed in part continuously is obviously contrary to the actual form of the approximate circle of the individual plant epidemic wood crown. Aiming at the two phenomena, the technical scheme constructs a single wood information extraction method based on image morphology. A circular structure is constructed using r=n/2 pixels as radius. The single epidemic wood extracted based on the method can avoid false epidemic wood of other shapes from being extracted erroneously. And extracting the positions of the individual plant plague wood by using a closed operation. The extracted epidemic wood can be divided into a plurality of epidemic wood plants and is a single plant.
In the embodiment of the present invention, as shown in fig. 1, in step S42, the method for performing the expansion process is as follows: each pixel of the pine wood nematode disease pixel level sample is scanned and filtered by using a circular structure, structural elements of the circular structure and a binary image covered by the structural elements are processed and operated, if the result of the operation is 0, the pixel of the result image is 0, and otherwise, the pixel of the result image is 1;
the method for carrying out corrosion treatment comprises the following steps: and carrying out operation on structural elements of the circular structure and the covered binary image by using each pixel of the pine wood nematode disease pixel level sample after the circular structure scanning and filtering treatment, wherein if the operation result is 1, the pixel of the result image is 1, otherwise, the pixel of the result image is 0, and storing the result image with the pixel result of 1 as a closed operation result.
In the invention, the image closing operation is a process of sequentially performing expansion and corrosion treatment on the image. The image expands and erodes, which helps to fill small voids in the object, connect adjacent objects, smooth their boundaries, and does not significantly change their area. Thus, the mistakes can be divided into epidemic woods of a plurality of plants and are single plants. Inflation is the process of incorporating all background points in contact with an object into the object, expanding the boundary outward. Can be used to fill voids in objects. Corrosion is a process of eliminating boundary points and causing the boundary to shrink inward. Can be used to eliminate small and meaningless objects.
In the embodiment of the present invention, as shown in fig. 1, in step S43, pixels with a size smaller than (pi×r 2)/2 are discarded to complete extraction of the single wood information, where r represents a radius of the graphic structure.
In the invention, because the closing operation brings about a few false targets with small size, the false targets with the size smaller than (pi multiplied by r 2)/2 pixels are finally discarded, and the false small targets are eliminated.
The working principle and the working process of the invention are as follows: firstly, constructing a pine wood nematode epidemic wood remote sensing index based on R, G and a B wave band; extracting a pine wood nematode epidemic wood pixel level sample according to a pine wood nematode epidemic wood remote sensing index; filtering the pixel-level sample of the pine wood nematode epidemic disease wood; and extracting single wood information of the pixel-level sample of the pine wood nematode disease after the filtering treatment, and finishing the extraction of the single plant infected pine wood nematode disease pine wood information.
The beneficial effects of the invention are as follows:
(1) According to the extraction method, only 3 visible light wave band images with minimum aerial remote sensing of the unmanned aerial vehicle and most common are utilized, the vegetation index is extracted based on the spectral features of red light (R), green light (G) and blue light (B), and the single plant epidemic wood identification and information extraction are performed by utilizing the spectral features and the high spatial resolution characteristics of the vegetation index. On the premise of considering the geometric characteristics of the image of the crown, a complete technical scheme for automatically identifying the affected pine tree is provided by combining an image processing method. The method can provide more accurate information of the contamination or not and the spatial position for the identification of the contamination epidemic wood at the single plant level.
(2) In the research of the remote sensing monitoring field of forestry diseases and insect pests, the extraction method can simply, rapidly and accurately acquire the information of the individual plant epidemic wood, can effectively reduce the consumption and waste of time, manpower and economic cost in the traditional monitoring of the forestry diseases and insect pests, effectively suppress the spread of the forestry diseases and insect pests and the ecological damage and property loss caused by the spread of the forestry diseases and insect pests in time, and promote the ecological civilization construction.
Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific modifications and combinations from the teachings of the present disclosure without departing from the spirit thereof, and such modifications and combinations remain within the scope of the present disclosure.
