Method for constructing intelligent recognition model of himalayan marmot hole fused with habitat information

Abstract

The embodiment of the application discloses a Himalayan marmot hole intelligent identification model construction method fusing habitat information, comprising the following steps: acquiring multiple unmanned aerial vehicle image graphs of Himalayan marmot holes; marking marmot holes and their surrounding habitat features in each unmanned aerial vehicle image graph, wherein the habitat features include hole mouth mounds, vegetation, stone blocks and animal traces; training a target detection model by using each marked unmanned aerial vehicle image graph, and the trained model is used to improve the marmot hole missed identification condition. The embodiment can overcome the marmot hole missed detection condition.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of image target detection, and particularly relates to a Himalayan marmot hole intelligent recognition model construction method fusing habitat information. BACKGROUND

[0002] The Himalayan marmot is the natural host of plague, and by identifying the Himalayan marmot hole, the habitat and survival condition of the Himalayan marmot can be understood, and the risk of plague can be prevented.

[0003] The patent application CN112699852A provides an intelligent marmot recognition and monitoring system, which first builds a set of marmot image sample library based on unmanned aerial vehicle images, and trains a deep learning model to obtain a marmot recognition model. Then, the unmanned aerial vehicle is used to take pictures in a large range in an epidemic area to obtain marmot images in the epidemic area, and the marmot density in the area and the potential risk of plague are predicted. However, since the marmot is active, the shooting time and range of the unmanned aerial vehicle are limited, resulting in the omission of some marmot objects in the image during recognition. In addition, the marmot is sensitive to the abnormal "intrusion" of the unmanned aerial vehicle, and will escape from the visual range of the unmanned aerial vehicle. In combination with the above two points, there is a large error in determining the marmot density through the unmanned aerial vehicle image, and the uncertainty of predicting the risk of plague is high.

[0004] The patent application CN114155973A provides a plague prediction method based on deep learning, which combines the unmanned aerial vehicle image, the unmanned vehicle data and the plague prediction model of the YOLO4 model. This patent collects the hole data of small rodents such as yellow mice, Brandt's voles and long-clawed gerbils, and trains a corresponding hole target detection model. According to the model, the hole position and number of small rodents directly exposed to the ground in the source of the epidemic can be obtained, and on this basis, a plurality of indexes are selected to construct a plague risk prediction model, the changes of these indexes are monitored, and the possible risk area in the region is obtained combined with the sampling vehicle data. In this way, when the label data is selected, the target object obviously exposed to the unmanned aerial vehicle image is directly labeled, so that only the obvious holes in the image can be identified, and the objects hidden by vegetation, stones and other objects are easy to be missed, and thus there is an error in the number of holes. SUMMARY

[0005] The embodiment of the present application provides a Himalayan marmot hole intelligent recognition model construction method to solve the problem of missing recognition of the marmot hole.

[0006] In a first aspect, the embodiment of the present application provides a Himalayan marmot hole intelligent recognition model construction method fusing habitat information, comprising:

[0007] Obtain a plurality of unmanned aerial vehicle image pictures of the Himalayan marmot hole;

[0008] Label the habitat features of the marmot burrows and their surroundings in each drone image, including burrow entrance mounds, vegetation, rocks, and animal tracks.

[0009] The target detection model is trained using the labeled drone images, and the trained model is used to improve the identification of marmot burrow leaks.

[0010] In a second aspect, an embodiment of the present invention provides an electronic device, comprising:

[0011] one or more processors;

[0012] a memory for storing one or more programs,

[0013] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for constructing an intelligent recognition model of Himalayan marmot holes that integrates habitat information as described in any embodiment.

[0014] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for constructing an intelligent recognition model of Himalayan marmot holes that integrates habitat information as described in any embodiment.

