A method and system for regional biodiversity monitoring based on drone technology

Abstract

Multiple embodiments of this specification relate to the field of information technology, and specifically to a method and system for regional biodiversity monitoring based on drone technology. The method includes: reading multiple aerial images; identifying plant populations in the area based on visible light aerial images, and obtaining predicted animal species and their predicted scales based on the plant populations; identifying individual animals that appear in visible light aerial images and infrared aerial images, and obtaining the species and number of occurrences of individual animals; obtaining each occurrence species, likelihood scale and confidence level based on the occurrence species, number of occurrences, predicted animal species and their predicted scales; establishing a regional ecological model based on the likelihood scales of all occurrence species; correcting the likelihood scale of each occurrence species based on the food chain relationship of the regional ecological model and the confidence level of each occurrence species; and obtaining regional biodiversity monitoring results based on the plant populations and the regional ecological model after the corrected likelihood scales.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the field of information technology, and specifically to a method and system for regional biodiversity monitoring based on drone technology. Background Art

[0002] With the increasing awareness of ecological and environmental protection and the advancement of technology, biodiversity monitoring has become increasingly important. Traditional ground survey methods are time-consuming and labor-intensive and difficult to cover large areas, while the development of drone technology has provided a new solution for biodiversity monitoring. UAVs combine advanced sensor technology and geographic information systems to provide strong support for ecological research. UAVs can carry a variety of cameras. Among them, visible light cameras are used to obtain high-resolution ground images; multispectral / hyperspectral cameras are used to distinguish different types of vegetation and monitor plant health; thermal infrared cameras are used to detect temperature differences and identify animal heat signals; lidar is used for three-dimensional terrain modeling and to measure vegetation height and density. Through these data, plant species, quantity and distribution can be evaluated, and then the level of biodiversity can be understood. It provides strong technical support for ecological protection, scientific research and disaster management. Although plants can be identified and analyzed more accurately, there are technical difficulties in identifying animals. First of all, animals are not easy to be photographed directly, and on the other hand, it is impossible for all animals to be photographed. Therefore, there is still a lack of effective technology for monitoring and analysis of animal populations. Summary of the invention

[0003] Multiple embodiments of this specification describe a method and system for regional biodiversity monitoring based on drone technology.

[0004] In a first aspect, the embodiments of this specification provide a method for regional biodiversity monitoring based on drone technology, comprising the steps of:

[0005] Reading multiple aerial images in the area, wherein the aerial images include visible light aerial images and infrared aerial images;

[0006] Identify plant populations in an area based on the visible light aerial image, and obtain predicted animal species and their predicted sizes based on the plant populations;

[0007] Identify individual animals that appear in the visible light aerial image and the infrared aerial image, and obtain the type and number of the individual animals that appear;

[0008] According to the occurrence types, occurrence numbers, predicted animal species and predicted scales, obtaining each occurrence type, likelihood scale and confidence level;

[0009] Establish regional ecological models based on the likelihood scale of all described species;

[0010] Correcting the likelihood scale of each occurrence species according to the food chain relationship of the regional ecological model and the confidence of each occurrence species;

[0011] Based on the plant population and the regional ecological model after correcting the likelihood scale, the regional biodiversity monitoring results are obtained.

[0012] In a second aspect, the embodiments of this specification provide a system for regional biodiversity monitoring based on drone technology, which is characterized by comprising:

[0013] A shooting module is used to read multiple aerial images in the area, wherein the aerial images include visible light aerial images and infrared aerial images;

[0014] A first recognition module is configured to recognize plant populations in an area according to the visible light aerial image, and to obtain predicted animal species and predicted scales according to the plant populations;

[0015] A second recognition module is used to recognize individual animals that appear in the visible light aerial image and the infrared aerial image, and obtain the type and number of the individual animals that appear;

[0016] A prediction module, which obtains each occurrence type, likelihood scale and confidence level according to the occurrence type, occurrence quantity, predicted animal type and predicted scale;

[0017] Ecological module, which builds a regional ecological model based on the likelihood scale of all the described species;

[0018] A correction module, which corrects the likelihood scale of each occurrence species according to the food chain relationship of the regional ecological model and the confidence of each occurrence species;

[0019] The result module obtains the regional biodiversity monitoring result according to the plant population and the regional ecological model after the likelihood scale is corrected.

[0020] In a third aspect, an embodiment of this specification provides an electronic device, including a processor and a memory;

[0021] The processor is connected to the memory;

[0022] The memory is used to store executable program code;

[0023] The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method described in any one of the above aspects.

[0024] In a fourth aspect, an embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in any of the above aspects.

[0025] In a fifth aspect, an embodiment of this specification provides a computer program product, including a computer program, which implements the method described in any of the above aspects when executed by a processor.

[0026] The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least:

[0027] In multiple embodiments of this specification, the regional biodiversity monitoring method provided can identify plant populations based on visible light aerial images and infrared aerial images, and predict and correct animal species and likelihood scales, which can solve the industry difficulties such as low accuracy of animal data and animals being easily missed in aerial images obtained by drone aerial photography, and realize the monitoring of regional biodiversity and obtain monitoring results. The number of occurrences, predicted scale and likelihood scale are used for related calculations to finally obtain the scale and confidence of the animal population, making the monitoring of the animal population scale more accurate.

