CN116152177A - Epidemic wood identification method, device, computer equipment and computer readable storage medium - Google Patents

Epidemic wood identification method, device, computer equipment and computer readable storage medium Download PDF

Info

Publication number
CN116152177A
CN116152177A CN202211725894.1A CN202211725894A CN116152177A CN 116152177 A CN116152177 A CN 116152177A CN 202211725894 A CN202211725894 A CN 202211725894A CN 116152177 A CN116152177 A CN 116152177A
Authority
CN
China
Prior art keywords
remote sensing
epidemic wood
epidemic
image
sensing image
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.)
Pending
Application number
CN202211725894.1A
Other languages
Chinese (zh)
Inventor
尤勇敏
请求不公布姓名
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiuling Zhejiang Digital Technology Co ltd
Original Assignee
Jiuling Zhejiang Digital Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiuling Zhejiang Digital Technology Co ltd filed Critical Jiuling Zhejiang Digital Technology Co ltd
Priority to CN202211725894.1A priority Critical patent/CN116152177A/en
Publication of CN116152177A publication Critical patent/CN116152177A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30188—Vegetation; Agriculture

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Image Analysis (AREA)

Abstract

The application provides a epidemic wood identification method, a device, computer equipment and a computer readable storage medium, wherein the method comprises the following steps: acquiring a remote sensing image set acquired aiming at a region to be identified; inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition to obtain a target remote sensing image containing epidemic wood point location information; and acquiring space coordinate information corresponding to the epidemic wood point position information, and screening and combining the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified. By adopting the method and the device, the epidemic wood recognition accuracy can be effectively improved.

