CN117636185B - Pine wood nematode disease detecting system based on image processing - Google Patents
Pine wood nematode disease detecting system based on image processing Download PDFInfo
- Publication number
- CN117636185B CN117636185B CN202410107688.7A CN202410107688A CN117636185B CN 117636185 B CN117636185 B CN 117636185B CN 202410107688 A CN202410107688 A CN 202410107688A CN 117636185 B CN117636185 B CN 117636185B
- Authority
- CN
- China
- Prior art keywords
- area
- image
- pest
- sub
- early warning
- 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 66
- 201000010099 disease Diseases 0.000 title claims abstract description 66
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 66
- 241000607479 Yersinia pestis Species 0.000 claims abstract description 144
- 238000001514 detection method Methods 0.000 claims abstract description 112
- 238000012216 screening Methods 0.000 claims abstract description 33
- 238000004364 calculation method Methods 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims description 120
- 239000011347 resin Substances 0.000 claims description 86
- 229920005989 resin Polymers 0.000 claims description 86
- 235000008331 Pinus X rigitaeda Nutrition 0.000 claims description 66
- 235000011613 Pinus brutia Nutrition 0.000 claims description 66
- 241000018646 Pinus brutia Species 0.000 claims description 66
- 241000238631 Hexapoda Species 0.000 claims description 28
- 238000004659 sterilization and disinfection Methods 0.000 claims description 12
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 239000002023 wood Substances 0.000 description 4
- 239000003086 colorant Substances 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000003902 lesion Effects 0.000 description 2
- 241000243770 Bursaphelenchus Species 0.000 description 1
- 241000244206 Nematoda Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- QDOXWKRWXJOMAK-UHFFFAOYSA-N dichromium trioxide Chemical compound O=[Cr]O[Cr]=O QDOXWKRWXJOMAK-UHFFFAOYSA-N 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
- Y02A50/30—Against vector-borne diseases, e.g. mosquito-borne, fly-borne, tick-borne or waterborne diseases whose impact is exacerbated by climate change
Abstract
The invention relates to the technical field of image processing, and discloses a pine wood nematode disease detection system based on image processing; obtaining a target image based on screening criteria, obtaining a target area in the target image, collecting comprehensive pest data of the target area, generating a pest detection index, judging whether to generate a pest early warning prompt based on a pest detection difference value, generating a pest early warning level based on the pest detection difference value, and formulating a killing instruction based on the pest early warning level; according to the invention, through screening and identifying the satellite remote sensing image, the non-target image with interference factors can be removed, and the target area with small area and accurate position is identified based on the target image, so that on one hand, the acquisition amount and the calculation amount of detection data are effectively reduced, the data calculation rate is improved, and on the other hand, the error influence possibly caused by useless data on the detection result is avoided, and the accuracy of detecting the pine wood nematode disease is greatly improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a pine wood nematode disease detection system based on image processing.
Background
Pine wood nematode disease is a forest disease caused by pine wood nematodes and has strong destructiveness, a large number of pine wood diseases are bred and spread in multiple provinces in China in recent years, so that a large number of pine wood diseases are dead, ecological balance is seriously destroyed, and in order to effectively protect healthy growth of pine wood, the pine wood nematode disease needs to be timely and accurately detected, so that early warning prompt is timely made when the pine wood nematode disease occurs, and for this reason, analysis and detection of the pine wood nematode disease need to be carried out by means of a pine wood remote sensing image.
The Chinese patent with the application publication number of CN115841492A discloses a pine wood nematode lesion color standing tree remote sensing intelligent identification method based on cloud edge cooperation, wherein a plurality of neighborhood enhancement ranges calculate new component values of pixel points, an enhancement chart corresponding to a collected pine tree image is obtained, the new component values calculated by the neighborhood enhancement ranges can weaken the mutual restriction between coverage and image resolution, the identification problem caused by overlarge or undersize resolution is avoided, the enhancement chart enlarges the saliency of pixel point image information in a target area, the extracted image features can more accurately correspond to a color-changing standing tree area in the collected image, and the nematode lesion area identification of the pine tree remote sensing image is completed more reliably;
The prior art has the following defects:
when the existing pine wood nematode disease detection system detects a pine wood nematode disease remote sensing image, massive data contained in the remote sensing image are required to be collected and detected, the calculation amount of subsequent data is large due to the fact that the data contained in the remote sensing image is large, the calculation efficiency is reduced, meanwhile, a large amount of useless data exists in the remote sensing image, when the excessive useless data participates in detection calculation, the accuracy of a detection result is easily affected negatively, the accuracy of the detection result is further reduced, and practical detection application of forestry personnel to the pine wood nematode disease is not facilitated.
In view of the above, the present invention proposes a pine wood nematode disease detection system based on image processing to solve the above-mentioned problems.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides the following technical scheme for achieving the purposes: the pine wood nematode disease detection system based on image processing is applied to an image processing server and comprises:
target image screening module for obtainingThe satellite remote sensing images are based on screening criteria, for ∈K>Screening the satellite remote sensing images to obtain a target image;
target objectRegion identification module for marking target image Individual zone demarcation point, will->After the area demarcation points are connected in sequence, a target area is obtained;
the data acquisition module acquires comprehensive pest data of the target area and generates a pest detection index based on the comprehensive pest data;
the early warning prompt module is used for comparing the pest detection index with a preset pest detection threshold value, generating a pest detection difference value and judging whether pest early warning prompt is generated or not;
the level dividing module is used for generating pest early warning levels based on the pest detection difference value;
and the instruction making module is used for making a killing instruction based on the insect pest early warning level.
Further, the screening criteria are: the actual resolution of the image is greater than 90% of the rated resolution, and the ratio of the pixel points of the pine tree is greater than the ratio of the lowest pixel points;
the target image acquisition method comprises the following steps:
acquisition of selected pine areas by means of an image databaseA satellite remote sensing image;
obtaining by looking at image propertiesThe actual resolution of each satellite remote sensing image is based on a screening criterion, and the satellite remote sensing image with the actual resolution being more than 90% of the rated resolution is marked as a marked image;
identifying the region of the pine tree in the marked image by a computer vision technology, and counting the number of pixels in the region of the pine tree;
Comparing the number of the pixels in the area where the pine tree is positioned with the total number of the pixels in the marked image to obtain the ratio of the pixels in the pine tree;
based on the screening criterion, screening out that the pixel point occupation ratio of pine is larger than the lowest pixel point occupation ratioMarking images->Less than->；
From the slaveAnd selecting a mark image corresponding to the maximum value of the pixel point occupation ratio of the pine from the mark images, namely the target image.