Claims (3)
1. The method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image is characterized by comprising the following steps:
s1: constructing a pine wood nematode epidemic wood remote sensing index based on R, G and a B wave band;
s2: extracting a pine wood nematode epidemic wood pixel level sample according to a pine wood nematode epidemic wood remote sensing index;
S3: filtering the pixel-level sample of the pine wood nematode disease epidemic;
S4: extracting single wood information of the pixel-level sample of the pine wood nematode disease after the filtering treatment, and finishing the extraction of the single plant infected pine wood nematode disease pine wood information;
said step S1 comprises the sub-steps of:
S11: for pine trees in different growth stages, measuring R, G and B wave band spectral emissivity corresponding to the pine tree in different growth stages by using a spectrometer;
S12: determining a late index T 1 and an early index T 2 of the pine wood nematode disease wood according to R, G and B wave band spectral emissivity, and completing construction of a pine wood nematode disease wood remote sensing index based on R, G and B wave bands;
In the step S12, the calculation formula of the intermediate and late stage index T 1 of pine wood nematode disease wood is as follows:
T1＝DNR/DNB
wherein DN R represents an R-band image pixel value, and DN B represents a B-band image pixel value;
The calculation formula of the early index T 2 of pine wood nematode disease epidemic is as follows:
T2＝DNG/DNR
wherein DN G represents a G-band image pixel value, and DN R represents an R-band image pixel value;
Said step S2 comprises the sub-steps of:
S21: extracting a suspected epidemic wood sample and a suspected non-epidemic wood sample according to the late index T 1 in the pine wood nematode epidemic wood;
S22: performing secondary judgment on the suspected non-epidemic wood sample according to the early index T 2 of the pine wood nematode epidemic wood to obtain a pine wood nematode epidemic wood pixel-level sample;
in the step S21, the method for extracting the suspected epidemic wood sample and the suspected non-epidemic wood sample comprises the following steps: if the intermediate and late period index T 1 of pine wood nematode epidemic wood of each pixel is greater than 2, the pixel is a suspected epidemic wood sample of Pinus yunnanensis, a suspected epidemic wood sample of Pinus kusnezoffii or a suspected epidemic wood sample of Pinus massoniana in an intermediate and late death state, otherwise, the pixel is a suspected non-epidemic wood sample;
In the step S22, in the suspected non-epidemic wood sample, if the early index T 2 of pine wood nematode disease wood is smaller than 1.1, the pixel is the suspected epidemic wood sample of masson pine in early half-dead state, otherwise is the suspected non-epidemic wood sample;
Taking a suspected epidemic wood sample of Pinus yunnanensis, a suspected epidemic wood sample of Pinus thunbergii, a suspected epidemic wood sample of Pinus massoniana in a middle and late dead state and a suspected epidemic wood sample of Pinus massoniana in an early half-dead state as pine wood nematode epidemic wood pixel-level samples;
Said step S4 comprises the sub-steps of:
s41: constructing a circular structure with a pixel size r=n/2 as a radius, wherein n represents a window length;
s42: sequentially performing expansion treatment and corrosion treatment on the filtered pine wood nematode epidemic wood pixel level sample according to a graph closure operation by utilizing a circular structure;
s43: discarding false targets after the graphic closed operation processing to finish the extraction of single wood information;
In the step S42, the expansion processing method includes: each pixel of the pine wood nematode disease pixel level sample is scanned and filtered by using a circular structure, structural elements of the circular structure and a binary image covered by the structural elements are processed and operated, if the result of the operation is 0, the pixel of the result image is 0, and otherwise, the pixel of the result image is 1;
The method for carrying out corrosion treatment comprises the following steps: each pixel of the pixel-level sample of the pine wood nematode disease after the circular structure scanning and filtering treatment is utilized, structural elements of the circular structure and a binary image covered by the structural elements are processed and operated, if the result of the operation is 1, the pixel of the result image is 1, otherwise, the pixel of the result image is 0, and a result image with the pixel result of 1 is stored and is used as a closed operation result;
In the step S43, pixels with a size smaller than (pi×r 2)/2 are discarded to complete extraction of the single wood information, where r represents a radius of the graphic structure.
2. The method according to claim 1, wherein the growing stage of pine tree comprises health, early stage, middle stage and late stage in step S11.