[0028] Other features and advantages of the various embodiments of the present specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 This is a schematic diagram of the application area of ​​the regional biodiversity monitoring method according to the embodiment of this specification.

[0031] Figure 2 This is a schematic diagram of obtaining aerial images according to an embodiment of this specification.

[0032] Figure 3 This is a flow chart of the method for monitoring regional biodiversity according to an embodiment of this specification.

[0033] Figure 4 It is a schematic diagram of multiple sub-region images in the embodiments of this specification.

[0034] Figure 5 This is a flow chart of a method for obtaining the types and numbers of individual animals according to an embodiment of this specification.

[0035] Figure 6This is a schematic diagram of an animal individual being photographed in the embodiments of this specification.

[0036] Figure 7 This is a flow chart of a method for obtaining the likelihood scale and confidence level of occurrence of a species according to an embodiment of this specification.

[0037] Figure 8 A flowchart of a method for modifying the likelihood scale of each occurrence category according to an embodiment of the present specification.

[0038] Fig. 9 This is a schematic diagram of a regional biodiversity monitoring system according to an embodiment of this specification.

[0039] Fig.10 A schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0040] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification, but the following embodiments are only preferred embodiments of this specification, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of this specification.

[0041] The terms "first", "second", "third", etc. in the description and claims of this specification and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0042] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation of this specification.

[0043] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.

[0044] Before describing the technical solution in this specification, an introduction is given to the application scenarios and related technologies.

[0045] With the impact of climate change, increased pollution, land use change and other factors, the global ecosystem is facing unprecedented challenges. Rising global temperatures have led to an increase in extreme weather events, affecting the living environment of organisms and the stability of ecosystems. Pollution from industrial emissions and agricultural activities not only damages water quality and soil, but also threatens the health and survival of wild animals and plants. Large-scale urbanization and agricultural expansion have led to the loss and fragmentation of biological habitats, further exacerbating the decline in biodiversity. Ecological monitoring can assess and track the health of ecosystems, biodiversity and environmental changes. Please refer to the attached Figure 1 , scientists and decision makers can obtain key information about the state of ecosystems, which is essential for formulating effective protection measures, responding to environmental changes and promoting sustainable development. At the same time, reasonable natural resource management and development and utilization also need to be based on an accurate understanding of the state of ecosystems. Ecological monitoring provides the necessary data support for resource management and helps achieve sustainable utilization.

[0046] In recent years, with the development of new technologies such as remote sensing technology, drones 20, and Internet of Things devices, ecological monitoring has become more efficient and accurate. Drones 20 include multi-rotor drones 20, fixed-wing drones 20, compound-wing drones 20, and vertical take-off and landing fixed-wing drones 20. Multi-rotor drones 20 usually have four or more propellers and can take off and land vertically and hover in the air. Suitable for detailed scanning in a smaller area, especially for complex terrain or scenes that require close observation. Fixed-wing drones 20 are similar to small aircraft and require a runway to take off and land, but are capable of long-term flight over a larger range. Suitable for rapid scanning of a large area, especially in flat or open areas.

[0047] The area in this specification refers to a designated target area, such as a designated valley, a designated wetland, a forest farm with demarcated boundaries, etc. The area 10 can be large or small according to the actual demarcation, but the technical solution disclosed in this specification is applicable. In order to cover a larger area 10, the drone 20 needs to have a longer flight time, usually at least 1 hour. When conducting ecological monitoring of area 10, it is necessary to carry a variety of sensors, such as visible light cameras, multi-spectral cameras, thermal infrared cameras, etc. In addition, in order to be able to identify biological features on the ground in detail, a high-resolution camera is required. Please refer to the attached Figure 2When the drone 20 is taking aerial photos, it takes photos while flying, so the drone 20 also needs an accurate positioning system for navigation and location recording. Therefore, the same place will be photographed by the drone 20 more than once, and the angles will be slightly different. For smaller areas 10, such as within a few square kilometers, the multi-rotor drone 20 can complete the task within an hour. For larger areas 10, such as tens of square kilometers or even larger areas to be monitored 10, the fixed-wing drone 20 can usually complete the task within a few hours to a day. In this specification, the shooting equipment carried by the drone 20 includes at least a visible light camera and an infrared camera. The shooting equipment carried by the drone 20 uploads the captured images to the server 30 via a wireless network. The server 30 receives, stores, retrieves and analyzes them. The shooting equipment can also store the captured data locally. After the shooting task is completed, the storage medium is connected to the server 30 to upload the data. The method for regional biodiversity monitoring based on drone technology provided in this specification runs on the server 30. The server 30 can be an independent physical server 30, or a server 30 cluster or distributed system composed of multiple physical servers 30, or a cloud server 30 that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks, as well as big data and artificial intelligence platforms.

[0048] Since this application involves some professional terms, these professional terms will be introduced below.