Further, the method for acquiring the target area comprises the following steps:
scanning the target image to obtain a scanning gray image, and labeling the pixel value of each pixel point in the scanning gray image;
marking the positions of pixel points with pixel values larger than a preset pixel threshold value to obtain effective positions;
marking along boundary lines of effective positions according to preset lengthsPersonal location, get->A plurality of zone demarcation points;
will be clockwiseAnd after the area demarcation points are connected in sequence, obtaining the target area.
Further, the comprehensive pest data comprises a healthy area occupation ratio, crack dryness and resin distribution density;
the method for acquiring the health area occupation ratio comprises the following steps:
scanning a target area to obtain a crown scanning image;
marking pixel values of all pixel points in the crown scanning image, and counting the total amount of the pixel points;
Comparing the pixel values of all the pixel points with a preset lower limit value and a preset upper limit value one by one;
the pixel points with the pixel values between the preset lower limit value and the preset upper limit value are marked as effective pixel points, and the number of the effective pixel points is counted;
comparing the number of the effective pixel points with the total number of the pixel points to obtain a healthy area occupation ratio;
the expression of the healthy area ratio is:
；
in the method, in the process of the invention,for the ratio of healthy area, ++>For the number of effective pixels, +.>Is the total number of pixels.
Further, the method for acquiring the dryness of the crack comprises the following steps:
scanning the target area to obtain a trunk scanning image, and equally dividing the trunk scanning image into parts according to a preset heightA sub-image;
identification by computer vision techniquesBark texture information on the sub-images and drawing lines along the positions of bark textures to obtain texture lines;
sequentially measuring horizontal distance values between vertical direction extension lines of two adjacent horizontal distribution texture lines along the horizontal direction, and marking the two adjacent texture lines with the horizontal distance values larger than a preset distance threshold value as target texture linesObtainingA group target texture line;
are positioned at the position of the measuring device in the vertical directionThe vertical distance value between two adjacent vertical distribution texture line endpoints in the group target texture line is horizontally drawn by taking the texture line endpoint where the minimum value of the vertical distance value is positioned as a base point, and is matched with +. >Group target texture line contact, get +.>A plurality of fracture zones;
one-to-one measurementThe length of each crack region and the height of each crack region are calculated by an area formula to obtain +.>The area of each crack;
the expression of the crack area is:
；
in the method, in the process of the invention,is->Sub-picture +.>Crack area of the individual crack region, +.>Is->Sub-picture +.>Length of the crack region->Is->Sub-picture +.>The height of the individual fracture zones;
will beSub-picture +.>After the area accumulation of the individual cracks, obtain +.>Sub-crack area;
the expression of the sub-crack area is:
；
in the method, in the process of the invention,is->Sub-slit area of sub-image +.>Is->No. of sub-image>The area of each crack;
measuring the length of the trunk scanning image and the height of the trunk scanning image, and obtaining the trunk scanning image area through an area calculation formula;
the expression of the trunk scan image area is:
；
in the method, in the process of the invention,scanning the image area for the trunk>Scanning the length of the image for the trunk, < >>Scanning the height of the image for the trunk;
removingMaximum and minimum in the sub-crack area, the remaining +.>After the sub-crack areas are accumulated, comparing the sub-crack areas with the trunk scanning image area to obtain crack dryness;
The expression of the crack dryness is:
；
in the method, in the process of the invention,for crack dryness>Is->Sub-slit area.
Further, the method for obtaining the distribution density of the resin comprises the following steps:
acquiring pixel values of all pixel points in the trunk scanning image, and comparing the pixel values of all pixel points with a preset first resin threshold value and a preset second resin threshold value;
the pixel points with the pixel values between the preset first resin threshold value and the preset second resin threshold value are marked as resin pixel points, and the obtained pixel points are obtainedA set of resin pixels;
edge of the frameThe boundary drawing line of the set of the resin pixel points minimizes and surrounds the resin pixel points to obtain +.>The areas to be identified;
identification by computer vision techniquesMarking the region to be identified, where bark texture information does not appear, as a resin region to obtain +.>Resin region->Less than->；
Measuring separatelyThe horizontal length and vertical height of the resin region, and based on the maximum value of the horizontal length and the maximum value of the vertical height, obtaining +.>A ratio of length to height;
the expression of the long-to-high ratio is:
；
in the method, in the process of the invention,is->Length to height ratio of individual resin areas, +.>Is->Maximum value of horizontal length of individual resin region, +. >Is->Maximum vertical height of each resin region;
the method comprises the steps of marking a resin area with a length-height ratio smaller than a preset length-height threshold value as a first distribution area, and marking a resin area with a length-height ratio larger than or equal to the preset length-height threshold value as a second distribution area;
drawing a rectangle on the boundary line of the first distribution area, measuring the length and width of the rectangle, and respectively recording the length and the height of the first distribution area to obtainArea of the first distribution area, +.>Less than->；
The area of the first distribution area is expressed as:
；
in the method, in the process of the invention,is->Area of the first distribution area, +.>Is->The length of the first distribution area is,is->The height of the first distribution areas;
will beAfter the areas of the first distribution areas are accumulated, a first sub-area is obtained;
the expression for the first sub-area is:
；
in the method, in the process of the invention,for the first sub-area->Is->The area of the first distribution area;
the boundary line of the second distribution area is drawn to obtain a circle, and the radius of the circle is measured and recorded as a second distribution areaRadius of domain, obtainArea of the second distribution area +.>Less than->；
The area of the second distribution area is expressed as:
；
in the method, in the process of the invention, Is->Area of the second distribution area +.>Is of circumference rate>Is->Radius of the second distribution area;
will beAfter the areas of the second distribution areas are accumulated, a second sub-area is obtained;
the expression for the second sub-area is:
；
in the method, in the process of the invention,for the second sub-area->Is->The area of the second distribution area;
after the first sub-area and the second sub-area are added, comparing the first sub-area and the second sub-area with the trunk scanning image area to obtain the resin distribution density;
the expression of the resin distribution density is:
；
in the method, in the process of the invention,is the resin distribution density.
Further, the expression of the pest detection index is:
；
in the method, in the process of the invention,for pest detection index, & lt & gt>、/>、/>Is a weight factor.
Further, the expression of the pest detection difference is:
；
in the method, in the process of the invention,for pest detection difference, & gt>Is a preset pest detection threshold;
the judging method for generating the pest early warning prompt comprises the following steps:
when (when)When the insect pest early warning prompt is larger than or equal to 0, judging to generate the insect pest early warning prompt;
when (when)And when the value is smaller than 0, judging that the insect pest early warning prompt is not generated.
Further, the insect pest early warning level comprises a first early warning level and a second early warning level;
the generation method of the first-level early warning level and the second-level early warning level comprises the following steps:
difference value of pest detection And a preset level threshold->Comparison of (I)>Greater than 0;
when (when)Less than->Generating a first-level early warning level;
when (when)Greater than or equal to->And generating a second-level early warning level.