[0015] In summary, this embodiment provides a method for constructing an intelligent recognition model for Himalayan marmot burrows that integrates habitat information. This model assists in plague natural foci investigations by identifying Himalayan marmot burrows rather than Himalayan marmots. First, a large-scale image of Himalayan marmot burrows is obtained through drone photography, and the iconic burrow entrance features are combined with the marmot burrow to define a broad-based marmot burrow. This is then applied as a labeling strategy to the Himalayan marmot burrow training sample set and the recognition model construction method, reducing the risk of missing targets in drone imagery and improving the target detection model's ability to recognize Himalayan marmot burrows in large-scale scenarios. Compared to narrow-sense marmot burrows directly exposed to the surface, broad-sense marmot burrows include not only marmot burrows but also habitat information such as burrow entrance attachments and obstructions. When applied to marmot burrow labeling and recognition, even if the target object is exposed in a small area in the image, the habitat information features near the burrow entrance can still be used to identify the marmot burrow, effectively solving the problem of missed model detection caused by partial or complete occlusion of the target in drone imagery. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 This is a flow chart of a method for constructing an intelligent identification model for Himalayan marmot holes that integrates habitat information, provided by an embodiment of the present invention;

[0018] Figure 2 This is a diagram of the size distribution of marmot holes provided by an embodiment of the present invention;

[0019] Figure 3 This is a flowchart of another method for constructing an intelligent recognition model for Himalayan marmot holes that integrates habitat information, provided by an embodiment of the present invention;

[0020] Figure 4 1 is a schematic diagram of detection results of a broad sense marmot burrow and a narrow sense marmot burrow provided by an embodiment of the present invention;

[0021] Figure 5 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention are described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0023] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0024] In the description of the present invention, it should also be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0025] Figure 1 This is a flow chart of a method for constructing an intelligent recognition model for Himalayan marmot holes that integrates habitat information, provided by an embodiment of the present invention. The method is executed by an electronic device, such as Figure 1 As shown, the specific steps include:

[0026] S110. Acquire multiple drone images of Himalayan marmot caves.

[0027] This embodiment constructs Himalayan marmot burrow data, which is obtained by drone photography. Considering the wide distribution range of Himalayan marmot burrows, the drone can use a higher altitude and be equipped with a high-resolution camera to capture orthophotos of the marmot burrows, which include both the marmot burrow objects and various habitat features around them.

[0028] S120. Label the marmot burrows and their surrounding habitat features in each drone image.

[0029] This embodiment, combined with image analysis, can extract unique habitat features around Himalayan marmot burrow entrances from images. For example, the length and width of the burrow entrances are generally larger than those of common pika burrows, and there are distinct mounds, slopes, and vegetation around the burrow entrances. These can serve as hallmark features that distinguish Himalayan marmot burrows from those of other pika burrows.

[0030] Optionally, the dimensions of the marmot burrows can be determined based on the environmental characteristics of the burrows and the features of drone images, and the habitat features to be labeled around the burrows can be selected from a large number of data sets. In a specific embodiment, the process can include the following steps:

[0031] Step 1. Determine the basic marking dimensions of the marmot holes based on the normal distribution of the sizes of the marmot holes in each drone image. This step determines the marking frame dimensions of the marmot holes based on the morphological characteristics of the hole openings. Specifically, the shapes of the marmot hole openings in the sampling area are mostly circular or elliptical. The direction of the marmot hole opening is used as the measurement direction of the length of the marmot hole opening, and the direction perpendicular to it is used as the measurement direction of the hole opening width. The measured hole opening length is between 10-74cm, and the hole opening width is between 8-70cm. The statistical results show that the average length of the marmot hole opening is 27.26cm, and the average width is 29.84cm. The data is normally distributed with the mean as the center, indicating that the size of the marmot holes in the survey area is highly consistent, such as Figure 2 As shown, the size of a labeling box that only considers marmot holes can be determined based on the statistical size.

[0032] Step 2: Based on the environment around the marmot hole and the features of the drone image, determine the habitat features that need to be marked around the marmot hole. Optionally, this embodiment provides three screening paths:

[0033] The first screening path identifies habitat features that require annotation around marmot burrows based on the shape and size of the burrow entrance features, as well as their differences from the surface and vegetation. Alternatively, the significance of the shape and size of various burrow entrance features, as well as their differences from the surface and vegetation, can be analyzed one by one. Ultimately, burrow entrance mounds are identified as habitat features that require annotation based on their fan-shaped or comet-shaped shape, large size, significant color difference from the surface, and contrast with exposed patches of vegetation. Specifically, burrow entrance mounds are often fan-shaped or comet-shaped. Of the 159 marmot burrows measured in the Qinghai-Tibet Plateau sampling area, only three had no obvious mounds. Nearly 98% of the burrow entrances had mounds around them, and the surrounding mound slopes are clearly visible in drone imagery. Measurements also show that the slopes of the mounds around the burrow entrances are approximately 23-290 cm long and 12-250 cm wide. Statistical results show that the average length and width of the mounds are 126.25 cm and 126.04 cm, respectively. The specific width varies depending on the type of marmot burrow. In bare areas, the mounds differ significantly in color from the surrounding surface, making them easily identifiable. In vegetated areas, the mounds may be covered or destroyed by vegetation, leaving exposed patches that indirectly indicate hidden burrow entrances. Therefore, these mounds can serve as a key marker for identifying marmot burrows.

[0034] The second screening approach identifies habitat features requiring annotation around marmot burrows based on their burrow entrance location and environmental adaptability. Alternatively, the importance of various burrow entrance features in marmot burrow site selection and their adaptability to the environment can be analyzed individually. Ultimately, based on the shelter and soil properties preferred by marmots during burrow site selection, tree roots, rocks, and earth mounds can be identified as habitat features requiring annotation. Specifically, regarding environmental adaptability of burrow entrance location, most marmot burrows tend to be located in areas with natural shelter, such as beneath tree roots or rocks, using these shelters to enhance concealment and reduce exposure risk. Furthermore, they have certain soil requirements, such as prioritizing areas with loose soil to minimize digging difficulty. Alternatively, based on the distribution of burrow entrances and surrounding earth mounds, the distribution of marmot burrows can be classified into the following seven categories: large earth mounds, burrow entrances along vegetation roots, burrow entrances along rocks, areas with dense vegetation, areas with dense rocks, agricultural-pastoral transition zones, and mixed distribution with pika burrows. These can all be used as markers for marmot holes and are key areas for capture.

[0035] The third screening path is to determine the habitat features that need to be marked around the marmot holes based on the vertical shooting characteristics of the drone. Optionally, the obstructions in the vertical shooting of the drone can be analyzed one by one, and finally the obstructing elements such as vegetation at the hole entrance, stones, and traces of animal activities are used as habitat features that need to be marked. Specifically, there is a natural challenge in identifying marmot holes based on drone images: the vertical shooting characteristics of drone images lead to a low exposure rate of marmot holes, especially when the hole entrance is parallel or inclined to the ground, the hole entrance is obscured by vegetation, stones, traces of animal activities, etc., which can easily form a "visual blind spot". Therefore, for the obscured hole entrance, the obstruction itself can also be used as a landmark to identify the marmot hole.

[0036] For ease of distinction and description, this embodiment refers to the habitat features required for labeling obtained through the three aforementioned screening paths as the first, second, and third habitat features, respectively. The union of these three habitat features forms the set of habitat features required for labeling around marmot burrows. Ultimately, this set includes the burrow entrance mound, vegetation, rocks, and animal tracks. In this embodiment, these habitat features, along with the marmot burrow, constitute the broad definition of a marmot burrow.

[0037] Step three, adjust the annotation size of the generalized marmot burrows according to the characteristics of each habitat. In subsequent operations, this application will annotate the generalized marmot burrows, so it is necessary to adjust the size of the initial annotation frame to accommodate the above-mentioned habitat characteristics. The generalized marmot burrows can not only effectively deal with the obstruction of the burrow entrance, but also highlight the difference between marmot burrows and symbiotic animal burrows. For example, the length and width of the common pika burrow entrance are generally less than 10 cm, which is significantly smaller than the marmot burrows. It can be used as a basis for distinction. Even if marmot burrows and pika burrows coexist in some areas, the size difference is still a key feature for classification; for example, the definition of the generalized marmot burrows introduces fan-shaped or comet-shaped mounds, and their area is generally larger than 100 cm. 2, and there are vegetation or local exposed patches, stones and other features around it. Even if the entrance to the cave is completely blocked, its existence can still be inferred through related habitat characteristics; in addition, the burrowing behavior of marmots is closely related to their living habits, which is essentially different from the ordinary pika burrows associated with slender mounds and no shelter.