Description

Epidemic wood identification method, device, computer equipment and computer readable storage medium
Technical Field
The application relates to the technical field of pine wood nematode disease epidemic wood monitoring, in particular to an epidemic wood identification method, an epidemic wood identification device, computer equipment and a computer readable storage medium.
Background
Pine wood nematode disease is a destructive disease of pine tree caused by pine wood nematode, is a forestry quarantine pest which has the most serious harm to forest in China at present, and has rapid spread and strong adaptability, so that infected tree plants (commonly called epidemic trees) are rapidly found and removed, and the method is important for protecting ecological health of forest areas.
At present, in the industry, in a epidemic wood searching method for pine wood nematode disease, an unmanned aerial vehicle flies in a low altitude and shoots a remote sensing image, the remote sensing image is spliced, so that after a synthesized image of a monitoring area is obtained, a target detection is carried out on the spliced image by using a deep learning method, and the epidemic wood position is searched. However, in order to ensure that no region is missed in the shooting process, the images of adjacent shooting points have higher overlapping rate, so that the problems of misplacement, distortion, stretching and the like of the images in the overlapping region are caused when the images are spliced, and epidemic wood identification is not facilitated.
Therefore, the existing epidemic wood recognition mode has the technical problem of low recognition accuracy.
Disclosure of Invention
Based on the foregoing, it is necessary to provide a method, a device, a computer device and a computer readable storage medium for identifying epidemic wood for improving the accuracy of identifying the pine wood nematode epidemic wood.
In a first aspect, the present application provides a method for identifying epidemic wood, including:
acquiring a remote sensing image set acquired aiming at a region to be identified;
inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition to obtain a target remote sensing image containing epidemic wood point location information;
And acquiring space coordinate information corresponding to the epidemic wood point position information, and screening and combining the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified.
In some embodiments of the present application, the area to be identified includes at least two route areas to be identified, and acquiring a remote sensing image set acquired for the area to be identified includes: acquiring a first remote sensing image acquired aiming at a first route area, and preprocessing the first remote sensing image to obtain a first remote sensing image set; acquiring a second remote sensing image acquired aiming at a second route area, and preprocessing the second remote sensing image to obtain a second remote sensing image set; the first route area and the second route area are adjacent route areas with corresponding image overlapping rates meeting preset conditions; and determining the first remote sensing image set and the second remote sensing image set as remote sensing image sets.
In some embodiments of the present application, acquiring a first remote sensing image acquired for a first route area, and preprocessing the first remote sensing image to obtain a first remote sensing image set, including: acquiring a first remote sensing image acquired for a first route area; performing frame extraction processing on the first remote sensing image to obtain a first ground remote sensing image with at least two frames; based on a preset cutting size and an overlapping size, performing sub-graph interception processing on each first ground remote sensing image according to a preset screenshot direction to obtain a first remote sensing image set; the screenshot direction comprises a horizontal screenshot direction and/or a vertical screenshot direction.
In some embodiments of the present application, inputting a remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition to obtain a target remote sensing image including epidemic wood point location information, including: inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition, and outputting candidate remote sensing images marked with target epidemic wood points and image coordinate information of the target epidemic wood points; determining origin coordinate information of the candidate remote sensing image, and carrying out coordinate reduction processing on the image coordinate information based on the origin coordinate information to obtain epidemic wood point position information of the target epidemic wood point; and determining candidate remote sensing images containing epidemic timber point position information as target remote sensing images.
In some embodiments of the present application, the target remote sensing image is at least two frames of target remote sensing images, and the epidemic wood point location information includes first epidemic wood point location information and second epidemic wood point location information belonging to different target remote sensing images; the method for acquiring the space coordinate information corresponding to the epidemic wood point position information, so as to screen and combine the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified comprises the following steps: acquiring and according to the image acquisition time of the target remote sensing image, acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information, and obtaining space coordinate information; calculating a distance value between the first space coordinate information and the second space coordinate information to screen epidemic wood point position information with the distance value smaller than a preset distance threshold value as a target point position combination; extracting respective point location image features of the target point location combinations, merging the epidemic wood point location information based on the point location image features, and analyzing to obtain the epidemic wood position in the area to be identified.
In some embodiments of the present application, the target remote sensing image is at least two frames of target remote sensing images including GPS coordinate information; the method for acquiring the space coordinate information comprises the steps of acquiring and according to the image acquisition time of a target remote sensing image, acquiring first space coordinate information corresponding to first epidemic wood point position information and second space coordinate information corresponding to second epidemic wood point position information to obtain the space coordinate information, and comprises the following steps: acquiring and sequentially arranging the target remote sensing images according to the image acquisition time of the target remote sensing images to obtain a remote sensing image sequence; extracting target remote sensing images of two adjacent frames in a remote sensing image sequence, wherein the target remote sensing images are respectively used as a first image and a second image, epidemic wood point position information contained in the first image is used as first epidemic wood point position information, and epidemic wood point position information contained in the second image is used as second epidemic wood point position information; and acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information based on the GPS coordinate information to obtain the space coordinate information.
In some embodiments of the present application, extracting respective point location image features of the target point location combination, and performing merging processing on the epidemic wood point location information based on the point location image features, and analyzing to obtain an epidemic wood position in the area to be identified, including: extracting respective point location image features of epidemic wood point location information contained in the target point location combination to be respectively used as a first point location image feature and a second point location image feature; calculating a similarity value between the first point image feature and the second point image feature, screening out a target point combination with the similarity value larger than a preset similarity threshold value, and carrying out point combination processing to obtain target point information; and acquiring space coordinate information corresponding to the target point position information and space coordinate information corresponding to the uncombined epidemic wood point position information as the epidemic wood position in the area to be identified.
In a second aspect, the present application provides a epidemic wood recognition device, including:
the image acquisition module is used for acquiring a remote sensing image set acquired aiming at the area to be identified;
the epidemic wood recognition module is used for inputting the remote sensing image set into the trained epidemic wood recognition model to carry out epidemic wood recognition, so as to obtain a target remote sensing image containing epidemic wood point location information;
the position analysis module is used for acquiring space coordinate information corresponding to the epidemic wood point position information, and screening and combining the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified.
In a third aspect, the present application also provides a computer device comprising:
one or more processors;
a memory; and one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the epidemic prevention method.
In a fourth aspect, the present application further provides a computer readable storage medium having stored thereon a computer program, the computer program being loaded by a processor to perform the steps of the above-described epidemic wood identification method.
In a fifth aspect, embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the method provided in the first aspect.
According to the epidemic wood identification method, the device, the computer equipment and the computer readable storage medium, the server acquires the remote sensing image set acquired for the area to be identified, inputs the remote sensing image set into the trained epidemic wood identification model for epidemic wood identification, so that a target remote sensing image containing epidemic wood point location information can be obtained, then acquires spatial coordinate information corresponding to the epidemic wood point location information, further acquires and performs screening and merging processing on the epidemic wood point location information according to the image acquisition time and the spatial coordinate information of the target remote sensing image, and finally the epidemic wood position in the area to be identified can be obtained. Therefore, the epidemic wood recognition scheme provided by the application does not need to carry out image splicing processing, so that the time required for splicing is saved, the problems of image feature damage and the like caused by splicing can be avoided, and the accuracy of epidemic wood recognition is effectively improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is an application scenario diagram of a epidemic wood recognition method provided in an embodiment of the present application;