[0049] Aerial images

[0050] Aerial imagery refers to images of the earth's surface taken from the air, usually by airplanes, helicopters, unmanned aerial vehicles (UAVs) or other aircraft. Aerial images are widely used in many fields, especially in ecological monitoring, environmental management, urban planning, agriculture, disaster assessment, etc. Aerial images have the characteristics of wide viewing angle, high resolution, and multi-angle shooting. Modern aerial photography equipment is able to take high-resolution images that can clearly show the details of the ground. By regularly taking images of the same area, it is possible to monitor vegetation coverage, species distribution, forest health, etc., providing a basis for ecological protection.

[0051] Plant population

[0052] Plant population refers to the collection of all individuals of the same plant in a specific area. Specifically, plant populations have the characteristics of homogeneity, geographical concentration, and ecological consistency. All individuals in a population belong to the same species and have the same genetic and morphological characteristics. They are usually distributed in relatively concentrated geographical areas, with a certain spatial proximity to each other. Individuals in a population live in similar ecological environments and are affected by the same ecological factors (such as light, water, soil type, etc.). The study of plant populations is of great significance for understanding plant growth and development, reproductive strategies, competitive relationships, and the functions and stability of ecosystems. Through the study of plant populations, we can reveal the distribution patterns, population dynamics, genetic diversity, and interactions with other biological and abiotic factors of species, thereby providing a scientific basis for ecological protection, agricultural production, and resource management.

[0053] Ecological Model

[0054] Ecological models are mathematical models used to simulate the interactions between various organisms in an ecosystem and the environment. These models help understand and predict the behavior and changes of ecosystems by quantifying and simulating ecological processes. Ecological models include biological and environmental factors and the interactions between the two. Ecological models describe the population dynamics, including birth rates, mortality rates, immigration and emigration, etc.

[0055] Food Chain

[0056] The food chain is one of the basic concepts that describe the flow of energy and matter in an ecosystem. It represents the order in which different organisms are connected by food relationships. The food chain reflects the predator-prey relationship between different organisms in an ecosystem and the process of energy transfer from one organism to another. The basic components of a food chain include producers, primary consumers, secondary consumers, and tertiary consumers. Producers are green plants that convert solar energy into chemical energy through photosynthesis and store it in organic matter. Primary consumers are herbivores that directly feed on producers, such as rabbits. Secondary consumers mainly feed on primary consumers. For example, foxes prey on rabbits. Tertiary consumers prey on secondary consumers. For example, eagles prey on foxes.

[0057] First, this manual provides a method for regional biodiversity monitoring based on drone technology. Figure 3 , including the steps of:

[0058] Step S101) reading a plurality of aerial images within an area, wherein the aerial images include visible light aerial images and infrared aerial images.

[0059] The drone 20 can be equipped with a variety of shooting equipment, such as visible light cameras, infrared cameras, radars, etc. This embodiment requires the use of visible light cameras and infrared cameras. Visible light cameras are sufficient for distinguishing plant species and areas. However, animals are not only usually on the move, but many types of animals hide in their habitats most of the time. The use of infrared cameras can directly capture the outline of animals, and the type and number of animals can be identified through the outline. Combining visible light aerial images and infrared aerial images can provide data support for regional biodiversity monitoring.

[0060] Step S102) identifying plant populations in an area based on the visible light aerial image, and obtaining predicted animal species and predicted sizes based on the plant populations.

[0061] During the aerial photography process, the drone 20 takes photos while flying. The same place will be photographed multiple times, and the angle of each photo will be slightly different. When the drone 20 flies at a higher altitude, the difference in angle will be less obvious, but it will not affect the implementation of this embodiment. Since the position of the plant will not change, each identified plant can be marked and distinguished by its position. For plants that are difficult to distinguish individually, the distribution range can be used to mark the range position. For example, use the range to mark the grassland, and use the position coordinates to mark a tree or multiple trees that are close in distance.

[0062] Because animals move, they may appear in different positions in multiple aerial images, and the distance of this position change may be relatively large. If the individual animals are not labeled, the individual may be counted repeatedly. If the individual animals are labeled, it will be time-consuming and laborious, not only consuming a lot of computing power, but also difficult to guarantee accuracy. For example, when a bird flies from one place to another, it will appear in multiple aerial images. When there is only one individual bird in the image, it is easy to distinguish the individual bird labels. When there are multiple individual birds in the image, and they fly in different directions or there are individual birds flying from other places, it is difficult to accurately distinguish and identify the individual birds because the differences between the individual birds are small.

[0063] The present application solves the problem by dividing the sub-regions 11. Specifically, the method for identifying plant populations in an area based on the visible light aerial image includes:

[0064] Acquire a plurality of the visible light aerial images, and perform brightness and color correction on the visible light aerial images;

[0065] The positions of the multiple visible light aerial images are aligned to obtain position annotations of the visible light aerial images.