[0038] Step 4: Determine the hierarchical structure of the annotations based on the planar distribution map of marmot burrows. This step takes into account the hierarchical distribution of marmot burrows. Analysis of the planar distribution map reveals a distinction between main burrows and satellite burrows. Main burrows have larger mounds and are centrally located, while satellite burrows have smaller mounds and are distributed around the main burrows, forming a hierarchical network. Therefore, when sampling the annotations, both main and satellite burrows should be evenly distributed to ensure representative data for subsequent model training.

[0039] After the above preparations are completed, a labeling frame containing the marmot burrow and its surrounding habitat features can be generated in each drone image; and the category, horizontal coordinate of the center point, vertical coordinate of the center point, width and height of the labeling frame can be stored. In a specific embodiment, the labeling process includes:

[0040] First, image screening is performed. Since target identification will be performed later, a high-quality dataset must be selected based on the aforementioned drone image dataset. Specifically, when selecting the data, images that do not contain the target object or have poor quality are eliminated, incorporating manual visual interpretation methods. At the same time, it is important to ensure that the habitat features identified in the above steps are as fully as possible around the cave entrance.

[0041] Then, data annotation and image cropping are performed. Based on the previous data collection, information acquisition, and knowledge conversion, a generalized marmot hole is defined in this step. Unlike the narrow marmot hole, which only includes the hole entrance, the generalized marmot hole also includes the habitat characteristics around the hole entrance. Specifically, the content of a single annotation object is as follows:

[0042]

[0043] in, Represents an object labeled with a narrow sense of marmot hole, represents an object labeled with a generalized marmot hole, Indicates the marmot hole object in the current annotation content. It represents the habitat characteristics around the target object, such as mounds, slopes, and vegetation, and k and i represent the number and index of the marked marmot holes, respectively.

[0044] Finally, we use the LabelImg tool to label the data. The broad sense of marmot holes is defined as hole-mounds, and the narrow sense of marmot holes is defined as holes. The data is stored in the YOLO format. The basic storage structure is as follows:

[0045]

[0046] in, Indicates the id of the current category, Indicates the x-coordinate of the center point of the annotation box. Indicates the y coordinate of the center point of the annotation box, and They represent the width and height of the annotation box respectively. The following four parameters are the normalized result values.

[0047] Furthermore, since the original captured images contain a lot of irrelevant information, it is necessary to crop the annotated images to reduce the time and space overhead of the model training process. Specifically, after the image is annotated, a 416×416 cropping area is drawn based on the annotated box range to crop the original image, so that all the images are ultimately 416×416 in size.

[0048] S130. Using the labeled drone images, a target detection model is trained. The trained model is used to improve the identification of marmot hole leaks.

[0049] Optionally, data partitioning and model training can be performed first. After a series of preprocessing operations are performed on the original images, model training can be performed. In one specific embodiment, the image dataset and the corresponding annotated dataset can be divided into a training set and a test set as needed. The training set is used to train the model parameters, and the trained parameters are further improved in combination with the test set. Different partitioning ratios are used depending on the characteristics of the data; the ratio used here is 9:1.

[0050] Based on the partitioned dataset, a transfer learning framework was used to assist in model training. Generally speaking, transfer learning involves relearning the features of a new dataset based on an existing model to create a new model. For example, the YOLO11 model has been trained on public datasets of varying sizes to produce a variety of models, ranging from small to large, including n, s, m, and l (corresponding to different model categories). Based on these basic models, transfer learning combined with the marmot cave data can reduce the time required to create recognition models.

[0051] Before training the model, you also need to set the appropriate parameters. Key training parameters include image size (imgsz), training epochs (epochs), batch size (batch), data augmentation strategy (close_mosaic), and the device used (device). A common augmentation operation involves random scaling and cropping of the original image, followed by image stitching. This enhances the features of the target being detected. During training, depending on hardware conditions, select imgsz 416, epochs 400, batch 8, close_mosaic 20, and the optimizer strategy, stochastic gradient descent. The device can be adjusted based on the computer's configuration.

[0052] After training, the model can be evaluated. mAP50 is the core metric used to evaluate the model's performance. It measures the degree of overlap between the predicted and true bounding boxes. Other metrics include precision and recall.

