CN112633161A - Pine wood nematode disease withered and dead tree detection and positioning method based on high-altitude pan-tilt recognition - Google Patents

Abstract

The invention discloses a detection and positioning method for pine wilt disease and dead trees based on high-altitude pan-tilt recognition, which is carried out according to the following steps: determining a detection target area, acquiring the surrounding environment in the detection target area by adopting a high-altitude pan-tilt camera, and generating PTZ scanning information; according to the PTZ scanning information, positioning the photographing position of the camera to generate an image to be identified; the high-altitude pan-tilt head carries out CNN model identification on the images to be identified in the grab image queue Q1, and suspicious images are extracted; the high-altitude cloud deck controls the camera to shoot and re-identify the corresponding position according to the suspicious image, and locks the PTZ coordinate position of the dead tree in the suspicious image; and determining longitude and latitude information of a corresponding camera according to the PTZ coordinate position of the withered and dead tree in the suspicious image, and calculating the longitude and latitude of the geographic position of the withered and dead tree. Has the advantages that: the high-altitude shooting, image recognition and accurate positioning of the tree withered and dead due to the pine wilt disease are realized, and the direction is guided for patrol.

Description

Pine wood nematode disease withered and dead tree detection and positioning method based on high-altitude pan-tilt recognition

Technical Field

The invention relates to the technical field of forest protection, in particular to a pine wilt disease dying tree detection and positioning method based on high-altitude pan-tilt recognition

Background

Pine wood nematode disease, also known as pine wilt disease, is a devastating forest disease caused by pine wood nematodes (Bursaphelenchus xylophilus), belongs to a major foreign invasive species in China, and has been listed as an internal and external forest plant quarantine object in China. The disease is rapidly diffused and spread since 1982, more than 14 provinces (cities and regions) occur in the whole country at present, the area reaches 7.7 hectares, so that a large number of pines die, pine forest resources, natural landscapes and ecological environments in China are seriously damaged, and serious economic and ecological losses are caused. Pine wood nematodes have become the first dangerous pest in forestry in China.

The diseased plants are discovered and felled as early as possible, the pine wood nematode disease can be prevented from expanding and spreading, and the damage of the disease to the ecological environment and the brought economic loss can be greatly reduced. At present, the method for detecting and treating the pine wood nematode disease in China generally comprises the steps of manually searching for diseased pine trees in a pine forest and then carrying out treatments such as pesticide application, cutting and the like on the pine trees, but most of the dead pine trees are firstly planted at the top ends of the pine trees, due to vegetation shielding, the pine trees are widely spread after artificial discovery, and many places are rare, so that the feasibility is low.

Some research institutions have used unmanned aerial vehicles to patrol the forest farm and have identified it using corresponding algorithms. However, the unmanned aerial vehicle has a short dead time and a limited patrol range, and still needs to manually process almost all the whole processes including the processes of take-off, return, path planning, video reading, uploading analysis and the like, so that the cost is quite high.

Therefore, the prior art has the disadvantages that: the method is lack of a technology which is cheap, timely, recyclable and less in artificial participation, and is convenient for detecting and positioning the existing tree which is withered and killed by the pine wood nematode.

Disclosure of Invention

Aiming at the problems, the invention provides a pine wood nematode disease dead tree detection and positioning method based on high-altitude pan-tilt recognition, which utilizes a high-altitude pan-tilt and a related camera to shoot forest trees at a long distance, and combines an image recognition technology and image characteristics of the pine wood nematode disease to recognize and position the dead trees, thereby being beneficial to forest maintenance personnel to control related diseases in time and preventing the pine wood nematode disease in a large area.

In order to achieve the purpose, the invention adopts the following specific technical scheme:

a pine wood nematode disease tree withering and dying detection and positioning method based on high-altitude pan-tilt recognition is characterized by comprising the following steps:

s1: determining a detection target area, acquiring the surrounding environment in the detection target area by adopting a high-altitude pan-tilt camera, and generating PTZ scanning information;

wherein P represents a horizontal angle value; t represents a vertical angle value; z represents the magnification value of the camera;

s2: according to the PTZ scanning information, positioning the photographing position of the camera to generate an image to be identified;

s3: the high-altitude pan-tilt head carries out CNN model identification on the images to be identified in the grab image queue Q1, and suspicious images are extracted;

s4: the high-altitude cloud deck controls the camera to shoot and re-identify the corresponding position according to the suspicious image, and locks the PTZ coordinate position of the dead tree in the suspicious image;

s5: and determining longitude and latitude information of a corresponding camera according to the PTZ coordinate position of the withered and dead tree in the suspicious image obtained in the S4, and calculating the longitude and latitude of the geographic position of the withered and dead tree.

Through the design, the existing forestry high-altitude fire prevention holder is fully utilized, the pine wood nematode withered and dead tree is automatically identified and positioned, any newly-added equipment is not required to be added at the front end, the problem of patrol and protection of the pine wood nematode withered and dead tree in a forest region from a rare person can be solved, and the purposes of saving manpower and material resources and protecting ecology by using an algorithm are really realized.

The further technical scheme is as follows: the specific content of the PTZ scanning information generated in step S1 is:

setting a starting PTZ value and an ending PTZ value of the high-altitude pan-tilt camera to be scanned, namely setting P_start、T_start、Z_start、P_end、T_end、Z_end；

Setting a camera photographing positioning value according to a stepping comparison lookup table of the Z value and the PT value between the initial PTZ value and the end PTZ value, and controlling the camera to photograph according to the photographing positioning value; sequentially putting the photographing positioning values into a photographing positioning array aP 1;

wherein, the step comparison lookup table of the Z value and the PT value is detailed in a table I;

z value range (Z)0-Zx)	δP	δT
			1-3	p1	t1
3-8	p2	t3
			8-12	p3	t4
12-15	p4	t5
			15-20	p5	t6
20-30	p6	t7
			Greater than 30	p7	t8

Setting a photographing positioning array aP1 as a null array, and setting initial values of a horizontal angle value P, a vertical angle value T and a camera magnification numerical value Z; let P₀＝P_start，T₀＝T_start，Z₀＝Z_start1 is ═ 1; and P is₀＝P_start，T₀＝T_start，Z₀＝Z_startPut into the photo location array aP 1;

according to the current magnification numerical value Z of the camera, the step-by-step comparison search of the Z value and the PT value is carried outFinding out a corresponding horizontal step value delta P and a corresponding vertical step value delta T from the table; and obtaining P_n＝P_n-1+ δ P and T_n＝T_n-1+ δ T, so that the camera is in the current photographing position; the value of n is an integer between 1 and x;

if P_n<P_startThen let P_n-1＝P_startOr P is_n>P_endLet P_n-1＝P_end(ii) a At this time, P_n＝P_n-1+δP’，δP’＝-δP；

If T<T_tartThen let T_n-1＝T_endOr T is>T_endThen let T_n＝T _start；

All the horizontal angle values P, the vertical angle values T, and the camera magnification power values Z, which are numbered 0 to x, are grouped into the photo location array aP1, and the final photo location array aP1 is made to be PTZ scanning information.

By adopting the scheme, the p and t values in the stepping comparison lookup table of the Z value and the PT value are different aiming at different regions, and progressive ptz scanning information is formed according to region self-adaptive setting.

Still further, the step S2 of generating the image to be recognized includes the following specific steps:

s21: the high-altitude pan-tilt head sequentially reads the PTZ values in the photographing positioning array aP1, and positions the camera to a photographing position corresponding to the appointed PTZ value by using a camera API of the camera;

s22: after the camera reaches any photographing position in the photographing positioning array aP1, pausing for m seconds, capturing an image to be recognized by using a camera API (application program interface), and placing the image to be recognized into a capturing queue Q1;

s23: controlling the camera to reach the next photographing position in the photographing positioning array aP1, and repeating the step S22;

adopt above-mentioned scheme to realize the preliminary image acquisition to withered and dead trees to set up the PTZ value through different position direction and distance, to same piece region, generally less step value changes the PTZ value, realizes the collection of all images in same piece region.

The further technical solution is that the specific steps of CNN recognition of the image to be recognized and extracting the suspicious image in step S3 are as follows:

the high-altitude pan-tilt head sequentially acquires images to be identified from the image capture queue Q1, and puts the images into a CNN (hidden network) model for feature identification to obtain an identification result; the identification result at least comprises frame selection content of the image and identification data of withered and dead trees; the identification data of the withered trees are as follows:

aBoxs＝[{x₁₁,y₁₁,x₁₂,y₁₂},{x₂₁,y₂₁,x₂₂,y₂₂},...,{x_i1,y_i1,x_i2,y_i2},...]；

wherein x_i1，y_i1Representing that the identified tree i corresponds to the coordinates of the upper left corner of the image to be identified; x is the number of_i2,y_i2Representing that the identified tree i corresponds to the lower right corner coordinate of the image to be identified;

if the withered trees exist in the identification result of the image to be identified, putting the identification result into a withered tree array R1, and taking the image as a suspicious image; otherwise the image is discarded.

In the case of a tree which is withered and dead by the pine wood nematode disease, the tree is generally characterized in that the trunk and bark part of the tree are fallen off due to leaf yellowing, the tree generally appears in groups along with a plurality of diseased trees or dead trees in the process of overhead shooting, and wood chips and barks appear on the root of the tree, so that the characteristic can be used for automatic identification from far to near in the process of characteristic identification. And acquiring the suspicious picture.

In a further technical scheme, the step S4 includes the following steps: s41: the high altitude cloud deck obtains withered and dead tree identification data corresponding to each suspicious image from withered and dead tree array R1 in sequence, namely:

aBoxs＝[{x₁₁,y₁₁,x₁₂,y₁₂},{x₂₁,y₂₁,x₂₂,y₂₂},...,{x_i1,y_i1,x_i2,y_i2},...]；

s42: the high-altitude tripod head sequentially takes the elements i: { x:_i1,y_i1,x_i2,y_i2using a cameraThe 3D positioning API sends a positioning command to the camera;

s43: the high-altitude pan-tilt control camera API captures images and stores the re-captured images;

s44: the high-altitude pan-tilt acquires a PTZ value according to the current camera API; and storing the image and the rephotograph image into a rephotograph queue Q2;

s45: the high-altitude pan-tilt head sequentially obtains the rephoto snap-shot images from the rephoto snap-shot queue Q2, puts the images into a CNN model for feature recognition to obtain the recognition result of the rephoto images, and puts the recognition result into the array R2 for re-recognizing dead trees; the identification result of the rephotograph comprises rephotograph frame selection content and identification data of dead trees;

s46: if withered trees exist in the recognition result of the double-shot snap-shot image, locking the PTZ coordinate positions of the withered trees in the suspicious image; otherwise, discarding the identification data of the withered and dead trees corresponding to the suspicious image;

s47: judging whether the identification data of the withered and dead trees in the suspicious image is identified again, if so, entering step S48; otherwise, identifying the identification data of the next withered tree in the suspicious image, and returning to the step S41;

s48: judging whether all the suspicious images are identified, if not, locking the next suspicious image, and returning to the step S41; if the identification is completed, all PTZ coordinate positions are output to step S5.

By adopting the steps, the suspicious images are further confirmed, the shooting accuracy is improved by adjusting the shooting angle and the shooting magnification factor in a rephotography mode, and the suspicious images are screened for the second time so as to reduce the workload of the patrol personnel.

Still further technical solution is that the specific steps of step S5 are:

s51: sequentially reading PTZ coordinates in the re-identified dead tree array R2, and converting the PTZ coordinates in the re-identified dead tree array R2 into a warp and weft value of the dead tree by using the longitude and the weft of the camera as an origin of coordinates by using a geotools;

s52: correspondingly storing the longitude and latitude values of the withered and dead trees in an array R2 for re-identifying the withered and dead trees one by one to obtain coordinate data R3 of the withered and dead trees;

s53: and judging whether all the deadly tree arrays R2 are completely converted, if so, outputting deadly tree coordinate data R3, and otherwise, returning to the step S51.

In the steps, the target tree is accurately positioned to obtain the corresponding longitude and latitude value, and the patrolman can carry out directional patrolling according to the data.

The invention has the beneficial effects that: the invention effectively utilizes the high-altitude cradle head to realize high-altitude shooting, image recognition and accurate positioning of the tree withered and dead due to the pine wilt disease, and guides the direction for patrol.

Drawings

FIG. 1 is a flow chart of the method of the present invention;

FIG. 2 is a flow chart of the high altitude pan/tilt/zoom camera shooting;

fig. 3 is a flowchart of the steps of the high altitude cloud platform calibrating and positioning the suspicious picture.

Detailed Description

The following provides a more detailed description of the embodiments and the operation of the present invention with reference to the accompanying drawings.

A detection and positioning method for pine wilt and dead trees based on high-altitude pan-tilt recognition is disclosed, and can be seen from figure 1, and is carried out according to the following steps:

s1: determining a detection target area, acquiring the surrounding environment in the detection target area by adopting a high-altitude pan-tilt camera, and generating PTZ scanning information;

wherein P represents a horizontal angle value; t represents a vertical angle value; z represents the magnification value of the camera;

the specific content of the PTZ scanning information generated in step S1 is:

setting a starting PTZ value and an ending PTZ value of the high-altitude pan-tilt camera to be scanned, namely setting P_start、T_start、Z_start、P_end、T_end、Z_end；

Setting a camera photographing positioning value according to a stepping comparison lookup table of the Z value and the PT value between the initial PTZ value and the end PTZ value, and controlling the camera to photograph according to the photographing positioning value; sequentially putting the photographing positioning values into a photographing positioning array aP 1;

in this embodiment, the step comparison lookup table of the Z value and the PT value is detailed in table one;

setting a photographing positioning array aP1 as a null array, and setting initial values of a horizontal angle value P, a vertical angle value T and a camera magnification numerical value Z; let P₀＝P_start，T₀＝T_start，Z₀＝Z_start1 is ═ 1; and P is₀＝P_start，T₀＝T_start，Z₀＝Z_startPut into the photo location array aP 1;

according to the current magnification numerical value Z of the camera, finding out a corresponding horizontal stepping value delta P and a corresponding vertical stepping value delta T from a stepping comparison lookup table of the Z value and the PT value; and obtaining P_n＝P_n-1+ δ P and T_n＝T_n-1+ δ T, so that the camera is in the current photographing position; the value of n is an integer between 1 and x;

if P_n<P_startThen let P_n-1＝P_startOr P is_n>P_endLet P_n-1＝P_end(ii) a At this time, P_n＝P_n-1+δP’，δP’＝-δP；

If T<T_tartThen let T_n-1＝T_endOr T is>T_endThen let T_n＝T _start；

All the horizontal angle values P, the vertical angle values T, and the camera magnification power values Z, which are numbered 0 to x, are grouped into the photo location array aP1, and the final photo location array aP1 is made to be PTZ scanning information.

S2: according to the PTZ scanning information, positioning the photographing position of the camera to generate an image to be identified;

in this embodiment, referring to fig. 2, the specific steps of generating the image to be recognized in step S2 are as follows:

s21: the high-altitude pan-tilt head sequentially reads the PTZ values in the photographing positioning array aP1, and positions the camera to a photographing position corresponding to the appointed PTZ value by using a camera API of the camera;

s22: after the camera reaches any photographing position in the photographing positioning array aP1, pausing for m seconds, capturing an image to be recognized by using a camera API (application program interface), and placing the image to be recognized into a capturing queue Q1;

s23: controlling the camera to reach the next photographing position in the photographing positioning array aP1, and repeating the step S22;

s3: the high-altitude pan-tilt head carries out CNN model identification on the images to be identified in the grab image queue Q1, and suspicious images are extracted;

the specific steps of CNN recognition of the image to be recognized and suspicious image extraction in step S3 are as follows:

the high-altitude pan-tilt head sequentially acquires images to be identified from the image capture queue Q1, and puts the images into a CNN (hidden network) model for feature identification to obtain an identification result; the identification result at least comprises frame selection content of the image and identification data of withered and dead trees; the identification data of the withered trees are as follows:

aBoxs＝[{x₁₁,y₁₁,x₁₂,y₁₂},{x₂₁,y₂₁,x₂₂,y₂₂},...,{x_i1,y_i1,x_i2,y_i2},...]；

wherein x_i1，y_i1Representing that the identified tree i corresponds to the coordinates of the upper left corner of the image to be identified; x is the number of_i2,y_i2Representing that the identified tree i corresponds to the lower right corner coordinate of the image to be identified;

if the withered trees exist in the identification result of the image to be identified, putting the identification result into a withered tree array R1, and taking the image as a suspicious image; otherwise the image is discarded.

S4: the high-altitude cloud deck controls the camera to shoot and re-identify the corresponding position according to the suspicious image, and locks the PTZ coordinate position of the dead tree in the suspicious image;

referring to fig. 3, the specific steps of step S4 are:

s41: the high altitude cloud deck obtains withered and dead tree identification data corresponding to each suspicious image from withered and dead tree array R1 in sequence, namely:

aBoxs＝[{x₁₁,y₁₁,x₁₂,y₁₂},{x₂₁,y₂₁,x₂₂,y₂₂},...,{x_i1,y_i1,x_i2,y_i2},...]；

s42: the high-altitude tripod head sequentially takes elements i of identification data of dead trees: { x_i1,y_i1,x_i2,y_i2Sending a positioning command to the camera by using a 3D positioning API of the camera;

s43: the high-altitude pan-tilt control camera API captures images and stores the re-captured images;

s44: the high-altitude pan-tilt acquires a PTZ value according to the current camera API; and storing the image and the rephotograph image into a rephotograph queue Q2;

s45: the high-altitude pan-tilt head sequentially obtains the rephoto snap-shot images from the rephoto snap-shot queue Q2, puts the images into a CNN model for feature recognition to obtain the recognition result of the rephoto images, and puts the recognition result into the array R2 for re-recognizing dead trees; the identification result of the rephotograph comprises rephotograph frame selection content and identification data of dead trees;

s46: if withered trees exist in the recognition result of the double-shot snap-shot image, locking the PTZ coordinate positions of the withered trees in the suspicious image; otherwise, discarding the identification data of the withered and dead trees corresponding to the suspicious image;

s47: judging whether the identification data of the withered and dead trees in the suspicious image is identified again, if so, entering step S48; otherwise, identifying the identification data of the next withered tree in the suspicious image, and returning to the step S41;

s48: judging whether all the suspicious images are identified, if not, locking the next suspicious image, and returning to the step S41; if yes, all PTZ coordinate positions are output and the process proceeds to step S5.

S5: and determining longitude and latitude information of a corresponding camera according to the PTZ coordinate position of the withered and dead tree in the suspicious image obtained in the S4, and calculating the longitude and latitude of the geographic position of the withered and dead tree.

In this embodiment, the specific steps of step S5 are:

s51: sequentially reading PTZ coordinates in the re-identified dead tree array R2, and converting the PTZ coordinates in the re-identified dead tree array R2 into a warp and weft value of the dead tree by using the longitude and the weft of the camera as an origin of coordinates by using a geotools;

s52: correspondingly storing the longitude and latitude values of the withered and dead trees in an array R2 for re-identifying the withered and dead trees one by one to obtain coordinate data R3 of the withered and dead trees;

s53: and judging whether all the deadly tree arrays R2 are completely converted, if so, outputting deadly tree coordinate data R3, and otherwise, returning to the step S51.

It should be noted that the above description is not intended to limit the present invention, and the present invention is not limited to the above examples, and those skilled in the art may make variations, modifications, additions or substitutions within the spirit and scope of the present invention.

Claims (6)

1. A detection and positioning method for pine wilt and nematode withered trees based on high-altitude pan-tilt recognition is characterized by comprising the following steps:

s1: determining a detection target area, acquiring the surrounding environment in the detection target area by adopting a high-altitude pan-tilt camera, and generating PTZ scanning information;

wherein P represents a horizontal angle value; t represents a vertical angle value; z represents the magnification value of the camera;

s2: according to the PTZ scanning information, positioning the photographing position of the camera to generate an image to be identified;

s3: the high-altitude pan-tilt head carries out CNN model identification on the images to be identified in the grab image queue Q1, and suspicious images are extracted;

s4: the high-altitude cloud deck controls the camera to shoot and re-identify the corresponding position according to the suspicious image, and locks the PTZ coordinate position of the dead tree in the suspicious image;

s5: and determining longitude and latitude information of a corresponding camera according to the PTZ coordinate position of the withered and dead tree in the suspicious image obtained in the S4, and calculating the longitude and latitude of the geographic position of the withered and dead tree.

2. The detecting and positioning method for pine wilt disease and dead tree based on high altitude pan-tilt recognition according to claim 1, wherein the specific content of the PTZ scanning information generated in step S1 is as follows:

setting a starting PTZ value and an ending PTZ value of the high-altitude pan-tilt camera to be scanned, namely setting P_start、T_start、Z_start、P_end、T_end、Z_end；

Setting a camera photographing positioning value according to a stepping comparison lookup table of the Z value and the PT value between the initial PTZ value and the end PTZ value, and controlling the camera to photograph according to the photographing positioning value; sequentially putting the photographing positioning values into a photographing positioning array aP 1;

wherein, the step comparison lookup table of the Z value and the PT value is detailed in a table I;

z value range (Z)₀-Z_x) δP δT 1-3 p1 t1 3-8 p2 t3 8-12 p3 t4 12-15 p4 t5 15-20 p5 t6 20-30 p6 t7 Greater than 30 p7 t8

Setting a photographing positioning array aP1 as a null array, and setting initial values of a horizontal angle value P, a vertical angle value T and a camera magnification numerical value Z; let P₀＝P_start，T₀＝T_start，Z₀＝Z_start1 is ═ 1; and P is₀＝P_start，T₀＝T_start，Z₀＝Z_startPut into the photo location array aP 1;

according to the current magnification numerical value Z of the camera, finding out a corresponding horizontal stepping value delta P and a corresponding vertical stepping value delta T from a stepping comparison lookup table of the Z value and the PT value; and obtaining P_n＝P_n-1+ δ P and T_n＝T_n-1+ δ T, so that the camera is in the current photographing position; n has a value of 1 toAn integer between x;

if P_n<P_startThen let P_n-1＝P_startOr P is_n>P_endLet P_n-1＝P_end(ii) a At this time, P_n＝P_n-1+δP’，δP’＝-δP；

If T<T_tartThen let T_n-1＝T_endOr T is>T_endThen let T_n＝T_start；

All the horizontal angle values P, the vertical angle values T, and the camera magnification power values Z, which are numbered 0 to x, are grouped into the photo location array aP1, and the final photo location array aP1 is made to be PTZ scanning information.

3. The detecting and positioning method for pine wilt disease and dead tree based on high altitude pan-tilt recognition according to claim 2, wherein the specific steps of generating the image to be recognized in step S2 are as follows:

s21: the high-altitude pan-tilt head sequentially reads the PTZ values in the photographing positioning array aP1, and positions the camera to a photographing position corresponding to the appointed PTZ value by using a camera API of the camera;

s22: after the camera reaches any photographing position in the photographing positioning array aP1, pausing for m seconds, capturing an image to be recognized by using a camera API (application program interface), and placing the image to be recognized into a capturing queue Q1;

s23: controlling the camera to reach the next photographing position in the photographing positioning array aP1, and repeating the step S22;

4. the detecting and positioning method for pine wilt disease and dead tree based on high altitude pan-tilt recognition according to claim 3, wherein the specific steps of CNN recognition of the image to be recognized in step S3 to extract the suspicious image are as follows:

the high-altitude pan-tilt head sequentially acquires images to be identified from the image capture queue Q1, and puts the images into a CNN (hidden network) model for feature identification to obtain an identification result; the identification result at least comprises frame selection content of the image and identification data of withered and dead trees; the identification data of the withered trees are as follows:

aBoxs＝[{x₁₁,y₁₁,x₁₂,y₁₂},{x₂₁,y₂₁,x₂₂,y₂₂},...,{x_i1,y_i1,x_i2,y_i2},...]；

wherein x_i1，y_i1Representing that the identified tree i corresponds to the coordinates of the upper left corner of the image to be identified; x is the number of_i2,y_i2Representing that the identified tree i corresponds to the lower right corner coordinate of the image to be identified;

if the withered trees exist in the identification result of the image to be identified, putting the identification result into a withered tree array R1, and taking the image as a suspicious image; otherwise the image is discarded.

5. The method for detecting and positioning the pine wilt disease and dead tree based on high-altitude pan-tilt recognition according to claim 4, wherein the step S4 comprises the following steps:

s41: the high altitude cloud deck obtains withered and dead tree identification data corresponding to each suspicious image from withered and dead tree array R1 in sequence, namely:

aBoxs＝[{x₁₁,y₁₁,x₁₂,y₁₂},{x₂₁,y₂₁,x₂₂,y₂₂},...,{x_i1,y_i1,x_i2,y_i2},...]；

s42: the high-altitude tripod head sequentially takes the elements i: { x:_i1,y_i1,x_i2,y_i2sending a positioning command to the camera by using a 3D positioning API of the camera;

s43: the high-altitude pan-tilt control camera API captures images and stores the re-captured images;

s44: the high-altitude pan-tilt acquires a PTZ value according to the current camera API; and storing the image and the rephotograph image into a rephotograph queue Q2;

s45: the high-altitude pan-tilt head sequentially obtains the rephoto snap-shot images from the rephoto snap-shot queue Q2, puts the images into a CNN model for feature recognition to obtain the recognition result of the rephoto images, and puts the recognition result into the array R2 for re-recognizing dead trees; the identification result of the rephotograph comprises rephotograph frame selection content and identification data of dead trees;

s46: if withered trees exist in the recognition result of the double-shot snap-shot image, locking the PTZ coordinate positions of the withered trees in the suspicious image; otherwise, discarding the identification data of the withered and dead trees corresponding to the suspicious image;

s47: judging whether the identification data of the withered and dead trees in the suspicious image is identified again, if so, entering step S48; otherwise, identifying the identification data of the next withered tree in the suspicious image, and returning to the step S41;

s48: judging whether all the suspicious images are identified, if not, locking the next suspicious image, and returning to the step S41; if the identification is completed, all PTZ coordinate positions are output to step S5.

6. The method for detecting and positioning the pine wilt disease and dead tree based on high altitude pan-tilt recognition according to claim 5, wherein the step S5 comprises the following steps:

s51: sequentially reading PTZ coordinates in the re-identified dead tree array R2, and converting the PTZ coordinates in the re-identified dead tree array R2 into a warp and weft value of the dead tree by using the longitude and the weft of the camera as an origin of coordinates by using a geotools;

s52: correspondingly storing the longitude and latitude values of the withered and dead trees in an array R2 for re-identifying the withered and dead trees one by one to obtain coordinate data R3 of the withered and dead trees;

s53: and judging whether all the deadly tree arrays R2 are completely converted, if so, outputting deadly tree coordinate data R3, and otherwise, returning to the step S51.

Images