[0066] The stitched image is divided into a plurality of sub-regions 11, and a plurality of images corresponding to each sub-region 11 are obtained according to the position marking, and are recorded as sub-region images. Figure 4 As shown, after position alignment, the sub-region 11 divided by position will correspond to multiple images, and the perspectives of the multiple images will include the side perspective, the perspective close to the top, the perspective of the back, etc., so that images of plants from multiple angles can be seen. The sub-region 11 is divided by position, and multiple images of the same sub-region 11 are intercepted from multiple aerial images and recorded as sub-region images for centralized analysis of one sub-region 11.

[0067] The pre-connected image recognition model is used to respectively identify the plants in the multiple sub-region images corresponding to each sub-region 11, and the plant position annotation is obtained according to the position annotation. According to the position coordinates of the boundary of the sub-region 11 and the ratio of the plant position to the boundary of the region 10, the position of the plant can be approximately calculated.

[0068] Plants identified in multiple sub-region images are fused according to plant position annotation to obtain plant identification results for each sub-region 11. The positions of the same plant calculated in different sub-region images will be slightly different, and fusion can be performed if they do not exceed a preset error range.

[0069] According to the plant identification results of all sub-areas 11, the plant population in the area is obtained.

[0070] After the plant identification results of all sub-regions 11 are summarized, the types and quantities of all plants in the region can be obtained.

[0071] On the other hand, this embodiment provides a method for obtaining predicted animal species and predicted scales based on the plant population, specifically comprising:

[0072] The type and population size of each plant population are obtained respectively, and the population size is expressed by the number or coverage area. The number of some plant types is expressed by coverage area, such as daylilies and water grasses.

[0073] Read the pre-imported food chain data, and list the animal types and population sizes of herbivores or omnivores that can be supported according to the type and population size of the plant population. For example, there are 13 cherry trees in the area. According to the food chain data, each cherry tree can support 0.05 woodpeckers or 3 butterflies. The food chain data of cherry trees, woodpeckers 41 and butterflies are obtained from historical statistical data. The food chain data records the quantitative ratio relationship and food chain relationship between species. For example, during the past 10 years, by manually counting the relationship between cherry trees and woodpeckers 41 in multiple areas 10, the corresponding relationship between cherry trees and woodpeckers 41 is 1:0.05 and the relationship with butterflies is 1:3. Although such statistics cannot accurately reflect the quantitative correspondence between plants and animals, the data obtained based on statistics are sufficiently practical for ecological monitoring. Therefore, in this embodiment, when it is obtained that there are 13 cherry trees in the area, it will be concluded that the area can support 0.65 woodpeckers and 39 butterflies. Of course, there are other tree species in the area. For example, if there are 36 pine trees, and according to historical food chain data, the ratio of pine trees to woodpecker 41 is 1:0.3, it can be concluded that the area can support 10.8 woodpeckers. Assuming that there are no other plants, it can be concluded that the area can support 2.45 woodpeckers. That is, one predicted animal species is woodpecker 41, and its predicted size is 11.45. Another predicted animal species is butterfly, and its predicted size is 39.

[0074] According to the animal type and population size of the herbivores or omnivores, combined with the food chain data, the animal type and population size of the carnivores are obtained. Assuming that the food chain data shows that the number relationship between butterflies and spiders is 1:0.03, it can be obtained that there are 1.17 spiders. Spiders are insects. In this embodiment, carnivorous insects are counted as carnivores only according to food type.

[0075] According to the animal types and population sizes of the herbivores, omnivores, and carnivores, the predicted animal species and their predicted sizes are obtained. In area 10, the predicted animal species is 41 woodpeckers, with a predicted size of 11.45; the predicted animal species is butterflies, with a predicted size of 39; and the predicted animal species is spiders, with a predicted size of 1.17.

[0076] Step S103) identifying individual animals appearing in the visible light aerial image and the infrared aerial image, and obtaining the types and numbers of the individual animals.

[0077] For details, please refer to the attached Figure 5 The method for identifying individual animals appearing in the visible light aerial image and the infrared aerial image and obtaining the types and numbers of the individual animals includes:

[0078] Step S1031) aligning the positions of the plurality of visible light aerial images and infrared aerial images to obtain position annotations of the visible light aerial images and infrared aerial images. By using the position of the drone 20 when the visible light aerial images and infrared aerial images were taken, the visible light aerial images and infrared aerial images can be aligned and their positions can be annotated.

[0079] Step S1032) Obtain the position of the individual animal according to the infrared aerial image and the position mark, and obtain the visible light aerial image corresponding to the position, which is recorded as a captured image. Infrared aerial images can directly locate the position of the animal according to the temperature, but infrared aerial images can only provide the outline of the animal with blurred boundaries. Therefore, it is difficult to identify the type of animal, that is, the appearance species, only by relying on infrared aerial images. Use the visible light aerial image aligned with the infrared aerial image to obtain the visible light aerial image corresponding to the animal, that is, the captured image. Then use the image recognition model to recognize the captured image to obtain the type of animal, that is, the appearance species.

[0080] Step S1033) According to the response of the pre-connected image recognition model to the captured image, the type and location of the individual animal are obtained.

[0081] Step S1034) All the animal individuals identified by the visible light aerial images and the infrared aerial images are fused according to their positions, and the number of animal individuals of each type is counted to obtain the types and numbers of the animal individuals.

[0082] The occurrence species indicates that animals of this species have been observed. Since the predicted animal species have been obtained in the previous step, the animal species present in the area can be determined by combining the occurrence species. Specifically, although butterflies are included in the predicted animal species, butterflies have never been photographed, assuming that it is currently spring and summer. In the subsequent steps, butterflies will not be included in the ecological model of area 10. If woodpecker 41 is photographed, woodpecker 41 will be included in the ecological model of area 10, and the number of woodpecker 41 will be further determined by subsequent steps.

[0083] Since animals are mobile, please refer to the attached Figure 6, the same woodpecker was photographed in one sub-area 11, and then flew to another sub-area 11 and was photographed again. Since it is very difficult to distinguish and mark individuals, individuals are not distinguished in this embodiment, so the woodpecker 41 will be counted twice in this embodiment and is considered to be two woodpeckers. The difference in quantity caused by this will be compensated and corrected in subsequent steps. On the other hand, after identifying the type and position of the individual animal, the individual animals appearing in multiple aerial images are fused by position. That is, if the position change of the animal is within the preset range, it is considered that the same animal is moving and will not be counted as two animals. The condition for fusion, that is, the condition for being considered to be the same animal, is that it appears in different aerial images and the position change does not exceed the preset threshold. The same animal will not appear twice in the same aerial image, so fusion is the fusion of individual animals appearing in different aerial images.

[0084] Step S104) According to the occurrence type, occurrence quantity, predicted animal type and predicted scale, each occurrence type, likelihood scale and confidence level are obtained.

[0085] For details, please refer to the attached Figure 7 According to the occurrence species, occurrence quantity, predicted animal species and predicted scale, the method for obtaining the probability scale and confidence of each occurrence species includes:

[0086] Step S1041) Divide a single day into several time periods, and obtain the equivalent number of outings of a single animal in each time period based on the predicted animal species and animal life habits. Among them, the animal life habits include one or more of daytime activities, evening activities, nighttime activities, flying, large-scale foraging, small-scale foraging, not hiding, and hiding. For example, the animal life habits of woodpecker 41 include: daytime activities, flying, large-scale foraging, and not hiding. Another example is the animal life habits of voles, including: daytime activities, nighttime activities, small-scale foraging, and hiding. There are many types of voles, some of which are active and foraging during the day, some are active and foraging at night, and some are active and foraging both during the day and at night. Another example is the animal life habits of sheep, including: daytime activities, large-scale foraging, and no hiding. The distinction between large-scale foraging and small-scale foraging is made based on sub-area 11. If the animal's single foraging range is not larger than the range of sub-area 11, it is foraging in a small range. On the contrary, if the animal's single foraging range is larger than the range of sub-area 11, it is foraging in a large range. For example, a vole's foraging usually does not exceed the range of a sub-area 11, so it is foraging in a small range. However, a sheep's foraging is likely to exceed the range of a sub-area 11, so it is foraging in a large range. Of course, the animal's living habits may change according to the size of the sub-area 11. And the division of the animal's living habits is based on most cases. Whether resting or foraging, voles are usually in a hiding state, while sheep do not hide when resting or foraging. Therefore, even if there are more voles in the area, fewer voles are photographed, that is, the equivalent number of times the vole goes out is less.

[0087] The daytime activities, evening activities, night activities, flying, large-scale foraging, small-scale foraging, not hiding, hiding in the life habits of animals are the basis for influencing and determining the equivalent number of times of going out. Daytime activities, evening activities, and night activities can distinguish time periods. For example, dragonflies are active in the evening, and a small number of owls can also be seen active in the evening. Daytime activities mean that the equivalent number of times of going out in the evening and night is 0. Evening activities mean that the equivalent number of times of going out in the daytime and night is 0. Night activities mean that the equivalent number of times of going out in the daytime and evening is 0.

[0088] Large-scale foraging means that a single animal has a large number of equivalent outing times, for example, 3, which means that a single animal will appear in multiple sub-areas 11 during a foraging process, and will be photographed in 3 sub-areas 11 on average during the drone 20 shooting process. The specific equivalent number of outings can be obtained based on the historical shooting data after manual annotation. After obtaining the equivalent number of outings of some animals whose living habits are large-scale foraging, it can be inferred to the equivalent number of outings of other animals whose living habits are large-scale foraging. For example, sheep are observed and photographed, and it is obtained that sheep will be photographed in 3 sub-areas 11 on average during the daytime period during a drone 20 shooting process. Therefore, cattle, horses, deer, dogs, rabbits, etc. can all be determined with reference to the equivalent number of outings of sheep. Specifically, it is adjusted according to the movement rate and frequency of animals. The movement speed and movement frequency of cattle are lower than those of sheep. The equivalent number of outings of sheep after manual observation and annotation is 3 at this time, so cattle can be set to 2, that is, lower than sheep. The movement speed and frequency of rabbits are higher than those of sheep, so the equivalent number of outings is set to 5.

[0089] If the animal has a flying animal life habit, the equivalent number of outings obtained according to daytime activities, evening activities, nighttime activities, large-scale foraging, and small-scale foraging is multiplied by a coefficient greater than 1. For example, the equivalent number of outings of a single animal of woodpecker 41 is 6, and then because woodpecker 41 can fly, it is further multiplied by a coefficient greater than 1. For example, the coefficient is 1.2, and the final equivalent number of outings of woodpecker 41 is 7.2.

[0090] If the animal has a hiding habit, the equivalent number of outings obtained according to daytime activities, evening activities, nighttime activities, large-scale foraging, and small-scale foraging is multiplied by a coefficient less than 1. For example, the equivalent number of outings of the vole is 1, and then multiplied by a coefficient less than 1. For example, the coefficient is 0.2, and the final equivalent number of outings of the vole is 0.2.

[0091] If the animal has a non-hiding animal life habit, the equivalent number of outings obtained based on daytime activities, evening activities, nighttime activities, large-scale foraging, and small-scale foraging is multiplied by a coefficient equal to 1. For example, cattle have a non-hiding animal life habit, so the equivalent number of outings for cattle is 2, which is the final equivalent number of outings.

[0092] Step S1042) According to the equivalent number of outings and the predicted scale, the total equivalent number of outings of each animal type in the area 10 in each time period is obtained. That is, the total equivalent number of outings of each animal type in the area 10 in each time period can be obtained by multiplying the equivalent number of outings by the predicted scale. For example, the equivalent number of outings of the woodpecker 41 is 7.2, and the predicted scale of the woodpecker 41 is 11.45, so the total equivalent number of outings is 7.2×11.45=82.44 times.

[0093] Step S1043) According to the shooting time of the aerial image, the total equivalent number of outings of each animal type in the corresponding time period is obtained.

[0094] Step S1044) Calculate the ratio of the number of occurrences to the total outbound equivalent number of times.

[0095] Assuming that the aerial image is taken during the day, the total equivalent number of times the woodpecker 41 goes out during the corresponding period is 82.44. After the actual image is taken, the number of appearances obtained through the above steps is 68. The ratio of the number of appearances to the total equivalent number of times of going out is 68 divided by 82.44, which is 0.825.

[0096] Another aerial image was taken in the evening, so the total equivalent number of outings of woodpecker 41 in the corresponding period is 0. After actual sorting, the number of occurrences is 1. This number of occurrences of 1 is considered an unexpected event by this embodiment and will not be used to calculate the likelihood scale. In actual calculation, the ratio of the number of occurrences to the total equivalent number of outings cannot be calculated, because the total equivalent number of outings at this time is 0 and cannot be used as the denominator for calculation.

[0097] Step S1045) Multiply the predicted scale by the ratio as the likelihood scale, and calculate the proximity of the ratio to 1 as the confidence. The ratio corresponding to Woodpecker 41 is 0.825, so the confidence is (1-(1-0.825))÷1×100%, and the calculated confidence is 82.5%.

[0098] The predicted scale of the woodpecker 41 obtained above is 11.45, and the ratio is 0.825, so the likelihood scale of the woodpecker 41 is 11.45×0.825=9.44625, which is rounded to 9. On the other hand, in another embodiment, 9.44625 may be directly used as the likelihood scale of the woodpecker 41 without rounding.

[0099] When no butterfly is observed in the area, the animal species of the spider is obtained as butterfly, and the number of occurrences is 0. Even if the predicted number of butterflies is 39, the calculated likelihood number is 0.

[0100] Step S105) Establish an ecological model of region 10 according to the likelihood scale of all the occurrence species. The ecological model of region 10 records the occurrence species, likelihood scale and confidence of all plant populations and animals. The ecological model of region 10 can be established by directly recording these data. It can also be established by using structured data for recording.

[0101] Step S106) According to the food chain relationship of the ecological model of the region 10 and the confidence of each occurrence species, the likelihood scale of each occurrence species is corrected.

[0102] Please see attached Figure 8 According to the food chain relationship of the ecological model of the region 10 and the confidence of each occurrence species, the method of correcting the likelihood scale of each occurrence species includes:

[0103] Step S1061) Obtain the scale ratio between all occurring species according to the food chain relationship of the ecological model of the area 10. Through the recording of historical data and manual marking, the scale ratio between different species in the food chain relationship can be obtained. In the actual ecology, the scale ratio is not a fixed value, but a range value. However, in this embodiment, the average value calculated based on the historical data records is used as the scale ratio, so that the scale ratio in this embodiment is a fixed value. Exemplarily, sparrowhawks were also observed in area 10, and the scale ratio between sparrowhawks and woodpeckers 41 was 0.092.

[0104] Step S1062) Select an occurrence category, and calculate the revised scales of other occurrence categories based on the likelihood scale of the selected occurrence category and the scale ratio.

[0105] Step S1063) Calculate the proximity between the revised scale and the likelihood scale of each occurrence category, and use the proximity to revise the confidence of the occurrence category.

[0106] Step S1064) The sum of the confidences of all the appearing categories is used as the confidence of the ecological model of region 10.

[0107] Exemplarily, the likelihood size of sparrowhawks obtained through the aforementioned steps is 1.325, and the confidence level is 75.35%.

[0108] When the sparrowhawk is selected as the occurrence species, the scale of the woodpecker 41 is corrected, and the corrected scale of the woodpecker 41 is 1.325÷0.092=14.4. The closeness between the corrected scale of the woodpecker 41 and the likelihood scale is (14.4-(14.4-9.44625))÷14.4×100%=65.60%. The corrected scale of the sparrowhawk is equal to the likelihood scale, so the closeness is 1. The confidence of the occurrence species is corrected using the proximity, that is, 65.60% is multiplied by the confidence of the woodpecker 41 to correct the confidence of the woodpecker 41, and the corrected confidence of the woodpecker 41 is 82.5%×65.60%=54.12%. The corrected confidence of the sparrowhawk is still 75.35%. The confidence of the ecological model of region 10 is 54.12% + 75.35%, plus the confidence of other species. It is worth noting that when calculating the confidence of the ecological model of region 10, the influence of plants on the confidence is not considered, because plants can be accurately observed and their confidence is fixed at 1, so they are not considered. Assuming that only woodpecker 41 and sparrowhawk are considered here, the confidence of the ecological model of region 10 is 54.12% + 75.35% = 129.47%, that is, the confidence is 1.2947.

[0109] On the other hand, when woodpecker 41 is selected as the occurrence species, the scale of sparrowhawk is corrected, and the corrected scale of sparrowhawk is 9.44625×0.092=0.869. The closeness between the corrected scale of sparrowhawk and the likelihood scale is (1.325-(1.325-0.869))÷1.325×100%=65.58%. The confidence of the occurrence species is corrected using the proximity, that is, 65.58% is multiplied by the confidence of sparrowhawk to correct the confidence of sparrowhawk, and the corrected confidence of sparrowhawk is 75.35%×65.58%=49.41%. The corrected confidence of woodpecker 41 is still 82.5%. Then the confidence of the ecological model of region 10 is 49.41%+82.5%, plus the confidence of other occurrence species. Assuming that only woodpeckers 41 and sparrowhawks are considered here, the confidence level of the ecological model of region 10 is 49.41% + 82.5% = 131.91%, that is, the confidence level is 1.3191.

[0110] Step S1065) Based on the ecological model of region 10 with the highest confidence, the likelihood scale of each occurrence type is corrected.

[0111] Step S107) obtaining regional biodiversity monitoring results based on the plant population and the regional ecological model after the likelihood scale is corrected.

[0112] It can be seen that when woodpecker 41 is selected as the occurrence species, the confidence of the ecological model of area 10 after correction and final calculation is the highest. Therefore, woodpecker 41 is finally selected as the occurrence species, and the likelihood scale of other occurrence species is corrected. That is, the likelihood scale of woodpecker 41 is finally determined to be 9.44625, and the likelihood scale of sparrowhawk is 0.869. On the other hand, in another embodiment, the likelihood scale of woodpecker 41 is 9 and the likelihood scale of sparrowhawk is 1 by rounding. By analogy, the likelihood scale of all plant populations and animals recorded in the revised ecological model of area 10 is obtained, and the regional biodiversity monitoring results are obtained. In this embodiment, 13 cherry trees, 36 pine trees, 0 butterflies, 39 butterflies, 419 woodpeckers, and 1 sparrowhawk are obtained. And other unspecified plants and animals.

[0113] On the other hand, this manual provides a system for regional biodiversity monitoring based on drone technology. Fig. 9 ,include:

[0114] The shooting module 100 reads a plurality of aerial images in the area, wherein the aerial images include visible light aerial images and infrared aerial images;

[0115] A first recognition module 200 is configured to recognize plant populations in an area according to the visible light aerial image, and obtain predicted animal species and predicted scales according to the plant populations;

[0116] The second recognition module 300 recognizes individual animals that appear in the visible light aerial image and the infrared aerial image, and obtains the type and number of the individual animals that appear;

[0117] Prediction module 400, obtaining each occurrence type, likelihood scale and confidence level according to the occurrence type, occurrence quantity, predicted animal type and predicted scale;

[0118] The ecological module 500 establishes an ecological model of the region 10 according to the likelihood scale of all the occurrence species;

[0119] A correction module 600, which corrects the likelihood scale of each occurrence species according to the food chain relationship of the ecological model of the region 10 and the confidence of each occurrence species;

[0120] The result module 700 obtains the regional biodiversity monitoring result according to the plant population and the regional ecological model after the likelihood scale is corrected.

[0121] See also Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification is shown.

[0122] like Fig.10As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105 and at least one communication bus 1102. Among them, the communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. Among them, the user interface 1103 may include a button, and the optional user interface may also include a standard wired interface and a wireless interface. Among them, the network interface 1104 may include but is not limited to a Bluetooth module, an NFC module, a Wi-Fi module, etc. Among them, the processor 1101 may include one or more processing cores. The processor 1101 uses various interfaces and lines to connect various parts in the entire electronic device 1100, and executes various functions and processes data of the routing device 1100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1101 may integrate one or a combination of CPU, GPU and modem, etc. Among them, the CPU mainly processes the operating system, user interface and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication.