Further, the killing instruction comprises a primary killing instruction and a secondary killing instruction;
the method for formulating the primary disinfection instruction and the secondary disinfection instruction comprises the following steps:
when the insect pest early warning level is a first early warning level, a first-level killing instruction is formulated;
when the pest early warning level is the secondary early warning level, a secondary killing instruction is formulated.
The pine wood nematode disease detection system based on image processing has the technical effects and advantages that:
the invention is realized by obtainingThe satellite remote sensing images are based on screening criteria, for ∈K>Screening individual satellite remote sensing images to obtain a target image, marking area demarcation points in the target image, sequentially connecting the area demarcation points to obtain a target area, collecting comprehensive pest data of the target area, generating a pest detection index based on the comprehensive pest data, comparing the pest detection index with a preset pest detection threshold value to generate a pest detection difference value, judging whether to generate a pest early warning prompt, generating a pest early warning level based on the pest detection difference value, and formulating a killing instruction based on the pest early warning level; compared with the prior art, the method has the advantages that through screening and identifying the satellite remote sensing images, the non-target images with interference factors can be removed, the target areas with small areas and accurate positions can be identified based on the target images, and whether pine wood nematode diseases occur or not can be accurately detected through the target areas, so that on one hand, the collection amount and the calculation amount of detection data are effectively reduced, the data calculation rate is improved, on the other hand, the error influence possibly caused by useless data on detection results is avoided, and the accuracy of pine wood nematode disease detection is greatly improved.
Drawings
Fig. 1 is a schematic diagram of a pine wood nematode disease detection system based on image processing provided in embodiment 1 of the present invention;
fig. 2 is a flow chart of a pine wood nematode disease detection method based on image processing provided in embodiment 2 of the present invention;
FIG. 3 is a schematic view of a target area according to embodiment 1 of the present invention;
fig. 4 is a schematic view of a crack region provided in example 1 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: referring to fig. 1, the pine wood nematode disease detection system based on image processing in this embodiment is applied to an image processing server, and includes:
target image screening module for obtainingThe satellite remote sensing images are based on screening criteria, for ∈K>Screening the satellite remote sensing images to obtain a target image;
The screening criterion refers to whether the satellite remote sensing image meets the standard of the use requirement, and the satellite remote sensing image is subjected to the influence of factors such as weather fluctuation environment, shooting angle change and the like, so that the satellite remote sensing image is blurred, the shot object is incomplete and the like, and therefore the satellite remote sensing image is screened by the aid of the screening criterion to obtain the image meeting the use requirement;
the screening criteria are: the actual resolution of the image is greater than 90% of the rated resolution, and the ratio of the pixel points of the pine tree is greater than the ratio of the lowest pixel points;
the nominal resolution is the nominal resolution of a shooting camera on a satellite, and is acquired through a satellite control system, and as the shooting camera on the satellite can be subjected to interference factors when shooting satellite remote sensing images, 90% of the nominal resolution needs to be set as the lowest resolution in order to reject images with high interference degrees; the minimum pixel occupation ratio is the minimum pixel occupation ratio of an image occupied by pine in the satellite remote sensing image, and the follow-up identification and detection of pine wood nematode disease can be met only when the pixel occupation ratio of the image occupied by pine reaches the minimum pixel occupation ratio;
The target image acquisition method comprises the following steps:
acquisition of selected pine areas by means of an image databaseA satellite remote sensing image;
obtaining by looking at image propertiesThe actual resolution of each satellite remote sensing image is based on a screening criterion, and the satellite remote sensing image with the actual resolution being more than 90% of the rated resolution is marked as a marked image;
identifying the region of the pine tree in the marked image by a computer vision technology, and counting the number of pixels in the region of the pine tree;
comparing the number of the pixels in the area where the pine tree is positioned with the total number of the pixels in the marked image to obtain the ratio of the pixels in the pine tree;
based on the screening criterion, screening out that the pixel point occupation ratio of pine is larger than the lowest pixel point occupation ratioMarking images->Less than->；
From the slaveSelecting a mark image corresponding to the maximum value of the pixel point occupation ratio of the pine from the mark images, namely, a target image;
target area recognition module for marking a target imageIndividual zone demarcation point, will->After the area demarcation points are connected in sequence, a target area is obtained;
the target image can be divided into a pine area and a background area according to different contents in the image, the pine area is a corresponding area for providing pine wood nematode disease detection data, the background area is a corresponding area incapable of effectively providing pine wood nematode disease detection data, and the background area in the target image is not required to participate in actual pine wood nematode disease detection, so that the background area is removed to obtain the target area, and the pine tree area is the target area;
The acquisition method of the target area comprises the following steps:
scanning the target image to obtain a scanning gray image, and labeling the pixel value of each pixel point in the scanning gray image;
marking the positions of pixel points with pixel values larger than a preset pixel threshold value to obtain effective positions; the preset pixel threshold is used for distinguishing the pixel value of the pine area pixel point from the pixel value of the background area pixel point, and as the pine area is the pine, the pine is usually green, yellow and other colors, and the background color is usually gray, white and other colors, the pixel value of the pine area pixel point is different from the pixel value of the background area pixel point, so that the pine area and the background area can be effectively distinguished; the preset pixel threshold value is obtained by acquiring a large number of historical pine area pixel values and then optimizing the coefficient;
marking along boundary lines of effective positions according to preset lengthsPersonal location, get->A plurality of zone demarcation points; the preset length is used for limiting the interval length of the area demarcation points, when the interval length of the area demarcation points is too large, the continuous span of two adjacent area demarcation points is larger, and at the moment, part of pine areas or part of background areas are drawn in, so that the target area is inaccurate; the preset length is obtained by acquiring the connecting line length of two adjacent area demarcation points corresponding to a large number of historical target areas and then taking the minimum value of the connecting line length;
Will be clockwiseAfter the area demarcation points are connected in sequence, a target area is obtained;
referring to fig. 3, the target area shown in fig. 3 is obtained by the above-mentioned target area obtaining method, in which A1, A2, A3, A4, A5, A6, A7, A8, A9, a10, a11, a12, a13, a14, a15 are all area demarcation points;
by setting the area demarcation point, the area of the target area can be reduced, the data acquisition and calculation amount of the target area can be reduced, the calculation efficiency can be improved, and the influence of useless data of the background area on the subsequent detection of the pine wood nematode disease can be avoided, so that the accuracy of the detection result of the pine wood nematode disease can be improved;