[0053]

[0054]

[0055] Among them, TP represents true positive samples, that is, samples predicted as positive and correctly predicted, FP represents samples predicted as positive but incorrectly predicted, and FN represents samples predicted as negative and correctly predicted. IoU represents intersection over union, where Intersection represents the intersection between the predicted box and the actual box, and Union represents the union between the predicted box and the actual box. mAP50 is the average value of the average precision calculated under the condition that the IoU threshold is 0.5. The above method can also be used Figure 3 The flowchart shown is represented.

[0056] The effect of the model constructed in this embodiment is analyzed and verified below.

[0057] First, in theory, the labeling strategy based on the generalized marmot burrow will fully learn the habitat characteristics through the neural network in deep learning, and update the model parameters through gradient backpropagation:

[0058]

[0059] in, Represents the parameters of the model during training, Represents the learning rate of parameter update during back propagation of the neural network, represents the auxiliary weight parameter, Habitat characteristics around the target object, represents the loss function based on the actual model parameters, It means partial derivative.

[0060] Based on this labeling strategy, the model will fully consider the habitat characteristics of the cave entrance when identifying the target. In this way, even if the model does not detect the actual cave entrance features, it can still calculate the surrounding environment of the cave entrance. The feature expression in the image can also identify the target object to a certain extent.

[0061] Secondly, in the experiment, the dataset used in this embodiment was captured by a DJ1 Mavic 2 Pro drone. Due to the wide distribution of Himalayan marmot burrows, the flight altitude was set to 100 meters. The sampling area included 29 typical natural plague foci in the Tibet Autonomous Region. Through image sampling and annotation, 801 target objects were obtained. To ensure sample balance, nearly half of the objects were labeled "hole" and "hole-mounds". The "hole-mounds" label corresponds to the broad-defined marmot burrows proposed in this invention, and the "hole" label corresponds to the narrow-defined marmot burrows.

[0062] During model training, since YOLO8 and YOLO11 are currently widely used in target detection tasks and have better recognition accuracy than other target detection models, the YOLO8n and YOLO11n models were used to detect holes and hole-mounds to determine the recognition accuracy of different models for these two types of objects. In terms of setting the training parameters, in order to control the effects of other factors, the core training parameters were set to the same values, including imgsz 416, epochs 400, batch 8, close_mosaic 20, and the optimization strategy optimizer was set to stochastic gradient descent. The device was selected as the CPU. The final training results are shown in the following table:

[0063] Model YOLO8n YOLO11n mAP50 0.924 0.929 mAP50-95 0.490 0.443 hole(mAP50) 0.902 0.861 hole-mounds(mAP50) 0.935 0.876 precision 0.918 0.935 recall 0.803 0.831

[0064] As can be seen from the table, hole-mounds annotation makes it easier to identify in images because of its richer and more unique features. Whether using YOLO8n or YOLO11n, its accuracy has been improved to a certain extent, by 3.59% and 1.74% respectively.

[0065] From this, it can be seen that the use of the fusion habitat feature labeling strategy can improve the recognition accuracy of Himalayan marmot burrows and reduce the missed detection of target objects, proving that the strategy is effective. Figure 4 is the target detection result obtained through experiments.

[0066] In summary, this embodiment provides a method for constructing an intelligent recognition model for Himalayan marmot burrows that integrates habitat information. The model assists in the investigation of natural plague foci by identifying Himalayan marmot burrows rather than Himalayan marmots. First, a large-scale image of Himalayan marmot burrows is obtained through drone photography, and the habitat characteristics of various burrow entrances are analyzed from the perspectives of feature size, morphology, contrast, marmot burrow location, and obstructions. The landmark information such as mounds, stones, and vegetation at the burrow entrances are screened out, and defined together with the marmot burrows as a generalized marmot burrow. This is then used as a labeling strategy in the Himalayan marmot burrow training sample set and the recognition model construction method, reducing the risk of missing the target to be detected in the drone image and improving the target detection model's ability to recognize Himalayan marmot burrows in a large-scale scenario. Compared with the narrow-sense marmot burrows directly exposed on the surface, the broad-sense marmot burrows include not only marmot burrows but also habitat information such as burrow attachments and obstructions. By applying it to marmot burrow annotation and identification, even if the target object has a small exposed area in the image, the habitat information characteristics near the burrow entrance can still be used to identify the marmot burrow, effectively solving the problem of model missed detection caused by partial or complete occlusion of the target in drone images.