3. The method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image according to claim 1, wherein in the step S3, the method for filtering the pixel-level sample of pine wood nematode disease is as follows: filtering false epidemic wood targets with abnormal pixels by using a sliding filter window with pixel resolution and epidemic wood crown size matched, wherein the calculation formula of the size W of the sliding filter window is as follows:
W＝n×n＝ρ2/0.25m2
where n represents the window length and ρ represents the resolution size.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110075606.1A CN112652028B (en) | 2021-01-20 | 2021-01-20 | Method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110075606.1A CN112652028B (en) | 2021-01-20 | 2021-01-20 | Method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112652028A CN112652028A (en) | 2021-04-13 |
| CN112652028B true CN112652028B (en) | 2024-07-23 |
Family
ID=75371119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110075606.1A Active CN112652028B (en) | 2021-01-20 | 2021-01-20 | Method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112652028B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115115955B (en) * | 2022-07-08 | 2023-03-24 | 宁波大学 | Pine wood nematode disease tree monitoring method and device based on unmanned aerial vehicle hyperspectral remote sensing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764285A (en) * | 2018-04-23 | 2018-11-06 | 湖北同诚通用航空有限公司 | A kind of recognition methods of pine tree Deceased wood and system based on high resolution image |
| CN112016388A (en) * | 2020-07-08 | 2020-12-01 | 珠江水利委员会珠江水利科学研究院 | Vegetation information extraction method based on visible light waveband unmanned aerial vehicle remote sensing image |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7933451B2 (en) * | 2005-11-23 | 2011-04-26 | Leica Geosystems Ag | Feature extraction using pixel-level and object-level analysis |
| CN109978890B (en) * | 2019-02-25 | 2023-07-07 | 平安科技（深圳）有限公司 | Target extraction method and device based on image processing and terminal equipment |
| CN110096970A (en) * | 2019-04-12 | 2019-08-06 | 苏州市职业大学 | Pine forest discoloration standing tree single plant identification method based on WV3 satellite image |
| CN110779876B (en) * | 2019-11-07 | 2022-12-02 | 长光禹辰信息技术与装备(青岛)有限公司 | Method, device and equipment for identifying epidemic trees and computer readable storage medium |
| CN111767865A (en) * | 2020-06-30 | 2020-10-13 | 北部湾大学 | Method for inverting mangrove forest biomass by using aerial image and laser data |
| CN112053359B (en) * | 2020-09-30 | 2023-05-26 | 中国科学院空天信息创新研究院 | Remote sensing image change detection method and device, electronic equipment and storage medium |
-
2021
- 2021-01-20 CN CN202110075606.1A patent/CN112652028B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764285A (en) * | 2018-04-23 | 2018-11-06 | 湖北同诚通用航空有限公司 | A kind of recognition methods of pine tree Deceased wood and system based on high resolution image |
| CN112016388A (en) * | 2020-07-08 | 2020-12-01 | 珠江水利委员会珠江水利科学研究院 | Vegetation information extraction method based on visible light waveband unmanned aerial vehicle remote sensing image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112652028A (en) | 2021-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lacar et al. | Use of hyperspectral imagery for mapping grape varieties in the Barossa Valley, South Australia | |
| CN110796001B (en) | Satellite image film-covering farmland identification and extraction method and system | |
| CN112183209A (en) | Regional crop classification method and system based on multi-dimensional feature fusion | |
| CN107314990B (en) | Spring corn remote sensing identification method | |
| CN113505635A (en) | Method and device for identifying winter wheat and garlic mixed planting area based on optics and radar | |
| CN111007013B (en) | Crop rotation fallow remote sensing monitoring method and device for northeast cold region | |
| CN114199880A (en) | Citrus disease and insect pest real-time detection method based on edge calculation | |
| Izzuddin et al. | Analysis of airborne hyperspectral image using vegetation indices, red edge position and continuum removal for detection of Ganoderma disease in oil palm. | |
| CN105067532A (en) | Method for identifying early-stage disease spots of sclerotinia sclerotiorum and botrytis of rape | |
| CN112652028B (en) | Method for extracting pine information of single plant infected with pine wood nematode disease based on RGB image | |
| CN111199192A (en) | Method for detecting integral maturity of field red globe grapes by adopting parallel line sampling | |
| CN115456960A (en) | Citrus huanglongbing disease and pest monitoring and early warning system and method | |
| CN114612794A (en) | Remote sensing identification method for land covering and planting structure in finely-divided agricultural area | |
| CN113011354A (en) | Unmanned aerial vehicle hyperspectral image pine wood nematode disease identification method based on deep learning | |
| Lang et al. | Detection of chlorophyll content in maize canopy from UAV imagery | |
| CN111060455B (en) | Northeast cold-cool area oriented remote sensing image crop marking method and device | |
| CN114049564A (en) | Pine wood nematode disease grade prediction model construction method based on hyperspectral remote sensing image | |
| US20240212188A1 (en) | A system and method for spatial analysis of leaf images | |
| CN102184389B (en) | Automatic annual ring identification method used for tree species with high chromatic aberration of heartwood and sapwood | |
| TWI656334B (en) | A system for early detection of orchid pest by hyperspectral imaging techniques | |
| CN116721385A (en) | Machine learning-based RGB camera data cyanobacteria bloom monitoring method | |
| Pan et al. | Dynamic analysis of early stage pine wilt disease in Pinus massoniana using ground-level hyperspectral imaging | |
| Barjaktarovic et al. | Potential detection of Flavescence dorée in the vineyard using close-range hyperspectral imaging | |
| Poblete-Echeverria et al. | Detection of the spectral signature of Phytophthora root rot (PRR) symptoms using hyperspectral imaging | |
| Wang et al. | Extraction of Jujube Planting Areas in Sentinel-2 Image Based on NDVI Threshold—A case study of Ruoqiang County |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |