CN115909113A - Method for surveying forestry pests through remote sensing monitoring of unmanned aerial vehicle - Google Patents
Method for surveying forestry pests through remote sensing monitoring of unmanned aerial vehicle Download PDFInfo
- Publication number
- CN115909113A CN115909113A CN202310029641.9A CN202310029641A CN115909113A CN 115909113 A CN115909113 A CN 115909113A CN 202310029641 A CN202310029641 A CN 202310029641A CN 115909113 A CN115909113 A CN 115909113A
- Authority
- CN
- China
- Prior art keywords
- pest
- rlsub
- area
- point
- aerial vehicle
- 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.)
- Granted
Links
Images
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
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
- Y02A40/28—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture specially adapted for farming
Abstract
The invention belongs to the technical field of forestry big data, and provides a method for surveying forestry pests by remote sensing of an unmanned aerial vehicle, wherein a spectrum image sequence is formed by continuously collecting spectrum images of a region to be monitored, and insect pest regions of each spectrum image in the identification sequence are obtained according to an RX algorithm; and correcting the position of the insect pest area on the reference image through the identification sequence. The offset positions of the identification points on the insect pest area and the unmanned aerial vehicle air route can be identified, the strong interference of the insect pest area on the local change of the plant in the natural state to the accurate identification of the insect pest area can be removed, and the identification accuracy of the insect pest area is improved. The difference between the positions of the non-pest region and the pest region is judged in a balanced mode, and the pest region can be increased and deleted according to the difference on the reference image intelligently.
Description
Technical Field
The invention belongs to the technical field of forestry big data, and particularly relates to a method for surveying forestry pests by remote sensing of an unmanned aerial vehicle.
Background
At present, a method for monitoring forestry pests (forestry pests such as pine moth, dead tree of pine wood nematode disease, broad-leaved tree species leaf eating pests and the like) through remote sensing collects remote sensing images in a monitoring area, compares the remote sensing images with spectral characteristic information of pixel points in historical remote sensing images in the monitoring area, and further judges whether pests exist in the area, but due to the change of plants in the natural state (vegetation growth, water content change of leaves caused by dry environment or coverage change of the leaves), the change of the spectral characteristic information of the pixel points generates strong interference on the spectral characteristic information identification of the pests, and the error identification rate is high.
To solve the problem, china with publication number CN115131683A specially facilitates 9, 30 and 2022, and discloses a forestry information identification method based on high-resolution remote sensing images, wherein pixel points are clustered by referring to spectral feature vectors of the pixel points in an image, and different tree species of a to-be-detected area are divided; deformation of a spectral curve of the pixel point and offset of gray values of different wave bands are obtained by calculating difference of wavelength and gray values in a gray value sequence of the pixel point, and damage probability is comprehensively judged by combining uniformity of variation of gray values of all wave bands, so that the condition that curve offset is high due to high vegetation coverage degree is eliminated, and accuracy of forestry pest detection is improved; the damage probability is adjusted according to the uniformity of the damage probability of the same tree species to obtain the final damage probability so as to judge the possibility of pest and disease damage, thereby realizing pest and disease damage detection of the same tree species in different seasons; however, in many conventional techniques including this technique, if a pest region already exists in a reference image or when a plant itself is locally changed in a natural state, strong interference occurs, so that recognition accuracy is lowered, the pest region cannot be recognized even, and a large calculation cost is required because an image of a region is large and recognition speed is slow.
Disclosure of Invention
The invention aims to provide a method for remotely sensing and monitoring forestry pests by an unmanned aerial vehicle to solve one or more technical problems in the prior art and provide at least one beneficial choice or creation condition.
In order to achieve the purpose, the invention provides a method for remotely sensing, monitoring and surveying forestry pests by an unmanned aerial vehicle, which specifically comprises the following steps:
s1: the method comprises the steps of obtaining an electronic map of a forest land to be monitored as an area to be monitored, wherein the forest land to be monitored comprises Pinaceae plants, broad-leaved tree species, eucalyptus tree species and/or liana plants (Pinaceae plants, namely slash pine, chinese red pine, larch and the like; broad-leaved tree species, namely poplar, elm, camphor tree and the like; eucalyptus tree species, namely eucalyptus citriodora, guangjiu and the like; and liana plants, namely mikania micrantha, pointwort, bellflower and the like);
s2: setting the information of a starting point, a terminal point and a barrier position of the unmanned aerial vehicle flying in the area to be monitored, and setting a route path of the unmanned aerial vehicle from the starting point to the terminal point in the area to be monitored;
s3: starting from a starting point, taking a plurality of points on an underway path as identification points at a fixed acquisition distance;
s4: the unmanned aerial vehicle flies according to a flight path and continuously collects spectrum images of an area to be monitored, a sequence formed by various spectrum images is used as a spectrum image sequence in sequence, and a spectrum image is obtained at each identification point in the spectrum image sequence; (because the ground coverage of each spectral image is limited by the narrow field of view of the spectrometer carried by the unmanned aerial vehicle, the spectral images in the field of view of the spectrometer need to be continuously collected to form a spectral image data set in the flight process, and then the spectral images in the spectral image data set are spliced to obtain an integral spectral image, so that the integral spectral image can effectively cover the area to be monitored);
s5: acquiring a sequence formed by all spectral images at all identification point positions in the spectral image sequence as an identification sequence;
s6: acquiring insect pest areas of each spectral image in the identification sequence according to an RX algorithm;
s7: correcting the position of the insect pest area on the reference image through the identification sequence;
s8: and marking the positions of the insect pest areas on the corrected reference image on an electronic map of the forest land to be monitored.
Further, in S1, the electronic map is discrete data of ground elements and phenomena having determined coordinates and attributes within a certain coordinate system acquired through a satellite, an unmanned aerial vehicle.
Further, in S2, the route path of the unmanned aerial vehicle from the starting point to the end point is set in the area to be monitored, and the route path of the unmanned aerial vehicle from the starting point to the end point in the area to be monitored comprises any one of A-Algorithm, K-shortest path Algorithm-based, dijkstra Algorithm, APF (APF: artificial Potential Field Algorithm) and SAA (SAA: simulated Annealing Algorithm).
The unmanned aerial vehicle carries out flying movement on the electronic map according to the position coordinate data corresponding to the air route path in the area to be monitored on the electronic map.
The starting point is the takeoff position (generally the current position) of the unmanned aerial vehicle in the area to be monitored, the end points are the positions of the end points of the unmanned aerial vehicle flying in the area to be monitored, the barrier position information is the positions of the barriers in the area to be monitored, and the unmanned aerial vehicle needs to keep a preset barrier safety distance (2-5 m) with the positions of the barriers.
Further, setting the initial parameters of the unmanned aerial vehicle: the spectral imaging system has a field angle (15-25 degrees), a course distance (8-35 m), a flight height (15-55 m), a flight speed (10-20 m/s), a boundary safety distance (2-5 m) and an obstacle safety distance (2-5 m).
Preferably, the field angle of the spectral imaging system is set to be 16 degrees, the flying height is set to be 30m, and the flying speed is set to be 13m/s.
Further, in S3, the fixed acquisition pitch is 8 to 35m.
Further, in S4, continuously acquiring the spectral image of the area to be monitored is to acquire the spectral image of the area to be monitored by using a spectrometer carried by the unmanned aerial vehicle during the flight process according to the flight path.
Further, in S6, the method for acquiring the pest region of each spectral image in the identification sequence according to the RX algorithm includes: and inputting the spectral reflectivity of each pixel point in the spectral image into an RX (Reed-Xiaooli hyperspectral target detection algorithm) algorithm to identify the insect pest area at the corresponding position of the area to be monitored in the spectral image.
The remote sensing monitoring of the plant diseases and insect pests is judged by measuring the change of the chlorophyll content in the plant, and the spectral reflectivity of the chlorophyll has obvious characteristics and can change along with the change of the wavelength. The spectral reflectance of chlorophyll of plants is very low at 0.5-0.7 μm, and is significantly increased at 0.7-0.9 μm in the near infrared band, because green plants can absorb the radiant energy of this band. As chlorophyll in plant bodies of plant diseases and insect pests is gradually reduced, the light absorption capacity is reduced, the reflectivity of visible light is obviously improved, and the reflectivity of an infrared region is obviously reduced, especially in a near-infrared band.
Preferably, the method for acquiring the pest region of each spectral image in the identification sequence according to the RX algorithm comprises the following steps: insect pest areas at corresponding positions in an unmanned aerial vehicle operation area are obtained by the method disclosed in the patent publication No. CN107347849A (RX abnormal detection method).
The insect pest area is an area formed by the insect pest pixel points or an internal area of a margin line formed by the insect pest pixel points; the pest and disease pixel points are corresponding pixel points in the spectral image, wherein the visible light reflectivity of the pest and disease pixel points is higher than the average value of the visible light reflectivity of each pixel point in the spectral image, or the pest and disease pixel points are corresponding pixel points in the spectral image, wherein the near-infrared band spectral reflectivity of the pest and disease pixel points is lower than the average value of the near-infrared band spectral reflectivity of each pixel point in the spectral image.
Further, in S7, the reference image is an overall spectral image of the to-be-monitored area spliced by the spectral images acquired in the previous time interval in steps S1 to S6, and the locations of the insect pest areas identified in step S6 are marked on the corresponding locations of the to-be-monitored area in the overall spectral image. (the time interval is generally 7-30 days), wherein the spectral image stitching comprises the steps of geometric correction, image preprocessing, image registration and image fusion.
Preferably, in S7, the reference image is a spectral image obtained by an onboard full-spectrum multi-modal imaging spectrometer or a spectral image obtained by a hyperspectral remote sensing satellite, the pest region of the spectral image is identified according to an RX algorithm, and the pest region is marked on the pest region to be monitored at a position corresponding to the pest region of the spectral image.
Furthermore, the unmanned aerial vehicle is a coaxial double-rotor unmanned aerial vehicle, a micro-rotor unmanned aerial vehicle or a multi-rotor unmanned aerial vehicle carrying a spectrometer; preferably, the spectrometer is a GaiaField-mini spectrometer, a specific AFX series hyperspectral camera or an ATH9020 hyperspectral imager.
Further, in S7, the method for correcting the location of the pest region on the reference image by the identification sequence includes the steps of:
marking the identification sequence as RLocal, taking each spectral image in the RLocal as an identification partition, wherein RLocal = { RL (i) }, i is the serial number of the identification partition, i belongs to [1, N1], N1 is the number of the identification partitions, and RL (i) is the ith identification partition in the RLocal sequence;
in the value range of i, calculating Euclidean distances between the geometric gravity center point of each insect pest region in RL (i) and P1 (i) by taking the identification point corresponding to RL (i) as P1 (i), and then taking the mean value of all the Euclidean distances in RL (i) as RLmean (i); selecting a point of a corresponding position of the geometric gravity center point with the maximum distance value from the geometric gravity center point P1 (i) of each insect pest region in the RL (i) on the reference image as a far center point P2 (i), and selecting a point of a corresponding position of the geometric gravity center point with the minimum distance value from the geometric gravity center point P1 (i) of each insect pest region in the RL (i) on the reference image as a near center point P3 (i);
screening all the insect pest areas from the reference image, wherein a set formed by the insect pest areas with the distances from the geometric gravity center points of all the insect pest areas to P2 (i) being smaller than RLmean (i) is recorded as RLSUB (i); and/or screening out a set formed by insect pest regions, wherein the distances from the geometric gravity center points of all insect pest regions to P2 (i) are smaller than RLmean (i), and the mean value of the near infrared band spectral reflectances of all pixel points in the insect pest regions is lower than the mean value of the near infrared band spectral reflectances of all pixel points in the insect pest regions corresponding to remote center points P2 (i) in RL (i), and marking as RLSUB (i);
taking N2 as the number of elements in RLSUB (i), wherein RLSUB (i, j) is the geometric gravity center of the jth pest region in RLSUB (i), j is the serial number of the elements in RLSUB (i), and j belongs to [1, N2];
if N2=0, then RL (i) is noted as the identified partition that does not require correction; and if N2 is more than 0, correcting the position of the insect pest area inside the corresponding position of RL (i) on the reference image in the value range of i.
Above scheme can discern the offset position of identification point on insect pest area and the unmanned aerial vehicle airline, can discern the accurate relative position in insect pest area, guarantees the relative precision of insect pest position coordinate.
Further, the method for correcting the position of the insect pest area inside the position corresponding to the RL (i) on the reference image in the value range of i comprises the following steps:
in the value range of j, calculating the mean value of the distances between the far center point P2 (i) and each RLSUB (i, j) in the RLSUB (i) and recording the mean value as the far center distance AD; calculating the mean value of the distance between the near-center point P3 (i) and each RLSUB (i, j) in the RLSUB (i) as the near-center distance BD; recording a point PF at a corresponding position on the reference image of the RLSUB (i, j) having the shortest distance between the remote center point P2 (i) and each RLSUB (i, j); a point at a corresponding position on the reference image of the RLSUB (i, j) with the shortest distance between the approximate center point P3 (i) and each RLSUB (i, j) is represented as PN;
when AD is larger than or equal to BD, taking the direction from the center point P2 (i) to PF as an adjusting direction, moving the insect region with the geometric center point closest to PF in each insect region on the reference image by a telecentric distance AD towards the adjusting direction,
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center and a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if the correction range A has insect pest areas except the insect pest area corresponding to P2 (i) in the corresponding area on RL (i), the insect pest areas on RL (i) are copied to the corresponding positions of the insect pest areas on the reference image.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B has insect pest regions except the insect pest region corresponding to P3 (i) in the corresponding region on RL (i), the insect pest regions on RL (i) are copied to the corresponding positions of the insect pest regions on the reference image.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center of a circle and takes a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if the correction range A does not have a pest region except the pest region corresponding to P2 (i) in the corresponding region on RL (i), the pest region in the correction range A on the reference image is deleted.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a radius of a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B does not have a pest region except the pest region corresponding to P3 (i) in the corresponding region on RL (i), the pest region in the correction range B on the reference image is deleted.
According to the technical scheme, the telecentric distance AD and the telecentric distance BD are offset distances from the geometric center of gravity of the insect pest region by taking the identification point as a reference, strong interference generated by accurate identification of the insect pest region when the insect pest region changes through the local part of the plant in the natural state can be removed through correction of the offset distances, identification accuracy of the insect pest region is improved, the current local image is processed, so that the identification speed is greatly improved without identifying the whole image, and the operation cost is reduced.
In order to reduce the influence of local change of the plant in the natural state on part of the non-pest area, and further improve the positioning and correction accuracy of the pest area, the invention provides the following preferable scheme:
preferably, the method for correcting the position of the pest region on the reference image within the value range of i comprises the following steps:
calculating the insect pest position deviation index dev (i) of RL (i), wherein the specific method comprises the following steps:
wherein the deviation degree function
| P2 (i) -RLSUB (i, j) | is the distance between the far center point P2 (i) and RLSUB (i, j);
p3 (i) -RLSUB (i, j) | is the distance between the proximal point P3 (i) and RLSUB (i, j);
calculating pest position deviation indexes dev (i) of all RL (i) in the value range of i, calculating the average value of all dev (i) as meandev, and marking all identification partitions RL (i) of dev (i) being more than or equal to meandev as the to-be-optimized identification partitions;
when RL (i) is the partition to be identified for optimization, calculating the mean value of the distances between the far center point P2 (i) and each RLSUB (i, j) in RLSUB (i) in the range of the value of j and recording the mean value as the far center distance AD; calculating the mean value of the distances between the near center point P3 (i) and each RLSUB (i, j) in the RLSUB (i) and recording the mean value as the near center distance BD; recording a point PF at a corresponding position on the reference image of the RLSUB (i, j) having the shortest distance between the remote center point P2 (i) and each RLSUB (i, j); a point at a corresponding position on the reference image of the RLSUB (i, j) with the shortest distance between the approximate center point P3 (i) and each RLSUB (i, j) is represented as PN;
when AD is larger than or equal to BD, taking the direction from the center point P2 (i) to PF as an adjusting direction, and moving the insect pest area with the geometric center point closest to PF in each insect pest area on the reference image towards the adjusting direction by a telecentric distance AD;
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
When AD is larger than or equal to BD, taking the direction from the center point P2 (i) to PF as an adjusting direction, moving the insect region with the geometric center point closest to PF in each insect region on the reference image by a telecentric distance AD towards the adjusting direction,
when AD < BD, the direction from the near center point P3 (i) to PN is used as an adjusting direction, and the insect damage area with the geometric center point closest to PN in each insect damage area on the reference image is moved towards the adjusting direction by the near center distance BD.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center and a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if the correction range A has insect pest areas except the insect pest area corresponding to P2 (i) in the corresponding area on RL (i), the insect pest areas on RL (i) are copied to the corresponding positions of the insect pest areas on the reference image.
Preferably, when AD < BD, a range which takes P3 (i) as a center and takes a connecting line from P3 (i) to PN as a radius on the reference image is taken as a correction range B, and if the correction range B has insect damage areas except the insect damage area corresponding to P3 (i) in the corresponding area on RL (i), the insect damage areas on RL (i) are copied to the corresponding positions of the insect damage areas on the reference image.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center and a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if no pest region except the pest region corresponding to P2 (i) exists in the corresponding region of the correction range A on RL (i), the pest region in the correction range A on the reference image is deleted.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a radius of a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B does not have a pest region except the pest region corresponding to P3 (i) in the corresponding region on RL (i), the pest region in the correction range B on the reference image is deleted.
According to the preferred scheme, the pest position deviation index is the deviation degree between the far center point and the near center point of the pest position and the identification point, so that the difference between the positions of the non-pest region and the pest region is judged in a balanced manner, the pest region can be increased or deleted according to the difference, the position of the pest region can be adjusted accurately, and the influence on the pest region caused by the fact that part of the non-pest region passes through the plant in the natural state changes is reduced.
The invention also provides a system for remotely sensing, monitoring and surveying the forestry pests by the unmanned aerial vehicle, which comprises the following components: the processor executes the computer program to realize steps in the method for remotely sensing and monitoring and surveying the forestry pest by the unmanned aerial vehicle, the system for remotely sensing and surveying the forestry pest by the unmanned aerial vehicle can be operated in computing equipment such as a desktop computer, a notebook computer, a palm computer, a cloud data center and the like, the operable system can comprise, but is not limited to, the processor, the memory and a server cluster, and the processor executes the computer program to operate in units of the following systems:
the map acquisition unit is used for acquiring an electronic map of the forest land to be monitored as an area to be monitored;
the route setting unit is used for setting the information of a starting point, an end point and a barrier position of the unmanned aerial vehicle flying in the area to be monitored, and setting a route path of the unmanned aerial vehicle from the starting point to the end point in the area to be monitored;
the identification dividing unit is used for taking a plurality of points on the route of the underway as identification points at a fixed acquisition distance from a starting point;
the spectrum acquisition unit is used for enabling the unmanned aerial vehicle to fly according to a route path and continuously acquire spectrum images of an area to be monitored, a sequence formed by various spectrum images is sequentially used as a spectrum image sequence, and a spectrum image is acquired at each identification point in the spectrum image sequence;
the identification extraction unit is used for acquiring a sequence formed by all the spectral images at all the identification point positions in the spectral image sequence as an identification sequence;
the pest identification unit is used for acquiring pest areas of each spectral image in the identification sequence according to an RX algorithm;
a pest correction unit for correcting the location of the pest region on the reference image by the identification sequence;
and the map marking unit is used for marking the insect pest area position on the corrected reference image on the electronic map of the forest land to be monitored.
The invention has the beneficial effects that: the invention provides a method for surveying forestry pests by remote sensing of an unmanned aerial vehicle, which can identify the offset positions of an insect pest area and an identification point on an unmanned aerial vehicle air line, can identify the accurate relative position of the insect pest area, ensure the relative precision of the coordinates of the insect pest position, remove the strong interference generated by the accurate identification of the insect pest area when the insect pest area changes through the local part of a plant in the natural state in a reference image by correcting the offset distance, improve the identification accuracy of the insect pest area, and greatly improve the identification speed without identifying the whole image because the current local image is processed, thereby reducing the operation cost. The difference between the positions of the non-pest region and the pest region is judged in a balanced manner, pest regions can be increased and deleted intelligently according to the difference, the positions of the pest regions can be adjusted accurately, and the influence on the pest regions caused by local change of the plants in the natural state in part of the non-pest regions is reduced.
Drawings
The above and other features of the invention will be more apparent from the detailed description of the embodiments shown in the accompanying drawings in which like reference characters designate the same or similar elements, and it will be apparent that the drawings in the following description are merely exemplary of the invention and that other drawings may be derived by those skilled in the art without inventive effort, wherein:
FIG. 1 is a flow chart of a method for remotely sensing and monitoring forestry pests by an unmanned aerial vehicle;
FIG. 2 is a structure diagram of a system for remotely sensing, monitoring and surveying forestry pests by an unmanned aerial vehicle
FIG. 3 is an electronic map of a forest area to be monitored according to an embodiment of the present invention;
fig. 4 is a map marking position of an insect pest area in an electronic map of a forest land to be monitored, which is identified according to an embodiment of the invention.
Detailed Description
The conception, the specific structure and the technical effects produced by the present invention will be clearly and completely described in conjunction with the embodiments and the attached drawings, so as to fully understand the objects, the schemes and the effects of the present invention. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The same reference numbers will be used throughout the drawings to refer to the same or like parts.
Fig. 1 is a flow chart of a method for remotely monitoring and surveying forestry pests by an unmanned aerial vehicle according to the present invention, and fig. 1 is a flowchart illustrating a method for remotely monitoring and surveying forestry pests by an unmanned aerial vehicle according to an embodiment of the present invention, and a preferred embodiment is described in detail. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.
S1: acquiring an electronic map of a forest land to be monitored as an area to be monitored;
s2: setting the information of a starting point, a terminal point and a barrier position of the unmanned aerial vehicle flying in the area to be monitored, and setting a route path of the unmanned aerial vehicle from the starting point to the terminal point in the area to be monitored;
s3: starting from a starting point, taking a plurality of points on the route of the flight as identification points at a fixed acquisition distance;
s4: the unmanned aerial vehicle flies according to a route path and continuously collects spectrum images of an area to be monitored, a sequence formed by various spectrum images is used as a spectrum image sequence in sequence, and a spectrum image is obtained at each identification point in the spectrum image sequence; (because the ground coverage of each spectral image is limited by the narrow field of view of the spectrometer carried by the unmanned aerial vehicle, the spectral images in the field of view of the spectrometer need to be continuously collected to form a spectral image data set in the flight process, and then the spectral images in the spectral image data set are spliced to obtain an integral spectral image, so that the integral spectral image can effectively cover the area to be monitored);
s5: acquiring a sequence formed by all spectral images at all identification point positions in the spectral image sequence as an identification sequence;
s6: acquiring insect pest areas of each spectral image in the identification sequence according to an RX algorithm;
s7: correcting the position of the insect pest area on the reference image through the identification sequence;
s8: and marking the positions of the insect pest areas on the corrected reference image on the electronic map of the forest land to be monitored.
Further, in S1, the electronic map is discrete data of ground elements and phenomena having determined coordinates and attributes within a certain coordinate system acquired through a satellite, an unmanned aerial vehicle.
Further, in S2, the route path of the unmanned aerial vehicle from the starting point to the end point in the area to be monitored is set to comprise any one of A-Algorithm, K-shortest path Algorithm based, dijkstra Algorithm, APF Algorithm (APF: artificial Potential Field Algorithm) and SAA Algorithm (SAA: simulated Annealing Algorithm) from the starting point to the end point in the area to be monitored.
The starting point is the takeoff position (generally the current position) of the unmanned aerial vehicle in the area to be monitored, the end points are the positions of the end points of the unmanned aerial vehicle flying in the area to be monitored, the barrier position information is the positions of the barriers in the area to be monitored, and the unmanned aerial vehicle needs to keep a preset barrier safety distance (2-5 m) with the positions of the barriers.
Further, setting the initial parameters of the unmanned aerial vehicle: the spectral imaging system has a field angle (15-25 degrees), a course distance (8-35 m), a flight height (15-55 m), a flight speed (10-20 m/s), a boundary safety distance (2-5 m) and an obstacle safety distance (2-5 m).
Preferably, the field angle of the spectral imaging system is set to be 16 degrees, the flying height is set to be 30m, and the flying speed is set to be 13m/s.
Further, in S3, the fixed acquisition pitch is 8 to 35m.
Further, in S4, the continuously collecting the spectrum image of the area to be monitored is to collect the spectrum image of the area to be monitored in the process of flying according to the flight path by the spectrometer carried by the unmanned aerial vehicle.
Further, in S6, the method for acquiring the pest region of each spectral image in the identification sequence according to the RX algorithm includes: and inputting the spectral reflectivity of each pixel point in the spectral image into an RX (Reed-Xiaooli hyperspectral target detection algorithm) algorithm to identify the insect pest area at the corresponding position of the area to be monitored in the spectral image.
The remote sensing monitoring of the plant diseases and insect pests is judged by measuring the change of the chlorophyll content in the plant, and the spectral reflectivity of the chlorophyll has obvious characteristics and can change along with the change of the wavelength. The spectral reflectance of chlorophyll of plants is very low at 0.5-0.7 μm, and is significantly increased at 0.7-0.9 μm in the near infrared band, because green plants can absorb the radiant energy of this band. As chlorophyll in plant bodies of plant diseases and insect pests is gradually reduced, the light absorption capacity is weakened, the reflectivity of visible light is obviously improved, and the reflectivity of an infrared region is obviously reduced, especially near-infrared wave bands.
Preferably, the method for acquiring the pest region of each spectral image in the identification sequence according to the RX algorithm comprises the following steps: insect pest areas at corresponding positions in an unmanned aerial vehicle operation area are obtained through a method disclosed in patent publication No. CN107347849A (RX abnormal detection method).
The insect pest area is an area formed by the insect pest pixel points or an internal area of a margin line formed by the insect pest pixel points; the pest and disease pixel points are corresponding pixel points in the spectral image, wherein the visible light reflectivity of the pest and disease pixel points is higher than the average value of the visible light reflectivity of each pixel point in the spectral image, or the pest and disease pixel points are corresponding pixel points in the spectral image, wherein the near-infrared band spectral reflectivity of the pest and disease pixel points is lower than the average value of the near-infrared band spectral reflectivity of each pixel point in the spectral image.
Further, in S7, the reference image is an overall spectral image of the to-be-monitored area spliced by the spectral images acquired in the previous time interval in steps S1 to S6, and the locations of the insect pest areas identified in step S6 are marked on the corresponding locations of the to-be-monitored area in the overall spectral image. (the time interval is generally 7-30 days), wherein the spectral image stitching comprises the steps of geometric correction, image preprocessing, image registration and image fusion.
Preferably, in S7, the reference image is a spectral image obtained by an onboard full-spectrum multi-modal imaging spectrometer or a spectral image obtained by a hyperspectral remote sensing satellite, the pest region of the spectral image is identified according to an RX algorithm, and the pest region is marked on the pest region to be monitored at a position corresponding to the pest region of the spectral image.
Further, the unmanned aerial vehicle is a coaxial dual-rotor unmanned aerial vehicle, a micro rotor unmanned aerial vehicle or a multi-rotor unmanned aerial vehicle carrying a spectrometer; preferably, the spectrometer is a GaiaField-mini spectrometer, a specific AFX series hyperspectral camera or an ATH9020 hyperspectral imager.
Further, in S7, the method for correcting the location of the pest region on the reference image by the identification sequence includes the steps of:
marking the identification sequence as RLocal, taking each spectral image in the RLocal as an identification partition, wherein RLocal = { RL (i) }, i is the serial number of the identification partition, i belongs to [1, N1], N1 is the number of the identification partitions, and RL (i) is the ith identification partition in the RLocal;
in the value range of i, calculating Euclidean distances between the geometric gravity center point of each insect pest region in RL (i) and P1 (i) by taking the identification point corresponding to RL (i) as P1 (i), and then taking the mean value of all the Euclidean distances in RL (i) as RLmean (i); selecting a point of a corresponding position of the geometric gravity center point with the maximum distance value from the geometric gravity center point P1 (i) of each insect pest region in the RL (i) on the reference image as a far center point P2 (i), and selecting a point of a corresponding position of the geometric gravity center point with the minimum distance value from the geometric gravity center point P1 (i) of each insect pest region in the RL (i) on the reference image as a near center point P3 (i);
screening all the insect pest areas from the reference image, wherein a set formed by the insect pest areas with the distances from the geometric gravity center points of all the insect pest areas to P2 (i) being smaller than RLmean (i) is recorded as RLSUB (i); and/or screening out a set formed by insect pest regions, wherein the distances from the geometric gravity center points of all insect pest regions to P2 (i) are smaller than RLmean (i), and the mean value of the near infrared band spectral reflectances of all pixel points in the insect pest regions is lower than the mean value of the near infrared band spectral reflectances of all pixel points in the insect pest regions corresponding to remote center points P2 (i) in RL (i), and marking as RLSUB (i);
taking N2 as the number of elements in RLSUB (i), wherein RLSUB (i, j) is the geometric gravity center of the jth pest region in RLSUB (i), j is the serial number of the elements in RLSUB (i), and j belongs to [1, N2];
if N2=0, then record RL (i) as the identified partition without correction; and if N2 is larger than 0, correcting the position of the insect pest region inside the corresponding position of RL (i) on the reference image in the value range of i.
Above scheme can discern the offset position of identification point on insect pest area and the unmanned aerial vehicle airline, can discern the accurate relative position in insect pest area, guarantees the relative precision of insect pest position coordinate.
Further, the method for correcting the position of the insect pest area inside the position corresponding to the RL (i) on the reference image in the value range of i comprises the following steps:
in the value range of j, calculating the mean value of the distances between the far center point P2 (i) and each RLSUB (i, j) in the RLSUB (i) and recording the mean value as the far center distance AD; calculating the mean value of the distances between the near center point P3 (i) and each RLSUB (i, j) in the RLSUB (i) and recording the mean value as the near center distance BD; recording a point PF at a corresponding position on the reference image of the RLSUB (i, j) having the shortest distance between the remote center point P2 (i) and each RLSUB (i, j); a point at a corresponding position on the reference image of RLSUB (i, j) with the shortest distance between the center point P3 (i) and each RLSUB (i, j) is represented as PN;
when AD is larger than or equal to BD, taking the direction from the center point P2 (i) to PF as an adjusting direction, moving the insect region with the geometric center point closest to PF in each insect region on the reference image by a telecentric distance AD towards the adjusting direction,
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center of a circle and takes a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if the correction range A has insect damage areas except the insect damage area corresponding to P2 (i) in the corresponding area on RL (i), the insect damage areas on RL (i) are copied to the corresponding positions of the insect damage areas on the reference image.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B has insect pest regions except the insect pest region corresponding to P3 (i) in the corresponding region on RL (i), the insect pest regions on RL (i) are copied to the corresponding positions of the insect pest regions on the reference image.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center of a circle and takes a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if the correction range A does not have a pest region except the pest region corresponding to P2 (i) in the corresponding region on RL (i), the pest region in the correction range A on the reference image is deleted.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a radius of a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B does not have a pest region except the pest region corresponding to P3 (i) in the corresponding region on RL (i), the pest region in the correction range B on the reference image is deleted.
According to the technical scheme, the telecentric distance AD and the telecentric distance BD are offset distances from the geometric center of gravity of the insect pest region by taking the identification point as a reference, strong interference generated by accurate identification of the insect pest region when the insect pest region changes through the local part of the plant in the natural state can be removed through correction of the offset distances, identification accuracy of the insect pest region is improved, the current local image is processed, so that the identification speed is greatly improved without identifying the whole image, and the operation cost is reduced.
In order to reduce the influence of local change of the plant in the natural state on part of the non-pest area, and further improve the positioning and correction accuracy of the pest area, the invention provides the following preferable scheme:
preferably, the method for correcting the position of the pest region on the reference image within the value range of i comprises the following steps:
calculating the insect pest position deviation index dev (i) of RL (i), wherein the specific method comprises the following steps:
wherein the function of deviation degree
| P2 (i) -RLSUB (i, j) | is the distance between the far center point P2 (i) and RLSUB (i, j);
p3 (i) -RLSUB (i, j) | is the distance between the proximal point P3 (i) and RLSUB (i, j);
calculating pest position deviation indexes dev (i) of all RL (i) in the value range of i, calculating the average value of all dev (i) as meandev, and marking all identification partitions RL (i) of dev (i) being more than or equal to meandev as the to-be-optimized identification partitions;
when RL (i) is a partition to be identified optimally, calculating the mean value of the distances between a far center point P2 (i) and each RLSUB (i, j) in RLSUB (i) in the value range of j and recording as a telecentric distance AD; calculating the mean value of the distances between the near center point P3 (i) and each RLSUB (i, j) in the RLSUB (i) and recording the mean value as the near center distance BD; recording a point PF at a corresponding position on the reference image of the RLSUB (i, j) having the shortest distance between the remote center point P2 (i) and each RLSUB (i, j); a point at a corresponding position on the reference image of the RLSUB (i, j) with the shortest distance between the approximate center point P3 (i) and each RLSUB (i, j) is represented as PN;
when AD is larger than or equal to BD, taking the direction from the far center point P2 (i) to the PF as an adjusting direction, and moving the insect pest area with the geometric center point closest to the PF in each insect pest area on the reference image towards the adjusting direction by a far center distance AD;
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
When AD is larger than or equal to BD, taking the direction from the far center point P2 (i) to PF as an adjusting direction, moving the insect pest area with the geometric center point closest to PF in each insect pest area on the reference image by a far center distance AD towards the adjusting direction,
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center of a circle and takes a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if the correction range A has insect damage areas except the insect damage area corresponding to P2 (i) in the corresponding area on RL (i), the insect damage areas on RL (i) are copied to the corresponding positions of the insect damage areas on the reference image.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B has insect pest regions except the insect pest region corresponding to P3 (i) in the corresponding region on RL (i), the insect pest regions on RL (i) are copied to the corresponding positions of the insect pest regions on the reference image.
Preferably, when AD is larger than or equal to BD, a range which takes P2 (i) as a center and a connecting line from P2 (i) to PF as a radius on the reference image is taken as a correction range A, and if no pest region except the pest region corresponding to P2 (i) exists in the corresponding region of the correction range A on RL (i), the pest region in the correction range A on the reference image is deleted.
Preferably, when AD < BD, a range on the reference image with a center of P3 (i) as a center and a radius of a connecting line from P3 (i) to PN as a radius is taken as a correction range B, and if the correction range B does not have a pest region except the pest region corresponding to P3 (i) in the corresponding region on RL (i), the pest region in the correction range B on the reference image is deleted.
According to the preferred scheme, the insect pest position deviation index is the deviation degree between the far-center point and the near-center point of the judged insect pest position and the identification point, so that the difference between the positions of the non-insect pest area and the insect pest area is judged in a balanced manner, the insect pest area can be increased or deleted according to the difference on the reference image intelligently, the position of the insect pest area is adjusted accurately, and the influence on the insect pest area when the local part of a part of the non-insect pest area passing through the plant in the natural state changes is reduced.
The system for remotely sensing, monitoring and surveying the forest harmful organisms by the unmanned aerial vehicle provided by the embodiment of the invention is shown in fig. 2, and comprises: a processor, a memory and a computer program stored in the memory and operable on the processor, the processor when executing the computer program implementing the steps in an embodiment of the method for remotely sensing and monitoring by unmanned aerial vehicle for investigating forestry pests, the processor executing the computer program to operate in the units of the following system:
the map acquisition unit is used for acquiring an electronic map of the forest land to be monitored as an area to be monitored;
the route setting unit is used for setting the information of a starting point, an end point and a barrier position of the unmanned aerial vehicle flying in the area to be monitored, and setting a route path of the unmanned aerial vehicle from the starting point to the end point in the area to be monitored;
the identification dividing unit is used for taking a plurality of points on the route of the flight as identification points from a starting point at a fixed acquisition distance;
the system comprises a spectrum acquisition unit, a monitoring unit and a monitoring unit, wherein the spectrum acquisition unit is used for enabling the unmanned aerial vehicle to fly along a route path and continuously acquire spectrum images of an area to be monitored, a sequence formed by various spectrum images is used as a spectrum image sequence in sequence, and a spectrum image is acquired at each identification point in the spectrum image sequence;
the identification extraction unit is used for acquiring a sequence formed by all the spectral images at all the identification point positions in the spectral image sequence as an identification sequence;
the pest identification unit is used for acquiring pest areas of the spectral images in the identification sequence according to an RX algorithm;
the pest correction unit is used for correcting the position of the pest region on the reference image through the identification sequence;
and the map marking unit is used for marking the insect pest area position on the corrected reference image on the electronic map of the forest land to be monitored.
The system for remotely sensing, monitoring and surveying forestry pests by using the unmanned aerial vehicle comprises: the system for remotely sensing and surveying the forestry pests by the unmanned aerial vehicle can be operated in computing equipment such as a desktop computer, a notebook computer, a palm computer, a cloud data center and the like, and the operable system can comprise, but is not limited to, a processor, a memory and a server cluster.
The system for remotely sensing, monitoring and surveying the forestry pests by the unmanned aerial vehicle can be operated in computing equipment such as a desktop computer, a notebook computer, a palm computer and a cloud data center. The system for remotely sensing, monitoring and surveying forestry pests by the unmanned aerial vehicle comprises, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that the examples are merely illustrative of a method for remotely monitoring and surveying forestry pests by a drone, and do not constitute a limitation on a method for remotely monitoring and surveying forestry pests by a drone, and may include more or less components than the drone, or some components in combination, or different components, for example, the system for remotely monitoring and surveying forestry pests by a drone may further include an input-output device, a network access device, a bus, etc.
Preferably, according to the system for remotely sensing, monitoring and surveying the forest harmful organisms by the unmanned aerial vehicle provided by the embodiment of the invention, as shown in fig. 3, an electronic map of the forest land to be monitored according to the embodiment of the invention is provided; fig. 4 shows that the map of the pest region in the electronic map of the forest land to be monitored after the system for remotely sensing, monitoring and investigating forest pests by an unmanned aerial vehicle identifies according to an embodiment of the present invention, so that through multiple tests, the identification rate of the embodiment of the present invention on the pest region of the forest pest can reach more than 95%, and through comparison with manual identification, the position accuracy of the pest region is high.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete component Gate or transistor logic, discrete hardware components, etc. The general processor can be a microprocessor or the processor can also be any conventional processor and the like, the processor is a control center of the system for remotely sensing and surveying the forest harmful organisms by the unmanned aerial vehicle, and various interfaces and lines are utilized to connect various subareas of the whole system for remotely sensing and surveying the forest harmful organisms by the unmanned aerial vehicle.
The memory can be used for storing the computer program and/or the module, and the processor realizes various functions of the method for remotely sensing and surveying forestry pests by the unmanned aerial vehicle by operating or executing the computer program and/or the module stored in the memory and calling data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Although the present invention has been described in considerable detail and with reference to certain illustrated embodiments, it is not intended to be limited to any such details or embodiments or any particular embodiment, so as to effectively encompass the intended scope of the invention. Furthermore, the foregoing describes the invention in terms of embodiments foreseen by the inventor for which an enabling description was available, notwithstanding that insubstantial modifications of the invention, not presently foreseen, may nonetheless represent equivalent modifications thereto.
Claims (7)
1. A method for remotely sensing and surveying forestry pests by an unmanned aerial vehicle is characterized by comprising the following steps:
s1: acquiring an electronic map of a forest land to be monitored as an area to be monitored;
s2: setting the information of a starting point, a terminal point and a barrier position of the unmanned aerial vehicle flying in the area to be monitored, and setting a route path of the unmanned aerial vehicle from the starting point to the terminal point in the area to be monitored;
s3: starting from a starting point, taking a plurality of points on an underway path as identification points at a fixed acquisition distance;
s4: the unmanned aerial vehicle flies according to a route path and continuously collects spectrum images of an area to be monitored, a sequence formed by various spectrum images is used as a spectrum image sequence in sequence, and a spectrum image is obtained at each identification point in the spectrum image sequence;
s5: acquiring a sequence formed by all spectral images at all identification point positions in the spectral image sequence as an identification sequence;
s6: acquiring insect pest areas of each spectral image in the identification sequence according to an RX algorithm;
s7: correcting the position of the insect pest area on the reference image through the identification sequence;
s8: and marking the positions of the insect pest areas on the corrected reference image on the electronic map of the forest land to be monitored.
2. The method for remotely sensing and surveying the forestry pests by the unmanned aerial vehicle according to claim 1, wherein in S6, the method for acquiring the pest areas of each spectral image in the identification sequence according to the RX algorithm comprises the following steps: inputting the spectral reflectivity of each pixel point in the spectral image into an RX algorithm to identify the insect pest region at the corresponding position of the region to be monitored in the spectral image; the insect pest area is an area formed by the insect pest pixel points or an internal area of an edge line formed by the insect pest pixel points; the pest and disease pixel points are corresponding pixel points in the spectral image, wherein the visible light reflectivity of the pest and disease pixel points is higher than the average value of the visible light reflectivity of each pixel point in the spectral image, or the pest and disease pixel points are corresponding pixel points in the spectral image, wherein the near-infrared band spectral reflectivity of the pest and disease pixel points is lower than the average value of the near-infrared band spectral reflectivity of each pixel point in the spectral image.
3. The method for remotely sensing and monitoring by unmanned aerial vehicle and investigating forestry harmful organisms according to claim 1, wherein the unmanned aerial vehicle is a coaxial dual-rotor unmanned aerial vehicle, a micro-rotor unmanned aerial vehicle or a multi-rotor unmanned aerial vehicle carrying a spectrometer; the spectrometer is a GaiaField-mini spectrometer, a specific AFX series hyperspectral camera or an ATH9020 hyperspectral imager.
4. The method for remotely sensing and surveying forestry pests by an unmanned aerial vehicle according to claim 1, wherein in S7, the method for correcting the positions of pest areas on the reference image through the identification sequence comprises the following steps:
marking the identification sequence as RLocal, taking each spectral image in the RLocal as an identification partition, wherein RLocal = { RL (i) }, i is the serial number of the identification partition, i belongs to [1, N1], N1 is the number of the identification partitions, and RL (i) is the ith identification partition in the RLocal;
in the value range of i, calculating Euclidean distances between the geometric gravity center point of each insect pest region in RL (i) and P1 (i) by taking the identification point corresponding to RL (i) as P1 (i), and then taking the mean value of all the Euclidean distances in RL (i) as RLmean (i); selecting a point of a corresponding position of the geometric gravity center point with the maximum distance value from the geometric gravity center point P1 (i) of each insect pest region in the RL (i) on the reference image as a far center point P2 (i), and selecting a point of a corresponding position of the geometric gravity center point with the minimum distance value from the geometric gravity center point P1 (i) of each insect pest region in the RL (i) on the reference image as a near center point P3 (i);
screening all pest regions from the reference image, wherein a set formed by pest regions with distances from geometric gravity center points of all pest regions to P2 (i) smaller than RLmean (i) is marked as RLSUB (i); and/or screening out a set formed by insect pest regions, wherein the distances from the geometric gravity center points of all insect pest regions to P2 (i) are smaller than RLmean (i), and the mean value of the near infrared band spectral reflectances of all pixel points in the insect pest regions is lower than the mean value of the near infrared band spectral reflectances of all pixel points in the insect pest regions corresponding to remote center points P2 (i) in RL (i), and marking as RLSUB (i);
taking N2 as the number of elements in RLSUB (i), wherein RLSUB (i, j) is the geometric gravity center of the jth pest region in RLSUB (i), j is the serial number of the elements in RLSUB (i), and j belongs to [1, N2];
if N2=0, then RL (i) is noted as the identified partition that does not require correction; and if N2 is more than 0, correcting the position of the insect pest area inside the corresponding position of RL (i) on the reference image in the value range of i.
5. The method for remotely sensing, monitoring and surveying the forestry pests by the unmanned aerial vehicle according to claim 4, wherein the method for correcting the positions of the pest areas inside the positions corresponding to RL (i) on the reference image within the value range of i comprises the following steps:
in the value range of j, calculating the mean value of the distances between the far center point P2 (i) and each RLSUB (i, j) in the RLSUB (i) and recording the mean value as the far center distance AD; calculating the mean value of the distance between the near-center point P3 (i) and each RLSUB (i, j) in the RLSUB (i) as the near-center distance BD; recording a point PF at a corresponding position on the reference image of the RLSUB (i, j) having the shortest distance between the remote center point P2 (i) and each RLSUB (i, j); a point at a corresponding position on the reference image of the RLSUB (i, j) with the shortest distance between the approximate center point P3 (i) and each RLSUB (i, j) is represented as PN;
when AD is larger than or equal to BD, taking the direction from the far center point P2 (i) to PF as an adjusting direction, moving the insect pest area with the geometric center point closest to PF in each insect pest area on the reference image by a far center distance AD towards the adjusting direction,
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
6. The method for remotely sensing and surveying the forestry pests by the unmanned aerial vehicle according to claim 4, wherein the method for correcting the positions of the pest areas on the reference image within the value range of i comprises the following steps:
calculating the insect pest position deviation index dev (i) of RL (i), wherein the specific method comprises the following steps:
wherein the function of deviation degree
| P2 (i) -RLSUB (i, j) | is the distance between the far center point P2 (i) and RLSUB (i, j);
| P3 (i) -RLSUB (i, j) | is the distance between the centroids P3 (i) and RLSUB (i, j);
calculating pest and disease damage position deviation indexes dev (i) of all RL (i) in the value range of i, calculating the average value of all dev (i) as meandev, and marking all identification partitions RL (i) of which dev (i) is more than or equal to meandev as to-be-optimized identification partitions;
when RL (i) is the partition to be identified for optimization, calculating the mean value of the distances between the far center point P2 (i) and each RLSUB (i, j) in RLSUB (i) in the range of the value of j and recording the mean value as the far center distance AD; calculating the mean value of the distance between the near-center point P3 (i) and each RLSUB (i, j) in the RLSUB (i) as the near-center distance BD; recording a point PF at a corresponding position on the reference image of the RLSUB (i, j) having the shortest distance between the remote center point P2 (i) and each RLSUB (i, j); a point at a corresponding position on the reference image of the RLSUB (i, j) with the shortest distance between the approximate center point P3 (i) and each RLSUB (i, j) is represented as PN;
when AD is larger than or equal to BD, taking the direction from the center point P2 (i) to PF as an adjusting direction, and moving the insect pest area with the geometric center point closest to PF in each insect pest area on the reference image towards the adjusting direction by a telecentric distance AD;
when AD < BD, the direction from the near center point P3 (i) to PN is used as the adjusting direction, and the insect damage area with the geometric center point closest to the PN in each insect damage area on the reference image is moved by the near center distance BD in the adjusting direction.
7. The utility model provides a system for forestry harmful organism is surveyed in unmanned aerial vehicle remote sensing monitoring, its characterized in that, a system for forestry harmful organism is surveyed in unmanned aerial vehicle remote sensing monitoring includes: a processor, a memory and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the method for remotely sensing and surveying forestry pests by an unmanned aerial vehicle according to any one of claims 1 to 6 when executing the computer program, and the system for remotely sensing and surveying forestry pests by an unmanned aerial vehicle runs in a computing device of a desktop computer, a notebook computer, a palm computer or a cloud data center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310029641.9A CN115909113B (en) | 2023-01-09 | 2023-01-09 | Method for investigating forestry harmful organisms through unmanned aerial vehicle remote sensing monitoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310029641.9A CN115909113B (en) | 2023-01-09 | 2023-01-09 | Method for investigating forestry harmful organisms through unmanned aerial vehicle remote sensing monitoring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115909113A true CN115909113A (en) | 2023-04-04 |
| CN115909113B CN115909113B (en) | 2023-06-16 |
Family
ID=86497082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310029641.9A Active CN115909113B (en) | 2023-01-09 | 2023-01-09 | Method for investigating forestry harmful organisms through unmanned aerial vehicle remote sensing monitoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115909113B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115662A (en) * | 2023-09-25 | 2023-11-24 | 中国科学院空天信息创新研究院 | Jujube tree spider mite pest identification method and system |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7653218B1 (en) * | 2006-05-02 | 2010-01-26 | Orbimage Si Opco, Inc. | Semi-automatic extraction of linear features from image data |
| CN103364781A (en) * | 2012-04-11 | 2013-10-23 | 南京财经大学 | Remote sensing data and geographical information system-based grainfield ground reference point screening method |
| CN103456011A (en) * | 2013-09-02 | 2013-12-18 | 杭州电子科技大学 | Improved hyperspectral RX abnormal detection method by utilization of complementary information |
| CN104266982A (en) * | 2014-09-04 | 2015-01-07 | 浙江托普仪器有限公司 | Large-area insect pest quantization monitoring system |
| CN106915462A (en) * | 2017-02-14 | 2017-07-04 | 福建兴宇信息科技有限公司 | Forestry pests ＆ diseases intelligent identifying system based on multi-source image information |
| CN107347849A (en) * | 2017-07-18 | 2017-11-17 | 河海大学 | A kind of brainpower insufflation system based on EO-1 hyperion real-time detection technology |
| CN108693119A (en) * | 2018-04-20 | 2018-10-23 | 北京麦飞科技有限公司 | Pest and disease damage based on unmanned plane high-spectrum remote-sensing intelligently examines the system of beating |
| US20210027429A1 (en) * | 2019-07-26 | 2021-01-28 | Zhejiang University Of Technology | Noise detection method for time-series vegetation index derived from remote sensing images |
| US20210166019A1 (en) * | 2019-11-29 | 2021-06-03 | RxMaker, Inc. | Enhanced Management Zones for Precision Agriculture |
| CN113378912A (en) * | 2021-06-08 | 2021-09-10 | 长光卫星技术有限公司 | Forest area illegal reclamation land block detection method based on deep learning target detection |
| CN114694023A (en) * | 2022-03-18 | 2022-07-01 | 河南农业大学 | Cotton aphid severity grading method based on spectral index data reconstruction |
| CN114936971A (en) * | 2022-06-08 | 2022-08-23 | 浙江理工大学 | Unmanned aerial vehicle remote sensing multispectral image splicing method and system for water area |
| CN115131683A (en) * | 2022-08-25 | 2022-09-30 | 金乡县林业保护和发展服务中心(金乡县湿地保护中心、金乡县野生动植物保护中心、金乡县国有白洼林场) | Forestry information identification method based on high-resolution remote sensing image |
| CN115311588A (en) * | 2022-09-16 | 2022-11-08 | 航天宏图信息技术股份有限公司 | Pine wood nematode disease stumpage detection method and device based on unmanned aerial vehicle remote sensing image |
| WO2022257139A1 (en) * | 2021-06-11 | 2022-12-15 | 深圳市大疆创新科技有限公司 | Plant state determination method, terminal, and computer-readable storage medium |
-
2023
- 2023-01-09 CN CN202310029641.9A patent/CN115909113B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7653218B1 (en) * | 2006-05-02 | 2010-01-26 | Orbimage Si Opco, Inc. | Semi-automatic extraction of linear features from image data |
| CN103364781A (en) * | 2012-04-11 | 2013-10-23 | 南京财经大学 | Remote sensing data and geographical information system-based grainfield ground reference point screening method |
| CN103456011A (en) * | 2013-09-02 | 2013-12-18 | 杭州电子科技大学 | Improved hyperspectral RX abnormal detection method by utilization of complementary information |
| CN104266982A (en) * | 2014-09-04 | 2015-01-07 | 浙江托普仪器有限公司 | Large-area insect pest quantization monitoring system |
| CN106915462A (en) * | 2017-02-14 | 2017-07-04 | 福建兴宇信息科技有限公司 | Forestry pests ＆ diseases intelligent identifying system based on multi-source image information |
| CN107347849A (en) * | 2017-07-18 | 2017-11-17 | 河海大学 | A kind of brainpower insufflation system based on EO-1 hyperion real-time detection technology |
| CN108693119A (en) * | 2018-04-20 | 2018-10-23 | 北京麦飞科技有限公司 | Pest and disease damage based on unmanned plane high-spectrum remote-sensing intelligently examines the system of beating |
| US20210027429A1 (en) * | 2019-07-26 | 2021-01-28 | Zhejiang University Of Technology | Noise detection method for time-series vegetation index derived from remote sensing images |
| US20210166019A1 (en) * | 2019-11-29 | 2021-06-03 | RxMaker, Inc. | Enhanced Management Zones for Precision Agriculture |
| CN113378912A (en) * | 2021-06-08 | 2021-09-10 | 长光卫星技术有限公司 | Forest area illegal reclamation land block detection method based on deep learning target detection |
| WO2022257139A1 (en) * | 2021-06-11 | 2022-12-15 | 深圳市大疆创新科技有限公司 | Plant state determination method, terminal, and computer-readable storage medium |
| CN114694023A (en) * | 2022-03-18 | 2022-07-01 | 河南农业大学 | Cotton aphid severity grading method based on spectral index data reconstruction |
| CN114936971A (en) * | 2022-06-08 | 2022-08-23 | 浙江理工大学 | Unmanned aerial vehicle remote sensing multispectral image splicing method and system for water area |
| CN115131683A (en) * | 2022-08-25 | 2022-09-30 | 金乡县林业保护和发展服务中心(金乡县湿地保护中心、金乡县野生动植物保护中心、金乡县国有白洼林场) | Forestry information identification method based on high-resolution remote sensing image |
| CN115311588A (en) * | 2022-09-16 | 2022-11-08 | 航天宏图信息技术股份有限公司 | Pine wood nematode disease stumpage detection method and device based on unmanned aerial vehicle remote sensing image |
Non-Patent Citations (5)
| Title |
|---|
| NESREEN M. ABD EL-GHANY等: "A review: application of remote sensing as a promising strategy for insect pests and diseases management", 《 ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH》, pages 33503 * |
| RAN TAO等: "Hyperspectral Anomaly Detection by Fractional Fourier Entropy", 《IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 》, vol. 12, no. 12, pages 4920, XP011770035, DOI: 10.1109/JSTARS.2019.2940278 * |
| 孙岩: "湿地植物高光谱特征分析与物种识别模型构建", 《中国优秀硕士学位论文全文数据库 基础科学辑》, pages 006 - 203 * |
| 孙红;李民赞;周志艳;刘刚;罗锡文;: "基于光谱技术的水稻稻纵卷叶螟受害区域检测", 光谱学与光谱分析, no. 04, pages 1080 - 1083 * |
| 赵春晖;王玉磊;李晓慧;: "一种新型高光谱实时异常检测算法", 红外与毫米波学报, no. 01, pages 114 - 121 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115662A (en) * | 2023-09-25 | 2023-11-24 | 中国科学院空天信息创新研究院 | Jujube tree spider mite pest identification method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115909113B (en) | 2023-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Osco et al. | A CNN approach to simultaneously count plants and detect plantation-rows from UAV imagery | |
| US10614562B2 (en) | Inventory, growth, and risk prediction using image processing | |
| Kamilaris et al. | Deep learning in agriculture: A survey | |
| Gai et al. | Automated crop plant detection based on the fusion of color and depth images for robotic weed control | |
| Senthilnath et al. | Detection of tomatoes using spectral-spatial methods in remotely sensed RGB images captured by UAV | |
| Bah et al. | Deep learning based classification system for identifying weeds using high-resolution UAV imagery | |
| Calou et al. | The use of UAVs in monitoring yellow sigatoka in banana | |
| CN113591766B (en) | Multi-source remote sensing tree species identification method for unmanned aerial vehicle | |
| CN110148116A (en) | A kind of forest biomass evaluation method and its system | |
| Diago et al. | On‐the‐go assessment of vineyard canopy porosity, bunch and leaf exposure by image analysis | |
| CN110889394A (en) | Rice lodging recognition method based on deep learning UNet network | |
| Zortea et al. | Automatic citrus tree detection from UAV images based on convolutional neural networks | |
| Ramesh et al. | Detection of rows in agricultural crop images acquired by remote sensing from a UAV | |
| Starý et al. | Comparing RGB-based vegetation indices from UAV imageries to estimate hops canopy area | |
| CN115909113A (en) | Method for surveying forestry pests through remote sensing monitoring of unmanned aerial vehicle | |
| Pádua et al. | Vineyard classification using machine learning techniques applied to RGB-UAV imagery | |
| Mousavi et al. | A novel enhanced vgg16 model to tackle grapevine leaves diseases with automatic method | |
| OZDARICI-OK et al. | Using remote sensing to identify individual tree species in orchards: A review | |
| CN111815612A (en) | Red date disease and pest prediction system based on Internet of things | |
| Zou et al. | The fusion of satellite and unmanned aerial vehicle (UAV) imagery for improving classification performance | |
| Dehkordi et al. | Performance Evaluation of Temporal and Spatial-Temporal Convolutional Neural Networks for Land-Cover Classification (A Case Study in Shahrekord, Iran) | |
| KR20220106639A (en) | Method and System for Detecting Aquatic Plants in Small Reservoirs using Multispectral UAV Imagery and Vegetation index | |
| Bonaria | Grapevine yield estimation using image analysis for the variety Arinto | |
| Paul et al. | Smart Agriculture Using UAV and Deep Learning: A Systematic Review | |
| Marin et al. | Individual Olive Tree Detection in RGB Images |
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 |