fig. 2 is a schematic flow chart of a method for identifying epidemic wood according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of a sub-graph interception step provided in an embodiment of the present application;
FIG. 4 is a schematic flow chart of a point merging step according to an embodiment of the present disclosure;
FIG. 5 is a flow chart of a model reasoning step provided in an embodiment of the present application;
fig. 6 is a schematic structural diagram of a epidemic wood recognition device according to an embodiment of the present disclosure;
fig. 7 is a schematic structural diagram of a computer device in the embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
In the description of the present application, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration. Any embodiment described herein as "for example" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the invention. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the invention with unnecessary detail. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
In the embodiments of the present application, the epidemic identification method mainly relates to Computer Vision technology (CV) in artificial intelligence (Artificial Intelligence, AI). Wherein artificial intelligence is the intelligence of simulating, extending and expanding a person using a digital computer or a machine controlled by a digital computer, sensing the environment, obtaining knowledge, and using knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence.
The computer vision is a science for researching how to make a machine "see", and more specifically, a camera and a computer are used to replace human eyes to identify, track and measure targets, and the like, and further, graphic processing is performed, so that the computer is processed into images which are more suitable for human eyes to observe or transmit to an instrument to detect. As a scientific discipline, computer vision research-related theory and technology has attempted to build artificial intelligence systems that can acquire information from images or multidimensional data. Computer vision techniques typically include image processing, image recognition, image semantic understanding, image retrieval, optical character recognition (Optical Character Recognition, OCR), video processing, video semantic understanding, video content/behavior recognition, three-dimensional object reconstruction, 3D techniques, virtual reality, augmented reality, synchronous positioning, map construction, etc., as well as common biometric recognition techniques such as face recognition, fingerprint recognition, etc. In the application, for an Image to be detected, CV mainly realizes Image detection (Image detection) in Image semantic understanding (Image Semantic Understanding, ISU), detects a target object in the Image, and outputs a detection result. It is understood that the target object may be any object determined by actual business requirements, such as a person, a vehicle, a parcel, etc., but in embodiments of the present application, the target object may refer to a tree infected with pine wood nematode disease. Of course, the epidemic wood identification method provided by the embodiment of the application can also be used for identifying trees with other plant diseases.
The embodiment of the application provides a epidemic wood identification method, a device, computer equipment and a computer readable storage medium, and the detailed description is given below.
Referring to fig. 1, fig. 1 is a schematic view of a scene of a epidemic wood recognition method provided by the present application, where the epidemic wood recognition method can be applied to a epidemic wood recognition system. The epidemic wood recognition system comprises a terminal 102 and a server 104. The terminal 102 may be a device that includes both receive and transmit hardware, i.e., a device having receive and transmit hardware capable of performing bi-directional communications over a bi-directional communication link. Such a device may include: a cellular or other communication device having a single-line display or a multi-line display or no display. The terminal 102 may be a desktop terminal or a mobile terminal, and the terminal 102 may be one of a mobile phone, a tablet computer, a notebook computer, a monocular camera, a multi-view camera, and an unmanned aerial vehicle. The server 104 may be a stand-alone server, or may be a server network or a server cluster of servers, including but not limited to a computer, a network host, a single network server, a set of multiple network servers, or a cloud server of multiple servers. Wherein the Cloud server is composed of a large number of computers or web servers based on Cloud Computing (Cloud Computing). In addition, the terminal 102 and the server 104 establish a communication connection through a network, and the network may specifically be any one of a wide area network, a local area network, and a metropolitan area network.
Those skilled in the art will appreciate that the application environment shown in fig. 1 is only one application scenario applicable to the present application, and is not limited to the application scenario of the present application, and that other application environments may also include more or fewer computer devices than those shown in fig. 1, for example, only 1 server 104 is shown in fig. 1, and it will be appreciated that the epidemic identification system may also include one or more other servers, and is not limited herein. In addition, as shown in fig. 1, the epidemic wood recognition system may further include a memory for storing data, such as remote sensing images collected by the unmanned aerial vehicle.
It should be noted that, the schematic view of the scene of the epidemic wood recognition system shown in fig. 1 is only an example, and the description of the epidemic wood recognition system and the scene of the embodiment of the present invention is for more clearly describing the technical solution of the embodiment of the present invention, and does not constitute a limitation on the technical solution provided by the embodiment of the present invention, and as a person of ordinary skill in the art knows, along with the evolution of the epidemic wood recognition system and the appearance of a new service scene, the technical solution provided by the embodiment of the present invention is applicable to similar technical problems.
Referring to fig. 2, an embodiment of the present application provides a method for identifying epidemic wood, which is mainly applied to the server 104 in fig. 1 to illustrate the method, and the method includes steps S201 to S203, specifically as follows:
s201, acquiring a remote sensing image set acquired for a region to be identified.
The area to be identified may be a mountain forest, agriculture and forestry or forest area which needs to be identified whether epidemic wood exists or not, and the shape of the area may be any shape, such as a circle, a square, a rectangle or any polygon, which is not limited in the embodiment of the present application.
The remote sensing image set may refer to a set of at least two remote sensing images, and the remote sensing images may be images needing to identify epidemic positions, including but not limited to pictures, video frames in videos, and the like; the video includes, but is not limited to, short video, long video, etc., which may be video less than 10 minutes in length, and long video which may be video greater than 10 minutes in length. Of course, the division of the video length is not limited to "10 minutes", and those skilled in the art may set the division conditions of other long and short videos according to the actual service requirement, which is not limited in the embodiment of the present application.
In a specific implementation, in order to improve the accuracy of epidemic wood recognition, the embodiment of the application proposes that an artificial intelligence technology can be combined to perform epidemic wood recognition on a remote sensing image set acquired for a region to be recognized, namely, a target detection algorithm is used to directly recognize epidemic wood points of a remote sensing image, the recognized epidemic wood points are combined through a correlation algorithm, and finally, the combined epidemic wood points and the rest epidemic wood points which are not combined are output as results. The method bypasses the process of remote sensing image stitching, namely does not involve image stitching, so that the time required by image stitching can be greatly saved, the problem of reduced recognition accuracy caused by image feature damage to a target area in the stitching process is avoided, and time consumption and recognition result quality are both obviously improved.
Further, the remote sensing image set required for epidemic wood recognition may be obtained by the terminal 102 and then sent to the server 104, or may be obtained by other devices and then transmitted to the server 104 through the terminal 102, where the obtaining manner of the remote sensing image set includes but is not limited to one of the following manners: 1. in a common network architecture, the server 104 receives a remote sensing image set from the terminal 102 or other cloud device with a network connection established; 2. in a preset blockchain network, the server 104 can synchronously acquire a remote sensing image set from other terminal nodes or server nodes, and the blockchain network can be a public chain, a private chain and the like; 3. in the preset tree structure, the server 104 may request the remote sensing image set from the upper server or poll the remote sensing image set from the lower server. Specifically, the method for acquiring the remote sensing image set is not limited in particular, and may be determined by an actual service scenario or a requirement, and the terminal 102 may be an unmanned aerial vehicle, such as an unmanned aerial vehicle, carrying the image acquisition device.
In one embodiment, the area to be identified includes at least two route areas to be identified, and step S201 includes: acquiring a first remote sensing image acquired aiming at a first route area, and preprocessing the first remote sensing image to obtain a first remote sensing image set; acquiring a second remote sensing image acquired aiming at a second route area, and preprocessing the second remote sensing image to obtain a second remote sensing image set; the first route area and the second route area are adjacent route areas with corresponding image overlapping rates meeting preset conditions; and determining the first remote sensing image set and the second remote sensing image set as remote sensing image sets.
Wherein the course area is a sub-area of the area to be identified which is divided for the unmanned aerial vehicle to fly the course, the first course area and the second course area will be explained as examples.
The image overlapping rate refers to a percentage of similarity between the remote sensing images acquired by the terminal 102 (e.g., the drone) along the first route region and the pixel content compared to the remote sensing images acquired along the second route region. The corresponding preset condition can be set according to the actual application requirement, for example, the preset condition is set to be 60%, which means that the image overlapping rate of the selected adjacent route area is not lower than 60%, so as to ensure that no area is missed.