the data acquisition module acquires comprehensive pest data of the target area and generates a pest detection index based on the comprehensive pest data;
the comprehensive insect pest data refer to the pathological change data of the pine tree when the pine tree has pine wood nematode disease, and the comprehensive insect pest data can be collected to comprehensively and accurately know whether the pine tree has pine wood nematode disease and the severity of the pine wood nematode disease, so that data support is provided for subsequent pine tree treatment;
The comprehensive pest data comprise a healthy area occupation ratio, crack dryness and resin distribution density;
the healthy area occupation ratio refers to the ratio of the area occupied by the green leaves in the corresponding pine tree on the crown in the target area to the total area of the target area, and when the healthy area occupation ratio is larger, the more the number of the green leaves on the pine tree is, the larger the area occupied by the green leaves is, the lower the severity of the pine wood nematode disease on the pine tree is, and the pest detection index is smaller; the opposite is the case;
the method for acquiring the health area occupation ratio comprises the following steps:
scanning a target area to obtain a crown scanning image;
marking pixel values of all pixel points in the crown scanning image, and counting the total amount of the pixel points;
comparing the pixel values of all the pixel points with a preset lower limit value and a preset upper limit value one by one; the preset lower limit value and the preset upper limit value are numerical value ranges used for representing pixel values corresponding to pine tree green leaves, and the green of the pine tree leaves is not standard green and is divided into dark green and light green, so that the pixel values corresponding to the pine tree green leaves are different in size, a numerical value range exists, and the numerical values corresponding to the preset lower limit value and the preset upper limit value are obtained through coefficient optimization after collecting a large number of pixel values corresponding to different pine tree leaf green degrees of histories;
The pixel points with the pixel values between the preset lower limit value and the preset upper limit value are marked as effective pixel points, and the number of the effective pixel points is counted;
comparing the number of the effective pixel points with the total number of the pixel points to obtain a healthy area occupation ratio;
the expression of the healthy area ratio is:
；
in the method, in the process of the invention,for the ratio of healthy area, ++>For the number of effective pixels, +.>Is the total pixel point amount;
the crack dryness refers to the severity of cracks on the trunk of the pine tree in the target area due to the withered pine wood nematode disease, and can be expressed by the ratio of the area of the cracks on the trunk of the pine tree to the trunk area, when the crack dryness is larger, the severity of the withered phenomenon on the trunk of the pine tree due to the pine wood nematode disease is shown to be larger, and the pest detection index is larger; the opposite is the case;
the method for acquiring the crack dryness comprises the following steps:
scanning the target area to obtain a trunk scanning image, and equally dividing the trunk scanning image into parts according to a preset heightA sub-image; the preset height is a numerical value used for carrying out height segmentation on the trunk scanning image, and the trunk scanning image is segmented according to the preset height, so that the segmented sub-image can contain at least two pieces of complete bark texture information, the subsequent acquisition of texture lines is convenient, and the preset height is less than or equal to one fifth of the height of the trunk scanning image;
Identification by computer vision techniquesBark texture information on the sub-images and drawing lines along the positions of bark textures to obtain texture lines;
sequentially measuring horizontal distance values between vertical direction extension lines of two adjacent horizontal distribution texture lines along the horizontal direction, and marking the two adjacent texture lines with the horizontal distance values larger than a preset distance threshold value as target texture lines to obtainA group target texture line; the preset distance threshold is the basis for distinguishing whether the distance between two adjacent texture lines is in a normal range, whenWhen the distance value between two adjacent texture lines exceeds a preset distance threshold value, the phenomenon that bark cracks are missing between the two adjacent texture lines is indicated; the preset distance threshold is obtained by acquiring the distances between two adjacent texture lines corresponding to the phenomenon that a great number of bark cracks are missing and optimizing the coefficients;
are positioned at the position of the measuring device in the vertical directionThe vertical distance value between two adjacent vertical distribution texture line endpoints in the group target texture line is horizontally drawn by taking the texture line endpoint where the minimum value of the vertical distance value is positioned as a base point, and is matched with +.>Group target texture line contact, get +.>A plurality of fracture zones;
One-to-one measurementThe length of each crack region and the height of each crack region are calculated by an area formula to obtain +.>The area of each crack;
the expression of the crack area is:
；
in the method, in the process of the invention,is->Sub-picture +.>Crack area of the individual crack region, +.>Is->Sub-picture +.>Length of the crack region->Is->Sub-picture +.>The height of the individual fracture zones;
will beSub-picture +.>After the area accumulation of the individual cracks, obtain +.>Sub-crack area;
the expression of the sub-crack area is:
；
in the method, in the process of the invention,is->Sub-slit area of sub-image +.>Is->No. of sub-image>The area of each crack;
measuring the length of the trunk scanning image and the height of the trunk scanning image, and obtaining the trunk scanning image area through an area calculation formula;
the expression of the trunk scan image area is:
；
in the method, in the process of the invention,scanning the image area for the trunk>Scanning the length of the image for the trunk, < >>Scanning the height of the image for the trunk;
removingMaximum and minimum in the sub-crack area, the remaining +.>After the sub-crack areas are accumulated, comparing the sub-crack areas with the trunk scanning image area to obtain crack dryness;
the expression of the crack dryness is:
；
In the method, in the process of the invention,for crack dryness>Is->Sub-crack area;
referring to fig. 4, exemplary, the crack region shown in fig. 4 is obtained by the above-mentioned method for obtaining the crack region, where O1 and O2 are both base points, MB1 and MB2 are both target texture lines, and YC is a vertical extension line of the target texture lines;
the resin distribution density refers to resin effluent secreted by pine wood nematode disease on the trunk, when the resin just flows out, the resin is generally nearly transparent, after a period of time, the resin is yellow and is adsorbed on the trunk, when the resin distribution density is higher, the higher the severity of the pine wood nematode disease on the trunk is, the higher the pest detection index is; the opposite is the case;
the method for obtaining the resin distribution density comprises the following steps:
acquiring pixel values of all pixel points in the trunk scanning image, and comparing the pixel values of all pixel points with a preset first resin threshold value and a preset second resin threshold value; the preset first resin threshold and the preset second resin threshold are used for distinguishing whether the pixel value corresponding to the pixel point is in the pixel value range corresponding to the resin color, the preset first resin threshold and the preset second resin threshold are respectively the minimum value and the maximum value of the pixel value of the resin color, so that whether the pixel point on the trunk belongs to the pixel point of the resin is judged, and the preset first resin threshold and the preset second resin threshold are obtained through coefficient optimization after collecting the pixel values corresponding to a large number of resin colors of histories;
The pixel points with the pixel values between the preset first resin threshold value and the preset second resin threshold value are marked as resin pixel points, and the obtained pixel points are obtainedA set of resin pixels;
edge of the frameThe boundary drawing line of the set of the resin pixel points minimizes and surrounds the resin pixel points to obtain +.>The areas to be identified; the method has the advantages that the region position of the resin pixel point set can be acquired relatively accurately in a minimized surrounding mode, the trunk part is prevented from being drawn into the region to be identified, and the accuracy of the region to be identified is further improved;
identification by computer vision techniquesMarking the region to be identified, where bark texture information does not appear, as a resin region to obtain +.>Resin region->Less than->The method comprises the steps of carrying out a first treatment on the surface of the The bark area to be identified is identified and screened out by carrying out secondary identification of bark texture information on the area to be identified, so that the bark with the color similar to that of the resin is prevented from being mistakenly identified as the resin, and the accuracy of resin identification is further improved;
measuring separatelyThe horizontal length and vertical height of the resin region, and based on the maximum value of the horizontal length and the maximum value of the vertical height, obtaining +. >A ratio of length to height;
the expression of the long-to-high ratio is:
；
in the method, in the process of the invention,is->Length to height ratio of individual resin areas, +.>Is->Maximum value of horizontal length of individual resin region, +.>Is->Maximum vertical height of each resin region;
the method comprises the steps of marking a resin area with a length-height ratio smaller than a preset length-height threshold value as a first distribution area, and marking a resin area with a length-height ratio larger than or equal to the preset length-height threshold value as a second distribution area; the preset long-high threshold value is used for judging whether the shape of the resin area is similar to a rectangle or a circle, when the shape of the resin area is similar to the rectangle, a calculation mode of rectangular area is adopted, otherwise, a calculation mode of circular area is adopted, and the preset long-high threshold value is obtained through coefficient optimization after a large number of lengths and heights of the resin area, which correspond to the calculation mode of rectangular area and the calculation mode of circular area, are acquired;
drawing a rectangle on the boundary line of the first distribution area, measuring the length and width of the rectangle, and respectively recording the length and the height of the first distribution area to obtainArea of the first distribution area, +.>Less than->；
The area of the first distribution area is expressed as:
；
In the method, in the process of the invention,is->Area of the first distribution area, +.>Is->The length of the first distribution area is,is->The height of the first distribution areas;
will beAfter the areas of the first distribution areas are accumulated, a first sub-area is obtained;
the expression for the first sub-area is:
；
in the method, in the process of the invention,for the first sub-area->Is->The area of the first distribution area;
drawing a circle on the boundary line of the second distribution area, measuring the radius of the circle, and recording as the radius of the second distribution area to obtainArea of the second distribution area +.>Less than->；
The area of the second distribution area is expressed as:
；
in the method, in the process of the invention,is->Area of the second distribution area +.>Is of circumference rate>Is->Radius of the second distribution area;
will beAfter the areas of the second distribution areas are accumulated, a second sub-area is obtained;
the expression for the second sub-area is:
；/>
in the method, in the process of the invention,for the second sub-area->Is->The area of the second distribution area;
after the first sub-area and the second sub-area are added, comparing the first sub-area and the second sub-area with the trunk scanning image area to obtain the resin distribution density;
the expression of the resin distribution density is:
；
in the method, in the process of the invention,is the resin distribution density;
the insect pest detection index is used for indicating the severity of pine wood nematode disease of pine trees, when the insect pest detection index is larger, the severity of pine wood nematode disease of pine trees is indicated, and vice versa;
The expression of the pest detection index is:
；
in the method, in the process of the invention,for pest detection index, & lt & gt>、/>、/>For the rightHeavy factor, meta-L>The method comprises the steps of carrying out a first treatment on the surface of the Exemplary, ->0.47%>0.36%>0.17;
it should be noted that, the size of the weight factor is a specific numerical value obtained by quantizing each data, so that the subsequent comparison is convenient, and the size of the weight factor depends on the amount of the comprehensive pest data and the corresponding weight factor is preliminarily set for each group of comprehensive pest data by a person skilled in the art;
the early warning prompt module is used for comparing the pest detection index with a preset pest detection threshold value, generating a pest detection difference value and judging whether pest early warning prompt is generated or not;
the pest detection difference value is the difference value between the pest detection index and a preset pest detection threshold value, and is a numerical basis for judging whether the pest detection index reaches an early warning prompt or not;
the preset pest detection threshold is a numerical value basis for dividing the range of the pest detection indexes, so that the pest detection indexes are distinguished from an early warning range and a non-early warning range, different-size pest detection indexes can be conveniently screened, and the preset pest detection threshold is obtained by collecting a large number of pest detection indexes giving out early warning prompts in the history, selecting the minimum value of the pest detection indexes and optimizing the coefficient;
The expression of the pest detection difference is:
；
in the method, in the process of the invention,for pest detection difference, & gt>Is a preset pest detection threshold;
the judging method for generating the pest early warning prompt comprises the following steps:
when (when)When the pest detection index is greater than or equal to 0, the pest detection index is greater than or equal to a preset pest detection threshold, and when pine wood nematode disease appears, judging that pest early warning prompt is generated;
when (when)When the value is smaller than 0, the pest detection index is smaller than a preset pest detection threshold, and when pine does not have pine wood nematode disease, no pest early warning prompt is judged to be generated;
the level dividing module is used for generating pest early warning levels based on the pest detection difference value;
the insect pest early warning level is a basis for distinguishing the severity degree of pine wood nematode disease of pine trees, and different severity degrees of pine wood nematode disease correspond to different early warning levels, so that follow-up insect pest treatment measures on the pine trees are affected;
the insect pest early warning level comprises a first early warning level and a second early warning level, and the severity of the pine wood nematode corresponding to the first early warning level is smaller than that of the pine wood nematode corresponding to the second early warning level;
the generation method of the first-level early warning level and the second-level early warning level comprises the following steps:
difference value of pest detection And a preset level threshold->Comparison of (I)>Greater than 0; the preset level threshold is used for distinguishing the first-level early warning level and the second-level early warning level of the pest detection difference value, so that the pest detection difference values with different sizes can be corresponding to different degrees of severity of pine wood nematode disease, and the preset level threshold is obtained by acquiring a large number of pest detection difference values corresponding to the slight degree of pine wood nematode disease and the severity degree of pine wood nematode disease of the history and optimizing the coefficients;