[0067] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, the device includes a processor 60, a memory 61, an input device 62 and an output device 63; the number of processors 60 in the device can be one or more. Figure 5 In the embodiment, a processor 60 is used as an example; the processor 60, the memory 61, the input device 62 and the output device 63 in the device can be connected by a bus or other means. Figure 5 The bus connection is taken as an example.

[0068] Memory 61, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the method for constructing an intelligent model for identifying Himalayan marmot burrows that integrates habitat information, as described in the embodiments of the present invention. Processor 60 executes the software programs, instructions, and modules stored in memory 61 to execute various functional applications and data processing functions of the device, thereby implementing the aforementioned method for constructing an intelligent model for identifying Himalayan marmot burrows that integrates habitat information.

[0069] The memory 61 can include a program storage area and a data storage area, where the program storage area can store an operating system, application programs required by at least one function, and the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 61 can include a high-speed random access memory, and can also include a non-volatile memory such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some examples, the memory 61 can further include a memory disposed remotely with respect to the processor 60, which can be connected to the device through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0070] The input device 62 can be used to receive input digital or character information, and to generate key signal inputs related to user settings of the device and function control. The output device 63 can include a display device such as a display screen.

[0071] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the method for constructing a smart identification model of a Himalayan marmot burrow fused with habitat information.

[0072] The computer storage medium of the embodiment of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, device or apparatus.

[0073] The computer readable signal medium can include a data signal propagated in a baseband or as a part of a carrier wave, in which a computer readable program code is borne. Such a propagated data signal can take on multiple forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can send, propagate or transfer a program for use by or in connection with an instruction execution system, device or apparatus.

[0074] The program code embodied on the computer readable media can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0075] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0076] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the present application; although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the technical solutions recorded in the above-mentioned embodiments can be modified, or some or all of the technical features thereof can be replaced equivalently; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing an intelligent identification model for Himalayan marmot holes integrating habitat information, characterized in that: include: Acquire multiple drone images of Himalayan marmot caves; Based on the normal distribution of marmot burrow sizes in each UAV image, the basic dimensions of the marmot burrows were determined; Based on the environment around the marmot burrow and the characteristics of drone images, the habitat features that need to be marked around the marmot burrow are determined, wherein the habitat features include earth mounds at the burrow entrance, vegetation, stones, and animal traces; specifically, based on the fan-shaped or comet-shaped earth mounds at the burrow entrance, large size characteristics, significant color difference from the ground surface, and contrast with exposed patches formed by plants, the earth mounds at the burrow entrance are determined to be the first habitat feature that needs to be marked; based on the shelters and soil characteristics that marmots prefer when selecting burrow entrances, the tree roots, stones, and earth mounds at the burrow entrance are determined to be the second habitat features that need to be marked; based on the obstructions in the vertical photography of the drone, the vegetation, stones, and animal activity traces at the burrow entrance are determined to be the third habitat features that need to be marked; the first habitat feature, the second habitat feature, and the third habitat feature are taken as the union to form a set of habitat features that need to be marked around the marmot burrow; Adjust the annotation size of the generalized marmot burrow according to the characteristics of each habitat. The generalized marmot burrow includes the marmot burrow entrance and the habitat characteristics around the burrow entrance. Determine the hierarchical structure of the annotation based on the planar distribution map of the marmot holes; specifically, extract the hierarchical distribution of the marmot holes from the planar distribution map of the marmot holes, wherein the hierarchical distribution includes main holes and satellite holes, and the main holes and the satellite holes are evenly sampled during the annotation sampling; The target detection model is trained using the labeled drone images, and the trained model is used to improve the identification of marmot burrow leaks.

2. The method according to claim 1, characterized in that The target detection model adopts the YOLO model.

3. An electronic device, characterized in that: include: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method for constructing an intelligent recognition model of Himalayan marmot holes integrating habitat information as described in any one of claims 1-2.

4. A computer-readable storage medium, characterized in that A computer program is stored thereon, which, when executed by a processor, implements the method for constructing an intelligent recognition model of Himalayan marmot holes integrating habitat information as described in any one of claims 1-2.

Images