In a specific implementation, the server 104 may perform frame extraction processing on the first remote sensing image/the second remote sensing image by using an image processing tool, for example, using tools such as "OpenCV" or "ffmpeg", where the frame extraction frequency may be one time a second or multiple times a second; the number of frames may be one frame at a time or may be multiple frames at a time, which is not limited in this application. After obtaining the first remote sensing image and the second remote sensing image and performing frame extraction processing on the video image, the server 104 respectively obtains more than one first video frame as a first remote sensing image set and more than one second video frame as a second remote sensing image set.
Thus, the server 104 obtains the remote sensing image set required for completing the current epidemic wood recognition task by obtaining the first remote sensing image set and the second remote sensing image set.
It should be noted that, in this embodiment, the area to be identified is divided into the first route area and the second route area, so as to improve the accuracy of epidemic wood identification, and the lifting principle is mainly implemented by a subsequent correlation algorithm which is difficult to be in space dimension, that is, the area to be identified is divided into a plurality of subareas to respectively identify epidemic wood, and finally, the target correlation algorithm is used to analyze and combine epidemic wood points, so that the accuracy of epidemic wood identification can be improved, and further, the accuracy of epidemic wood identification is improved, and specific implementation steps will be described in detail below.
In one embodiment, acquiring a first remote sensing image acquired for a first route area, and preprocessing the first remote sensing image to obtain a first remote sensing image set, including: acquiring a first remote sensing image acquired for a first route area; performing frame extraction processing on the first remote sensing image to obtain a first ground remote sensing image with at least two frames; based on a preset cutting size and an overlapping size, performing sub-graph interception processing on each first ground remote sensing image according to a preset screenshot direction to obtain a first remote sensing image set; the screenshot direction comprises a horizontal screenshot direction and/or a vertical screenshot direction.
The cutting size may be a size of performing sub-image cutting with respect to the original image, if the original image is a first ground remote sensing image obtained by extracting frames from the first remote sensing image, and the original image size is "6504×6336", the cutting size may be "1024×1024", which means that the original image may be cut into several sub-images with a size of "1024×1024".
The overlapping size may be the size of an overlapping pixel allowed when the original image is intercepted, for example, the overlapping size is set to be "x=200", which means that when the image is captured in the horizontal direction, the overlapping area between the subgraphs is a "200" pixel value; as another example, the overlap size is set to "y=100", meaning that the overlap area between sub-images is "100" pixel value when the vertical direction is taken.
The screenshot direction comprises a horizontal screenshot direction and/or a vertical screenshot direction, the horizontal screenshot direction can be regarded as an X-axis direction of a screen coordinate system, the vertical screenshot direction can be regarded as a Y-axis direction of the screen coordinate system, and an origin of the screen coordinate system can be arranged at the upper left corner position.
In a specific implementation, referring to fig. 3, after the server 104 obtains a first remote sensing image collected by the terminal 102 for a first route area and performs video image frame extraction processing on the first remote sensing image to obtain a plurality of frames of first ground remote sensing images, screenshot processing may be performed on each frame of first ground remote sensing image, for example, according to a preset cutting size and an overlapping size, sub-image interception is performed along a screenshot direction, and then a plurality of frames of sub-images may be intercepted from each frame of first ground remote sensing image, and the first remote sensing image set is obtained by packaging. It can be understood that the screenshot mode in this embodiment is a sliding window type, and the screenshot direction of the sliding window type may be horizontal first and then vertical, or may be vertical first and then horizontal, so that only the cutting size and the overlapping size in the two directions need to be ensured to be consistent.
Further, after the server 104 processes the first remote sensing image set, the second remote sensing image set can be obtained in the same manner, which is not described in detail in the embodiment of the present application. After the first remote sensing image set and the second remote sensing image set are obtained, the image set may be input into a trained epidemic wood recognition model (e.g. "model one" shown in fig. 3) for epidemic wood detection, so that the model outputs a target remote sensing image containing epidemic wood point location information based on the trained and learned epidemic wood detection capability.
S202, inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition, and obtaining a target remote sensing image containing epidemic wood point location information.
The trained epidemic wood recognition model can be any model with target detection capability, such as a fast R-CNN, SSD or YOLO model, and the model is not limited in the embodiment of the application.
The epidemic wood point information may refer to image coordinate information of the detected epidemic wood point in the remote sensing image, such as "(x, y)".
In a specific implementation, before the server 104 inputs the remote sensing image set into the trained epidemic wood recognition model, model training needs to be performed on the epidemic wood recognition model, and the model training step may include: constructing an initial epidemic wood recognition model (such as a Faster R-CNN, SSD or YOLO model); acquiring a ground image set, and dividing the ground image set into a training set and a testing set; the ground image set comprises a plurality of ground images marked with epidemic wood positions; performing preliminary training on the initial epidemic wood recognition model by using the training set to obtain a preliminarily trained epidemic wood recognition model; and testing and adjusting the preliminarily trained epidemic wood recognition model by using the test set to obtain a trained epidemic wood recognition model.
Further, after the remote sensing image set is input to the trained epidemic wood recognition model by the server 104, the trained epidemic wood recognition model will infer the images in the remote sensing image set one by one, and then output the remote sensing image marked with the target object, "epidemic wood points" (for example, marked with a model, or framed with a rectangular frame), as the target remote sensing image, and the remote sensing image not detected with "epidemic wood points," that is, the non-target remote sensing image. It can be understood that the trained epidemic wood recognition model can not only recognize and mark the 'epidemic wood points', but also output the image coordinate information thereof, and the image coordinate information is used for determining the epidemic wood position in the real scene, which will be described in detail below.
In one embodiment, step S202 includes: inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition, and outputting candidate remote sensing images marked with target epidemic wood points and image coordinate information of the target epidemic wood points; determining origin coordinate information of the candidate remote sensing image, and carrying out coordinate reduction processing on the image coordinate information based on the origin coordinate information to obtain epidemic wood point position information of the target epidemic wood point; and determining candidate remote sensing images containing epidemic timber point position information as target remote sensing images.
The origin coordinate information may refer to image coordinate information corresponding to the origin of the upper left corner of the candidate remote sensing image, and the origin coordinate information may be expressed as "(x) 0 ，y 0 ) The image coordinate information may be expressed as "(x, y)".
In specific implementation, the above embodiment refers to that the remote sensing image marked with the target object "epidemic wood point" can be used as the target remote sensing image, but the remote sensing image needs to be used as the candidate remote sensing image in the embodiment, because the embodiment is implemented based on the sub-graph interception mentioned in the previous step, and if there is no sub-graph interception step, the coordinate reduction scheme proposed in the embodiment is not required to be applied.
Specifically, referring to fig. 3, when the server 104 cuts a sub-image of the first or second ground remote sensing image, the upper left corner coordinate (i.e., origin coordinate information) of the sub-image is "(x) 0 ，y 0 ) The coordinate information of the image of a certain target epidemic wood point obtained by using the epidemic wood detection algorithm is (x, y), and then the corresponding coordinate (x) of the target epidemic wood point on the original image is needed to be analyzed 0 +x，y 0 +y) ", as new epidemic wood point location information, the remote sensing image with the updated epidemic wood point location information can be used as the target remote sensing image. In other words, if the subgraph is not truncated, the target remote sensing image is an image containing epidemic wood point information "(x, y)"; however, if the subgraph is intercepted, the target remote sensing image contains epidemic point location information (x) 0 +x，y 0 +y) ".
S203, acquiring space coordinate information corresponding to the epidemic wood point position information, and screening and combining the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified.
The epidemic wood point location information is image coordinate information corresponding to the existence of a screen coordinate system, and the space coordinate information is space coordinate information corresponding to the existence of a world coordinate system, and can be represented by longitude and latitude.
In a specific implementation, after the server 104 analyzes the epidemic wood point location information of the remote sensing image set based on the scheme described in the above embodiment, the spatial coordinate information of each epidemic wood point location information contained in the map may be further obtained based on the GPS information contained in each target remote sensing image, so as to perform coordinate screening and merging processing based on the spatial coordinate information, thereby analyzing the real epidemic wood position in the area to be identified.