when (when)Less than->When the insect pest detection difference value is smaller than a preset level threshold value, the degree of pine wood nematode disease occurrence is slight, and a first-level early warning level is generated;
when (when)Greater than or equal to->When the insect pest detection difference value is larger than or equal to a preset level threshold value, the degree of pine wood nematode disease is serious, and a second-level early warning level is generated;
the instruction making module is used for making a killing instruction based on the insect pest early warning level;
the killing instruction is applied to the treatment measures of pine wood nematode diseases of different insect pest early warning levels, so that the pine wood nematode diseases of pine trees are reasonably and accurately killed, and different insect pest early warning levels correspond to different killing instructions and represent different emergency degrees of killing treatment;
The killing instruction comprises a first-level killing instruction and a second-level killing instruction, and the method for formulating the first-level killing instruction and the second-level killing instruction comprises the following steps:
when the insect pest early warning level is a first early warning level, the emergency degree of the pine tree needing to be killed is low, and a first-level killing instruction is formulated;
when the insect pest early warning level is a secondary early warning level, the emergency degree of the pine tree needing to be killed is high, and a secondary killing instruction is formulated;
the emergency degree of the disinfection treatment corresponding to the first-level disinfection instruction is lower than the emergency degree corresponding to the second-level disinfection instruction, wherein the emergency degree refers to the sequence of disinfection emergency measures for pine wood nematode diseases of pine trees, and the emergency degree corresponding to the second-level disinfection instruction is lower because the degree of the pine wood nematode diseases corresponding to the first-level disinfection instruction is slight and the degree of harm to the pine wood nematode diseases does not reach the degree of emergency treatment required for the first time, and the emergency degree of the pine wood nematode diseases corresponding to the second-level disinfection instruction is serious and the degree of harm to the pine wood nematode diseases corresponding to the second-level disinfection instruction reaches the degree of emergency treatment required for the first time;
In the present embodiment, by acquiringThe satellite remote sensing images are based on screening criteria, for ∈K>Screening individual satellite remote sensing images to obtain a target image, marking area demarcation points in the target image, sequentially connecting the area demarcation points to obtain a target area, collecting comprehensive pest data of the target area, generating a pest detection index based on the comprehensive pest data, comparing the pest detection index with a preset pest detection threshold value to generate a pest detection difference value, judging whether to generate a pest early warning prompt, generating a pest early warning level based on the pest detection difference value, and formulating a killing instruction based on the pest early warning level; compared with the prior art, the method has the advantages that the non-target image with interference factors can be removed through screening and identifying the satellite remote sensing image, the target area with small area and accurate position can be identified based on the target image, and whether pine wood nematode disease occurs or not can be accurately detected through the target area, so that the acquisition amount and the calculation amount of detection data are effectively reduced, the data calculation rate is improved, and the possibility that useless data bring about the detection result is avoidedThe error influence greatly improves the accuracy of pine wood nematode disease detection. / >
Example 2: referring to fig. 2, this embodiment, which is not described in detail in embodiment 1, provides a method for detecting pine wood nematode disease based on image processing, which is applied to an image processing server and implemented by a pine wood nematode disease detection system based on image processing, and includes:
s1: by obtainingThe satellite remote sensing images are based on screening criteria, for ∈K>Screening the satellite remote sensing images to obtain a target image;
s2: marking in a target imageIndividual zone demarcation point, will->After the area demarcation points are connected in sequence, a target area is obtained;
s3: collecting comprehensive pest data of a target area, and generating a pest detection index based on the comprehensive pest data;
s4: comparing the pest detection index with a preset pest detection threshold value to generate a pest detection difference value, and judging whether a pest early warning prompt is generated or not; if the pest early warning prompt is generated, executing S5-S6; if the pest early warning prompt is not generated, repeating the steps S1-S4;
s5: generating an insect pest early warning level based on the insect pest detection difference value;
s6: based on the insect pest early warning level, a killing instruction is formulated.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (7)
1. The pine wood nematode disease detecting system based on image processing is applied to an image processing server and is characterized by comprising the following components:
target image screening module for obtainingThe satellite remote sensing images are based on screening criteria, for ∈K>Screening the satellite remote sensing images to obtain a target image;
target area recognition module for marking a target imageIndividual zone demarcation point, will->After the area demarcation points are connected in sequence, a target area is obtained;
the data acquisition module acquires comprehensive pest data of the target area and generates a pest detection index based on the comprehensive pest data;
the comprehensive pest data comprise a healthy area occupation ratio, crack dryness and resin distribution density;
the method for acquiring the health area occupation ratio comprises the following steps:
scanning a target area to obtain a crown scanning image;
marking pixel values of all pixel points in the crown scanning image, and counting the total amount of the pixel points;
comparing the pixel values of all the pixel points with a preset lower limit value and a preset upper limit value one by one;
the pixel points with the pixel values between the preset lower limit value and the preset upper limit value are marked as effective pixel points, and the number of the effective pixel points is counted;
comparing the number of the effective pixel points with the total number of the pixel points to obtain a healthy area occupation ratio;
The expression of the healthy area ratio is:
；
in the method, in the process of the invention,for the ratio of healthy area, ++>For the number of effective pixels, +.>Is the total pixel point amount;
the method for acquiring the crack dryness comprises the following steps:
scanning the target area to obtain a trunk scanning image, and equally dividing the trunk scanning image into parts according to a preset heightA sub-image;
identification by computer vision techniquesBark texture information on the sub-images and drawing lines along the positions of bark textures to obtain texture lines;
sequentially measuring horizontal distance values between vertical direction extension lines of two adjacent horizontal distribution texture lines along the horizontal direction, and marking the two adjacent texture lines with the horizontal distance values larger than a preset distance threshold value as target texture lines to obtainA group target texture line;
are positioned at the position of the measuring device in the vertical directionGroup target texture lineThe vertical distance value between the two adjacent vertical distribution texture line endpoints in the interior is horizontally drawn by taking the texture line endpoint with the minimum value of the vertical distance value as the base point, and is combined with +.>Group target texture line contact, get +.>A plurality of fracture zones;
one-to-one measurementThe length of each crack region and the height of each crack region are calculated by an area formula to obtain +. >The area of each crack;
the expression of the crack area is:
；
in the method, in the process of the invention,is->Sub-picture +.>Crack area of the individual crack region, +.>Is->Sub-picture +.>Length of the crack region->Is->Sub-picture +.>The height of the individual fracture zones;
will beSub-picture +.>After the area accumulation of the individual cracks, obtain +.>Sub-crack area;
the expression of the sub-crack area is:
；
in the method, in the process of the invention,is->Sub-slit area of sub-image +.>Is->No. of sub-image>Individual crackSeam area;
measuring the length of the trunk scanning image and the height of the trunk scanning image, and obtaining the trunk scanning image area through an area calculation formula;
the expression of the trunk scan image area is:
；
in the method, in the process of the invention,scanning the image area for the trunk>Scanning the length of the image for the trunk, < >>Scanning the height of the image for the trunk;
removingMaximum and minimum in the sub-crack area, the remaining +.>After the sub-crack areas are accumulated, comparing the sub-crack areas with the trunk scanning image area to obtain crack dryness;
the expression of the crack dryness is:
；
in the method, in the process of the invention,for crack dryness>Is->Sub-crack area;
the method for obtaining the resin distribution density comprises the following steps:
acquiring pixel values of all pixel points in the trunk scanning image, and comparing the pixel values of all pixel points with a preset first resin threshold value and a preset second resin threshold value;
The pixel points with the pixel values between the preset first resin threshold value and the preset second resin threshold value are marked as resin pixel points, and the obtained pixel points are obtainedA set of resin pixels;
edge of the frameThe boundary drawing line of the set of the resin pixel points minimizes and surrounds the resin pixel points to obtain +.>The areas to be identified;
identification by computer vision techniquesMarking the region to be identified, where bark texture information does not appear, as a resin region to obtain +.>Resin region->Less than->；
Measuring separatelyHorizontal length and vertical of individual resin regionsHeight, and based on the maximum value of the horizontal length and the maximum value of the vertical height, obtaining +.>A ratio of length to height;
the expression of the long-to-high ratio is:
；
in the method, in the process of the invention,is->Length to height ratio of individual resin areas, +.>Is->Maximum value of horizontal length of individual resin region, +.>Is->Maximum vertical height of each resin region;
the method comprises the steps of marking a resin area with a length-height ratio smaller than a preset length-height threshold value as a first distribution area, and marking a resin area with a length-height ratio larger than or equal to the preset length-height threshold value as a second distribution area;
drawing a rectangle on the boundary line of the first distribution area, measuring the length and width of the rectangle, and respectively recording the length and the height of the first distribution area to obtain Area of the first distribution area, +.>Less than->；
The area of the first distribution area is expressed as:
；
in the method, in the process of the invention,is->Area of the first distribution area, +.>Is->The length of the first distribution area, +.>Is->The height of the first distribution areas;
will beAfter the areas of the first distribution areas are accumulated, a first sub-area is obtained;
the expression for the first sub-area is:
；
in the method, in the process of the invention,for the first sub-area->Is->The area of the first distribution area;
drawing a circle on the boundary line of the second distribution area, measuring the radius of the circle, and recording as the radius of the second distribution area to obtainArea of the second distribution area +.>Less than->；
The area of the second distribution area is expressed as:
；
in the method, in the process of the invention,is->Area of the second distribution area +.>Is of circumference rate>Is->Radius of the second distribution area;
will beAfter the areas of the second distribution areas are accumulated, a second sub-area is obtained;
the expression for the second sub-area is:
；
in the method, in the process of the invention,for the second sub-area->Is->The area of the second distribution area;
after the first sub-area and the second sub-area are added, comparing the first sub-area and the second sub-area with the trunk scanning image area to obtain the resin distribution density;
the expression of the resin distribution density is:
；
In the method, in the process of the invention,is the resin distribution density;
the early warning prompt module is used for comparing the pest detection index with a preset pest detection threshold value, generating a pest detection difference value and judging whether pest early warning prompt is generated or not;
the level dividing module is used for generating pest early warning levels based on the pest detection difference value;
and the instruction making module is used for making a killing instruction based on the insect pest early warning level.
2. The image processing-based pine wood nematode disease detection system of claim 1, wherein the screening criteria is: the actual resolution of the image is greater than 90% of the rated resolution, and the ratio of the pixel points of the pine tree is greater than the ratio of the lowest pixel points;
the target image acquisition method comprises the following steps:
acquisition of selected pine areas by means of an image databaseA satellite remote sensing image;
obtaining by looking at image propertiesThe actual resolution of each satellite remote sensing image is based on a screening criterion, and the satellite remote sensing image with the actual resolution being more than 90% of the rated resolution is marked as a marked image;
identifying the region of the pine tree in the marked image by a computer vision technology, and counting the number of pixels in the region of the pine tree;
comparing the number of the pixels in the area where the pine tree is positioned with the total number of the pixels in the marked image to obtain the ratio of the pixels in the pine tree;
Based on the screening criterion, screening out that the pixel point occupation ratio of pine is larger than the lowest pixel point occupation ratioMarking images->Less than->；
From the slaveSelecting the mark image corresponding to the maximum value of the pixel point occupation ratio of pine from the mark images, namelyAnd (5) marking images.
3. The image processing-based pine wood nematode disease detection system according to claim 2, wherein the target area acquisition method includes:
scanning the target image to obtain a scanning gray image, and labeling the pixel value of each pixel point in the scanning gray image;
marking the positions of pixel points with pixel values larger than a preset pixel threshold value to obtain effective positions;
marking along boundary lines of effective positions according to preset lengthsPersonal location, get->A plurality of zone demarcation points;
will be clockwiseAnd after the area demarcation points are connected in sequence, obtaining the target area.
4. A pine wood nematode disease detection system according to claim 3, wherein the expression of the pest detection index is:
；
in the method, in the process of the invention,for pest detection index, & lt & gt>、/>、/>Is a weight factor.
5. The image processing-based pine wood nematode disease detection system of claim 4, wherein the expression of the pest detection difference is:
；
In the method, in the process of the invention,for pest detection difference, & gt>Is a preset pest detection threshold;
the judging method for generating the pest early warning prompt comprises the following steps:
when (when)When the insect pest early warning prompt is larger than or equal to 0, judging to generate the insect pest early warning prompt;
when (when)And when the value is smaller than 0, judging that the insect pest early warning prompt is not generated.
6. The image processing-based pine wood nematode disease detection system of claim 5, wherein the pest early warning level comprises a primary early warning level and a secondary early warning level;
the generation method of the first-level early warning level and the second-level early warning level comprises the following steps:
difference value of pest detectionAnd a preset level threshold->Comparison of (I)>Greater than 0;
when (when)Less than->Generating a first-level early warning level;
when (when)Greater than or equal to->And generating a second-level early warning level.
7. The image processing-based pine wood nematode disease detection system of claim 6, wherein the kill instruction includes a primary kill instruction and a secondary kill instruction;
the method for formulating the primary disinfection instruction and the secondary disinfection instruction comprises the following steps:
when the insect pest early warning level is a first early warning level, a first-level killing instruction is formulated;
when the pest early warning level is the secondary early warning level, a secondary killing instruction is formulated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410107688.7A CN117636185B (en) | 2024-01-26 | 2024-01-26 | Pine wood nematode disease detecting system based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410107688.7A CN117636185B (en) | 2024-01-26 | 2024-01-26 | Pine wood nematode disease detecting system based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117636185A CN117636185A (en) | 2024-03-01 |
| CN117636185B true CN117636185B (en) | 2024-04-09 |
Family
ID=90021954
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410107688.7A Active CN117636185B (en) | 2024-01-26 | 2024-01-26 | Pine wood nematode disease detecting system based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117636185B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104881865A (en) * | 2015-04-29 | 2015-09-02 | 北京林业大学 | Forest disease and pest monitoring and early warning method and system based on unmanned plane image analysis |
| CN108596104A (en) * | 2018-04-26 | 2018-09-28 | 安徽大学 | A kind of wheat powdery mildew remote-sensing monitoring method with Disease Characters preprocessing function |
| CN111028255A (en) * | 2018-10-10 | 2020-04-17 | 千寻位置网络有限公司 | Farmland area pre-screening method and device based on prior information and deep learning |
| CN113011354A (en) * | 2021-03-25 | 2021-06-22 | 浙江农林大学 | Unmanned aerial vehicle hyperspectral image pine wood nematode disease identification method based on deep learning |
| CN113435252A (en) * | 2021-05-27 | 2021-09-24 | 广西壮族自治区烟草公司百色市公司 | Tobacco pest and disease monitoring and early warning method and system based on remote sensing |
| CN113468964A (en) * | 2021-05-31 | 2021-10-01 | 山东省邮电工程有限公司 | Hyperspectrum-based agricultural disease and pest monitoring method and device |
| CN114595975A (en) * | 2022-03-11 | 2022-06-07 | 安徽大学 | Unmanned aerial vehicle remote sensing pine wood nematode disease monitoring method based on deep learning model |
| CN116612192A (en) * | 2023-07-19 | 2023-08-18 | 山东艺术学院 | Digital video-based pest and disease damage area target positioning method |
| CN116863342A (en) * | 2023-09-04 | 2023-10-10 | 江西啄木蜂科技有限公司 | Large-scale remote sensing image-based pine wood nematode dead wood extraction method |
| CN116883847A (en) * | 2023-07-11 | 2023-10-13 | 重庆英卡电子有限公司 | Pine wood nematode disease spread monitoring and early warning method and system |
| CN116912583A (en) * | 2023-07-19 | 2023-10-20 | 久瓴(上海)智能科技有限公司 | Epidemic wood data processing method and device, storage medium and electronic equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018191442A1 (en) * | 2017-04-11 | 2018-10-18 | Agerpoint, Inc. | Forestry management tool for assessing risk of catastrophic tree failure due to weather events |
-
2024
- 2024-01-26 CN CN202410107688.7A patent/CN117636185B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104881865A (en) * | 2015-04-29 | 2015-09-02 | 北京林业大学 | Forest disease and pest monitoring and early warning method and system based on unmanned plane image analysis |
| CN108596104A (en) * | 2018-04-26 | 2018-09-28 | 安徽大学 | A kind of wheat powdery mildew remote-sensing monitoring method with Disease Characters preprocessing function |
| CN111028255A (en) * | 2018-10-10 | 2020-04-17 | 千寻位置网络有限公司 | Farmland area pre-screening method and device based on prior information and deep learning |
| CN113011354A (en) * | 2021-03-25 | 2021-06-22 | 浙江农林大学 | Unmanned aerial vehicle hyperspectral image pine wood nematode disease identification method based on deep learning |
| CN113435252A (en) * | 2021-05-27 | 2021-09-24 | 广西壮族自治区烟草公司百色市公司 | Tobacco pest and disease monitoring and early warning method and system based on remote sensing |
| CN113468964A (en) * | 2021-05-31 | 2021-10-01 | 山东省邮电工程有限公司 | Hyperspectrum-based agricultural disease and pest monitoring method and device |
| CN114595975A (en) * | 2022-03-11 | 2022-06-07 | 安徽大学 | Unmanned aerial vehicle remote sensing pine wood nematode disease monitoring method based on deep learning model |
| CN116883847A (en) * | 2023-07-11 | 2023-10-13 | 重庆英卡电子有限公司 | Pine wood nematode disease spread monitoring and early warning method and system |
| CN116612192A (en) * | 2023-07-19 | 2023-08-18 | 山东艺术学院 | Digital video-based pest and disease damage area target positioning method |
| CN116912583A (en) * | 2023-07-19 | 2023-10-20 | 久瓴(上海)智能科技有限公司 | Epidemic wood data processing method and device, storage medium and electronic equipment |
| CN116863342A (en) * | 2023-09-04 | 2023-10-10 | 江西啄木蜂科技有限公司 | Large-scale remote sensing image-based pine wood nematode dead wood extraction method |
Non-Patent Citations (3)
| Title |
|---|
| UAV remote sensing monitoring of pine forest diseases based on improved Mask R-CNN;Gensheng Hu et al.;International Journal of Remote Sensing;20220307;全文 * |
| 基于无人机图像分形特征的油松受灾级别判定;费运巧;刘文萍;陆鹏飞;骆有庆;;计算机应用研究;20170430(第04期);全文 * |
| 针对卫星图像的语义分割算法研究;宋天龙;中国优秀硕士学位论文全文数据库;20190115;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117636185A (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108759973B (en) | Water level measuring method | |
| Pouliot et al. | Automated tree crown detection and delineation in high-resolution digital camera imagery of coniferous forest regeneration | |
| WO2017190574A1 (en) | Fast pedestrian detection method based on aggregation channel features | |
| Awrangjeb et al. | Improved building detection using texture information | |
| CN109472261B (en) | Computer vision-based automatic monitoring method for grain storage quantity change of granary | |
| CN106910187B (en) | Image data set artificial amplification method for bridge crack detection and positioning | |
| Wspanialy et al. | Early powdery mildew detection system for application in greenhouse automation | |
| CN111007013B (en) | Crop rotation fallow remote sensing monitoring method and device for northeast cold region | |
| CN110376202B (en) | Tea tree anthracnose lesion identification method based on imaging hyperspectral technology | |
| Masood et al. | Plants disease segmentation using image processing | |
| Ji et al. | In-field automatic detection of maize tassels using computer vision | |
| CN111027446A (en) | Coastline automatic extraction method of high-resolution image | |
| Al-Tarawneh | An empirical investigation of olive leave spot disease using auto-cropping segmentation and fuzzy C-means classification | |
| CN106373134A (en) | Road index line detection method based on linear structured light three-dimensional measurement | |
| CN107610122A (en) | Insect pest detection method inside a kind of single grain cereal based on Micro CT | |
| CN110188657A (en) | Corn arid recognition methods based on crimping blade detection | |
| KR101874968B1 (en) | Visibility measuring system base on image information and method for using the same | |
| CN111060455B (en) | Northeast cold-cool area oriented remote sensing image crop marking method and device | |
| Wang et al. | An automatic system for pest recognition and forecasting | |
| CN117636185B (en) | Pine wood nematode disease detecting system based on image processing | |
| WO2021248599A1 (en) | Method and system for automatically identifying spot of which category is abnormal | |
| WO2021229248A1 (en) | Method for detection and classification of biotic - abiotic stress in crops from thermal photographs using artificial intelligence | |
| CN110929739B (en) | Automatic impervious surface range remote sensing iterative extraction method | |
| CN109993071B (en) | Method and system for automatically identifying and investigating color-changing forest based on remote sensing image | |
| Li et al. | The disease assessment of cucumber downy mildew based on image processing |
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 |