In one embodiment, the target remote sensing image is at least two frames of target remote sensing images, the epidemic wood point location information includes first epidemic wood point location information and second epidemic wood point location information belonging to different target remote sensing images, and step S203 includes: acquiring and according to the image acquisition time of the target remote sensing image, acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information, and obtaining space coordinate information; calculating a distance value between the first space coordinate information and the second space coordinate information to screen epidemic wood point position information with the distance value smaller than a preset distance threshold value as a target point position combination; extracting respective point location image features of the target point location combinations, merging the epidemic wood point location information based on the point location image features, and analyzing to obtain the epidemic wood position in the area to be identified.
In a specific implementation, in order to obtain the epidemic wood position in the area to be identified, the server 104 may perform a set of comparison analysis on the multiple target remote sensing images obtained by the analysis in the previous step, where the target remote sensing images include: A. b, C, D, analyzing A and B, analyzing B and C, analyzing C and D, analyzing more images, and so on.
Further, before analyzing the two target remote sensing images A and B, the first epidemic wood point location information "(x) contained in the target remote sensing image" A "needs to be obtained 1 ，y 1 ) "first spatial coordinate information" (x' 1 ，y′ 1 ) And obtaining second epidemic wood point location information (x) contained in the target remote sensing image B 2 ，y 2 ) "second spatial coordinate information" (x' 2 ，y′ 2 ) As the spatial coordinate information described above. Then, a distance value between the first space coordinate information and the second space coordinate information is calculated, namely, the distance value "d" can be calculated through a distance formula between two point coordinates, and the formula is as follows:
Figure BDA0004025881810000121
further, the server 104 may compare the distance value "d" with a predetermined distance threshold, and if the distance value "d" is smaller than the predetermined distance threshold, the corresponding epidemic wood point location information "(x) is retained 1 ，y 1 ) Sum (x) 2 ，y 2 ) "as the target point combination, namely as the analysis basis of the subsequent epidemic wood point combination treatment; if the distance is greater than or equal to the preset distance threshold value, corresponding space coordinate information "(x ')' 1 ，y′ 1 ) "sum" (x' 2 ，y′ 2 ) "is the final desired log position. The spatial coordinate information acquisition step and the epidemic wood position acquisition step, which are referred to in the present embodiment, will be described in detail below, respectively.
In one embodiment, the target remote sensing image is at least two frames of target remote sensing images containing GPS coordinate information; the method for acquiring the space coordinate information comprises the steps of acquiring and according to the image acquisition time of a target remote sensing image, acquiring first space coordinate information corresponding to first epidemic wood point position information and second space coordinate information corresponding to second epidemic wood point position information to obtain the space coordinate information, and comprises the following steps: acquiring and sequentially arranging the target remote sensing images according to the image acquisition time sum of the target remote sensing images to obtain a remote sensing image sequence; extracting target remote sensing images of two adjacent frames in a remote sensing image sequence, wherein the target remote sensing images are respectively used as a first image and a second image, epidemic wood point position information contained in the first image is used as first epidemic wood point position information, and epidemic wood point position information contained in the second image is used as second epidemic wood point position information; and acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information based on the GPS coordinate information to obtain the space coordinate information.
The GPS coordinate information may be coordinate information expressed in terms of longitude and latitude acquired based on a global positioning system. That is, the unmanned aerial vehicle not only can collect remote sensing images, but also can collect GPS coordinate information corresponding to the images through a global positioning system while collecting the images.
In the specific implementation, the flying height of the unmanned aerial vehicle during low-altitude flying shooting can be about 1000 meters, and the reason is that the shooting area is enlarged as much as possible and the ground area size corresponding to each pixel in the picture is ensured to be within 10 cm, so that the remote sensing image shot during the unmanned aerial vehicle low-altitude flying comprises GPS coordinate information besides RGB information, and real space coordinate information can be determined by epidemic wood point position information by determining the GPS coordinates corresponding to the origin coordinate information.
Further, in the above embodiment, it is only proposed that a set of comparison analysis should be performed for a plurality of target remote sensing images, for example, the target remote sensing images include: A. b, C, D, the four graphs need to be analyzed first for A and B, then for B and C, and finally for C and D, but the grouping basis is not clear. Therefore, in this embodiment, it is proposed that the grouping may be performed according to the image acquisition time, for example, the image acquisition time of each target remote sensing image is determined, and then the target remote sensing images are sequentially arranged according to the image acquisition time from small to large, so as to obtain a remote sensing image sequence, then the target remote sensing images of two adjacent frames in the sequence are extracted as a group, and the space coordinate information corresponding to the epidemic timber point information contained in each of the two previous and subsequent frames of target remote sensing images is obtained, so as to obtain the first space coordinate information and the second space coordinate information recorded by different target remote sensing images.
It should be noted that, if a certain target remote sensing image includes a plurality of target epidemic wood points, that is, there are a plurality of epidemic wood point location information, when performing distance analysis, permutation analysis is required to be performed on all the epidemic wood point location information, so as to screen out an accurate target point location combination.
In one embodiment, extracting respective point location image features of the target point location combination, and combining the epidemic wood point location information based on the point location image features, and analyzing to obtain the epidemic wood position in the region to be identified, including: extracting respective point location image features of epidemic wood point location information contained in the target point location combination to be respectively used as a first point location image feature and a second point location image feature; calculating a similarity value between the first point image feature and the second point image feature, screening out a target point combination with the similarity value larger than a preset similarity threshold value, and carrying out point combination processing to obtain target point information; and acquiring space coordinate information corresponding to the target point position information and space coordinate information corresponding to the uncombined epidemic wood point position information as the epidemic wood position in the area to be identified.
In a specific implementation, in the point location screening scheme based on the time dimension analysis distance value in the above embodiment, a target association algorithm based on the time dimension is essentially adopted, in this embodiment, image feature extraction can be performed on the target point location combination, then image feature vectors belonging to different target remote sensing images are compared, that is, feature similarity analysis is performed, and finally, epidemic wood point location information with high similarity is combined to avoid repeated output of the same epidemic wood location.
Further, the server 104 may use the feature extraction model "mapping" to extract the point image features of the two images, that is, the point image feature of a region with a certain side length (for example, "30m×30 m") centered on the target epidemic wood point (the feature is a feature vector with a dimension of "512" obtained by transforming the model after fusing a series of appearance information such as colors, shapes, textures, sizes, and position distributions of the epidemic wood and the peripheral trees), which are respectively used as the first point image feature "X" belonging to the target remote sensing image "a" and the second point image feature "Y" belonging to the target remote sensing image "B". And then, calculating cosine similarity of the two point location image features to obtain a similarity value. Finally, the target point combinations with the similarity values larger than the preset similarity threshold value are screened out to perform point combination processing, and the target point combinations can be combined into one to obtain target point information as shown in fig. 4. Therefore, the space coordinate information corresponding to the target point position information and the space coordinate information corresponding to the uncombined epidemic wood point position information are the current epidemic wood positions required to be identified.
It should be noted that, the feature extraction model "enabling" uses "res net18" as a backbone network, and the reasoning process can refer to fig. 5, and the cosine similarity formula is as follows:
Figure BDA0004025881810000141
It should be noted that, the above screening and merging processing for the epidemic wood point location information should be analysis for a single route area, if the area to be identified is divided into a plurality of route areas, after screening and merging processing for the epidemic wood point location information is performed for each route area, the epidemic wood point location information retained between two adjacent route areas is analyzed by using a space dimension, and the analysis mode is the distance analysis and the feature similarity analysis described in the above embodiment, and finally, the spatial coordinate information of each of the merged target point location information and the non-merged epidemic wood point location information is used as the epidemic wood location in the area to be identified.
According to the epidemic wood recognition method, the server obtains the target remote sensing image containing the epidemic wood point location information by obtaining the remote sensing image set collected for the region to be recognized and inputting the remote sensing image set into the trained epidemic wood recognition model for epidemic wood recognition, then obtains the space coordinate information corresponding to the epidemic wood point location information, further obtains and processes screening and merging the epidemic wood point location information according to the image collection time and the space coordinate information of the target remote sensing image, and finally obtains the epidemic wood position in the region to be recognized. Therefore, the epidemic wood recognition scheme provided by the application does not need to carry out image splicing processing, so that the time required for splicing is saved, the problems of image feature damage and the like caused by splicing can be avoided, and the accuracy of epidemic wood recognition is effectively improved.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
In order to better implement the epidemic wood recognition method provided in the embodiment of the present application, on the basis of the epidemic wood recognition method provided in the embodiment of the present application, an apparatus for identifying epidemic wood is further provided in the embodiment of the present application, as shown in fig. 6, where the apparatus 600 for identifying epidemic wood includes:
an image acquisition module 610, configured to acquire a remote sensing image set acquired for an area to be identified;
The epidemic wood recognition module 620 is configured to input the remote sensing image set into a trained epidemic wood recognition model to perform epidemic wood recognition, so as to obtain a target remote sensing image containing epidemic wood point location information;
the position analysis module 630 is configured to obtain spatial coordinate information corresponding to the epidemic wood point location information, so as to screen and combine the epidemic wood point location information according to the spatial coordinate information and the image acquisition time of the target remote sensing image, and obtain the epidemic wood position in the area to be identified.
In one embodiment, the area to be identified includes at least two route areas to be identified, and the image acquisition module 610 is further configured to acquire a first remote sensing image acquired for the first route area, and perform preprocessing on the first remote sensing image to obtain a first remote sensing image set; acquiring a second remote sensing image acquired aiming at a second route area, and preprocessing the second remote sensing image to obtain a second remote sensing image set; the first route area and the second route area are adjacent route areas with corresponding image overlapping rates meeting preset conditions; and determining the first remote sensing image set and the second remote sensing image set as remote sensing image sets.
In one embodiment, the image acquisition module 610 is further configured to acquire a first remote sensing image acquired for a first airline area; performing frame extraction processing on the first remote sensing image to obtain a first ground remote sensing image with at least two frames; based on a preset cutting size and an overlapping size, performing sub-graph interception processing on each first ground remote sensing image according to a preset screenshot direction to obtain a first remote sensing image set; the screenshot direction comprises a horizontal screenshot direction and/or a vertical screenshot direction.
In one embodiment, the epidemic wood recognition module 620 is further configured to input the set of remote sensing images into the trained epidemic wood recognition model for epidemic wood recognition, and output candidate remote sensing images marked with the target epidemic wood points, and image coordinate information of the target epidemic wood points; determining origin coordinate information of the candidate remote sensing image, and carrying out coordinate reduction processing on the image coordinate information based on the origin coordinate information to obtain epidemic wood point position information of the target epidemic wood point; and determining candidate remote sensing images containing epidemic timber point position information as target remote sensing images.
In one embodiment, the target remote sensing image is at least two frames of target remote sensing images, the epidemic wood point location information includes first epidemic wood point location information and second epidemic wood point location information which belong to different target remote sensing images, and the position analysis module 630 is further configured to obtain, according to an image acquisition time of the target remote sensing image, first space coordinate information corresponding to the first epidemic wood point location information, and second space coordinate information corresponding to the second epidemic wood point location information, so as to obtain space coordinate information; calculating a distance value between the first space coordinate information and the second space coordinate information to screen epidemic wood point position information with the distance value smaller than a preset distance threshold value as a target point position combination; extracting respective point location image features of the target point location combinations, merging the epidemic wood point location information based on the point location image features, and analyzing to obtain the epidemic wood position in the area to be identified.
In one embodiment, the target remote sensing images are at least two frames of target remote sensing images containing GPS coordinate information, and the position analysis module 630 is further configured to acquire and sequentially arrange each target remote sensing image according to the image acquisition time of the target remote sensing images, so as to obtain a remote sensing image sequence; extracting target remote sensing images of two adjacent frames in a remote sensing image sequence, wherein the target remote sensing images are respectively used as a first image and a second image, epidemic wood point position information contained in the first image is used as first epidemic wood point position information, and epidemic wood point position information contained in the second image is used as second epidemic wood point position information; and acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information based on the GPS coordinate information to obtain the space coordinate information.
In one embodiment, the position analysis module 630 is further configured to extract respective point location image features of the epidemic wood point location information contained in the target point location combination, as a first point location image feature and a second point location image feature, respectively; calculating a similarity value between the first point image feature and the second point image feature, screening out a target point combination with the similarity value larger than a preset similarity threshold value, and carrying out point combination processing to obtain target point information; and acquiring space coordinate information corresponding to the target point position information and space coordinate information corresponding to the uncombined epidemic wood point position information as the epidemic wood position in the area to be identified.
In the above embodiment, the server obtains the target remote sensing image including the epidemic wood point location information by obtaining the remote sensing image set collected for the area to be identified and inputting the remote sensing image set into the trained epidemic wood identification model for epidemic wood identification, and then obtains the space coordinate information corresponding to the epidemic wood point location information, and further obtains and combines the epidemic wood point location information according to the image collection time and the space coordinate information of the target remote sensing image, thereby finally obtaining the epidemic wood position in the area to be identified. Therefore, the epidemic wood recognition scheme provided by the application does not need to carry out image splicing processing, so that the time required for splicing is saved, the problems of image feature damage and the like caused by splicing can be avoided, and the accuracy of epidemic wood recognition is effectively improved.
It should be noted that, the specific limitation of the epidemic wood recognition device may be referred to the limitation of the epidemic wood recognition method, and will not be described herein. All or part of the modules in the epidemic wood recognition device can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or independent of a processor in the electronic device, or may be stored in software in a memory in the electronic device, so that the processor may call and execute operations corresponding to the above modules.
In some embodiments of the present application, the epidemic identification device 600 may be implemented in the form of a computer program that is executable on a computer apparatus as shown in fig. 7. The memory of the computer device may store various program modules constituting the epidemic wood recognition apparatus 600, such as the image acquisition module 610, the epidemic wood recognition module 620, and the position analysis module 630 shown in fig. 6; the computer program constituted by the respective program modules causes the processor to execute the steps in the epidemic wood recognition method of the respective embodiments of the present application described in the present specification. For example, the computer apparatus shown in fig. 7 may perform step S201 through the image acquisition module 610 in the epidemic wood recognition device 600 shown in fig. 6. The computer device may execute step S202 through the epidemic wood recognition module 620. The computer device may perform step S203 through the location analysis module 630. The computer device includes a processor, a memory, and a network interface coupled by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external computer device through a network connection. The computer program when executed by a processor implements a method of epidemic wood identification.
It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In some embodiments of the present application, a computer device is provided that includes one or more processors; a memory; and one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to perform the steps of the epidemic wood recognition method. The step of this epidemic wood recognition method may be the step in the above-described epidemic wood recognition method of each embodiment.
In some embodiments of the present application, a computer readable storage medium is provided, in which a computer program is stored, where the computer program is loaded by a processor, so that the processor performs the steps of the above epidemic identification method. The step of the epidemic wood recognition method here may be a step in the epidemic wood recognition method of each of the above embodiments.
Those of ordinary skill in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein can include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can take many forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above description of the method, the device, the computer equipment and the computer readable storage medium for identifying epidemic wood provided in the embodiments of the present application has been provided in detail, and specific examples are applied to illustrate the principles and the embodiments of the present invention, and the above description of the embodiments is only used to help understand the method and the core idea of the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.

Claims (10)

1. A method for identifying epidemic wood, comprising:
acquiring a remote sensing image set acquired aiming at a region to be identified;
inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition to obtain a target remote sensing image containing epidemic wood point location information;
And acquiring space coordinate information corresponding to the epidemic wood point position information, and screening and combining the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified.
2. The method of claim 1, wherein the area to be identified comprises at least two route areas to be identified, and the acquiring the set of remote sensing images acquired for the area to be identified comprises:
acquiring a first remote sensing image acquired aiming at a first route area, and preprocessing the first remote sensing image to obtain a first remote sensing image set; and
acquiring a second remote sensing image acquired aiming at a second route area, and preprocessing the second remote sensing image to obtain a second remote sensing image set; the first route area and the second route area are adjacent route areas with corresponding image overlapping rates meeting preset conditions;
and determining the first remote sensing image set and the second remote sensing image set as the remote sensing image sets.
3. The method of claim 2, wherein the acquiring a first remote sensing image acquired for a first airline area and preprocessing the first remote sensing image to obtain a first remote sensing image set comprises:
Acquiring a first remote sensing image acquired for a first route area;
performing frame extraction processing on the first remote sensing image to obtain a first ground remote sensing image with at least two frames;
based on a preset cutting size and an overlapping size, performing sub-graph interception processing on each first ground remote sensing image according to a preset screenshot direction to obtain a first remote sensing image set;
the screenshot directions comprise a horizontal screenshot direction and/or a vertical screenshot direction.
4. The method of claim 1, wherein inputting the set of remote sensing images into a trained epidemic wood recognition model for epidemic wood recognition to obtain a target remote sensing image comprising epidemic wood point location information comprises:
inputting the remote sensing image set into a trained epidemic wood recognition model for epidemic wood recognition, and outputting candidate remote sensing images marked with target epidemic wood points and image coordinate information of the target epidemic wood points;
determining origin coordinate information of the candidate remote sensing image, and carrying out coordinate reduction processing on the image coordinate information based on the origin coordinate information to obtain epidemic wood point position information of the target epidemic wood point;
and determining candidate remote sensing images containing the epidemic timber point position information as the target remote sensing images.
5. The method of any one of claims 1 to 4, wherein the target remote sensing image is at least two frames of target remote sensing images, the epidemic point location information comprising first and second epidemic point location information attributed to different target remote sensing images;
the step of obtaining the space coordinate information corresponding to the epidemic wood point location information, so as to screen and combine the epidemic wood point location information according to the space coordinate information and the image acquisition time of the target remote sensing image, and obtaining the epidemic wood position in the area to be identified comprises the following steps:
acquiring and according to the image acquisition time of the target remote sensing image, acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information, and obtaining the space coordinate information;
calculating a distance value between the first space coordinate information and the second space coordinate information to screen out epidemic wood point position information of which the distance value is smaller than a preset distance threshold value as a target point position combination;
extracting respective point location image features of the target point location combinations, combining the epidemic wood point location information based on the point location image features, and analyzing to obtain the epidemic wood position in the area to be identified.
6. The method of claim 5, wherein the target remote sensing image is at least two frames of target remote sensing images containing GPS coordinate information;
the obtaining and obtaining first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information according to the image acquisition time of the target remote sensing image, and obtaining the space coordinate information comprises the following steps:
acquiring and sequentially arranging the target remote sensing images according to the image acquisition time of the target remote sensing images to obtain a remote sensing image sequence;
extracting target remote sensing images of two adjacent frames in the remote sensing image sequence, wherein the target remote sensing images are respectively used as a first image and a second image, epidemic wood point position information contained in the first image is used as first epidemic wood point position information, and epidemic wood point position information contained in the second image is used as second epidemic wood point position information;
and acquiring first space coordinate information corresponding to the first epidemic wood point location information and second space coordinate information corresponding to the second epidemic wood point location information based on the GPS coordinate information to obtain the space coordinate information.
7. The method of claim 5, wherein the extracting the respective point location image features of the target point location combinations, and performing merging processing on the epidemic wood point location information based on the point location image features, and analyzing to obtain the epidemic wood position in the area to be identified, includes:
extracting respective point location image features of epidemic wood point location information contained in the target point location combination to be respectively used as a first point location image feature and a second point location image feature;
calculating a similarity value between the first point image feature and the second point image feature, screening out a target point combination with the similarity value larger than a preset similarity threshold value, and carrying out point combination processing to obtain target point information;
and acquiring space coordinate information corresponding to the target point position information and space coordinate information corresponding to the uncombined epidemic wood point position information as the epidemic wood position in the area to be identified.
8. A epidemic wood identification device, comprising:
the image acquisition module is used for acquiring a remote sensing image set acquired aiming at the area to be identified;
the epidemic wood recognition module is used for inputting the remote sensing image set into a trained epidemic wood recognition model to carry out epidemic wood recognition, so as to obtain a target remote sensing image containing epidemic wood point location information;
And the position analysis module is used for acquiring the space coordinate information corresponding to the epidemic wood point position information, and screening and combining the epidemic wood point position information according to the space coordinate information and the image acquisition time of the target remote sensing image to obtain the epidemic wood position in the area to be identified.
9. A computer device, comprising:
one or more processors;
a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the epidemic identification method of any one of claims 1 to 7.
10. A computer-readable storage medium, having stored thereon a computer program, the computer program being loaded by a processor to perform the steps of the epidemic identification method according to any one of claims 1 to 7.
CN202211725894.1A 2022-12-29 2022-12-29 Epidemic wood identification method, device, computer equipment and computer readable storage medium Pending CN116152177A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211725894.1A CN116152177A (en) 2022-12-29 2022-12-29 Epidemic wood identification method, device, computer equipment and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211725894.1A CN116152177A (en) 2022-12-29 2022-12-29 Epidemic wood identification method, device, computer equipment and computer readable storage medium

Publications (1)

Publication Number Publication Date
CN116152177A true CN116152177A (en) 2023-05-23

Family

ID=86355547

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211725894.1A Pending CN116152177A (en) 2022-12-29 2022-12-29 Epidemic wood identification method, device, computer equipment and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN116152177A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116934700A (en) * 2023-07-17 2023-10-24 农芯(南京)智慧农业研究院有限公司 Method, device, equipment and storage medium for controlling and controlling pine wood nematode disease epidemic prevention and treatment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116934700A (en) * 2023-07-17 2023-10-24 农芯(南京)智慧农业研究院有限公司 Method, device, equipment and storage medium for controlling and controlling pine wood nematode disease epidemic prevention and treatment
CN116934700B (en) * 2023-07-17 2024-05-28 农芯(南京)智慧农业研究院有限公司 Method, device, equipment and storage medium for controlling and controlling pine wood nematode disease epidemic prevention and treatment

Similar Documents

Publication Publication Date Title
CN110569721B (en) Recognition model training method, image recognition method, device, equipment and medium
US11928800B2 (en) Image coordinate system transformation method and apparatus, device, and storage medium
CN109960742B (en) Local information searching method and device
US10043097B2 (en) Image abstraction system
CN111667001B (en) Target re-identification method, device, computer equipment and storage medium
CN107346414B (en) Pedestrian attribute identification method and device
CN110298281B (en) Video structuring method and device, electronic equipment and storage medium
CN112101195B (en) Crowd density estimation method, crowd density estimation device, computer equipment and storage medium
CN112528974B (en) Distance measuring method and device, electronic equipment and readable storage medium
CN112101386B (en) Text detection method, device, computer equipment and storage medium
CN109508636A (en) Vehicle attribute recognition methods, device, storage medium and electronic equipment
CN113537180B (en) Tree obstacle identification method and device, computer equipment and storage medium
CN112200056A (en) Face living body detection method and device, electronic equipment and storage medium
CN116152177A (en) Epidemic wood identification method, device, computer equipment and computer readable storage medium
CN109657540B (en) Withered tree positioning method and system
CN115512238A (en) Method and device for determining damaged area, storage medium and electronic device
CN113033386B (en) High-resolution remote sensing image-based transmission line channel hidden danger identification method and system
KR102388777B1 (en) System for providing adjacent building pre-survey service usign 360 degree virtual reality camera
Bellavia et al. Image orientation with a hybrid pipeline robust to rotations and wide-baselines
CN115880765A (en) Method and device for detecting abnormal behavior of regional intrusion and computer equipment
CN113570615A (en) Image processing method based on deep learning, electronic equipment and storage medium
CN112329550A (en) Weak supervision learning-based disaster-stricken building rapid positioning evaluation method and device
CN111626212A (en) Method and device for identifying object in picture, storage medium and electronic device
CN111767839A (en) Vehicle driving track determining method, device, equipment and medium
CN112215205B (en) Target identification method and device, computer equipment and storage medium

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