[0123] It is understandable that the above-mentioned modem may not be integrated into the processor 1101, but may be implemented by a separate chip.

[0124] Among them, the memory 1105 may include RAM or ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1105 may also be at least one storage device located away from the aforementioned processor 1101. The memory 1105 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application. The processor 1101 may be used to call the application stored in the memory 1105 and execute the methods in the above-mentioned multiple embodiments.

[0125] The embodiments of this specification also provide a computer-readable storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the multiple steps in the above embodiments. If the components of the above electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0126] The embodiments of this specification also provide a computer program product, including a computer program, which implements multiple steps in the above embodiments when executed by a processor.

[0127] In the absence of conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0128] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of this specification is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server 30 or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server 30 or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server 30, a data center, etc. that includes multiple available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)), etc.

[0129] When implemented by hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and realize the corresponding function. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit, and its logic function is determined by the user programming the device. The designer programs by himself to "integrate" a digital system on a PLD, without asking a chip manufacturer to design and make a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when writing program development, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL), and HDL is not just one, but many. Those skilled in the art should also be aware that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit to easily obtain the hardware circuit that implements the logic method flow.

[0130] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A method for regional biodiversity monitoring based on drone technology, characterized in that: Includes steps: Reading multiple aerial images in the area, wherein the aerial images include visible light aerial images and infrared aerial images; Identify plant populations in an area based on the visible light aerial image, and obtain predicted animal species and their predicted sizes based on the plant populations; Identify individual animals that appear in the visible light aerial image and the infrared aerial image, and obtain the type and number of the individual animals that appear; According to the occurrence types, occurrence numbers, predicted animal species and predicted scales, obtaining each occurrence type, likelihood scale and confidence level; Establish regional ecological models based on the likelihood scale of all described species; Correcting the likelihood scale of each occurrence species according to the food chain relationship of the regional ecological model and the confidence of each occurrence species; Based on the plant population and the regional ecological model after correcting the likelihood scale, the regional biodiversity monitoring results are obtained.

2. The method for regional biodiversity monitoring based on drone technology according to claim 1 is characterized in that: The method for identifying plant populations in an area based on the visible light aerial image comprises: Acquire a plurality of the visible light aerial images, and perform brightness and color correction on the visible light aerial images; Aligning the positions of the plurality of visible light aerial images to obtain position annotations of the visible light aerial images; Divide the stitched image into multiple sub-regions, and obtain multiple images corresponding to each sub-region according to the position annotations, which are recorded as sub-region images; Using a pre-connected image recognition model, respectively identifying plants in a plurality of sub-region images corresponding to each sub-region, and obtaining plant position annotations according to the position annotations; The plants identified by fusing multiple sub-region images are labeled according to the plant positions to obtain the plant identification results for each sub-region; According to the plant identification results of all sub-areas, the plant population in the area is obtained.

3. A method for regional biodiversity monitoring based on drone technology according to claim 1 or 2, characterized in that: Methods for obtaining predicted animal species and their predicted sizes based on the plant population include: respectively obtaining the type and population size of each plant population, wherein the population size is expressed in terms of number or coverage area; Read the pre-imported food chain data, and list the animal types and population sizes of herbivores or omnivores that can be supported based on the types and population sizes of plant populations; According to the animal type and population size of the herbivores or the omnivores, in combination with the food chain data, the animal type and population size of the carnivores are obtained; According to the animal types and population sizes of the herbivores, the omnivores and the carnivores, predicted animal species and predicted sizes thereof are obtained.

4. A method for regional biodiversity monitoring based on drone technology according to claim 1 or 2, characterized in that: The method of identifying individual animals appearing in the visible light aerial image and the infrared aerial image and obtaining the type and number of individual animals includes: Aligning the positions of the plurality of visible light aerial images and infrared aerial images to obtain position annotations of the visible light aerial images and infrared aerial images; Obtaining the position of the individual animal according to the infrared aerial image and the position mark, and obtaining a visible light aerial image corresponding to the position, which is recorded as a captured image; Obtaining the type and location of the individual animal according to the response of the pre-connected image recognition model to the captured image; The individual animals identified in all the visible light aerial images and infrared aerial images are fused according to their positions, and the number of individual animals of each type is counted to obtain the types and numbers of individual animals.

5. A method for regional biodiversity monitoring based on drone technology according to claim 1 or 2, characterized in that: According to the occurrence species, occurrence quantity, predicted animal species and predicted scale, the method for obtaining the probability scale and confidence of each occurrence species includes: Divide a single day into several time periods, and obtain the equivalent number of outings of a single animal in each time period based on the predicted animal species and animal living habits; According to the equivalent number of outings and the predicted scale, the total equivalent number of outings of each animal type in the area in each time period is obtained; According to the shooting time of the aerial images, the total equivalent number of outings of each animal type in the corresponding period is obtained; Calculating the ratio of the number of occurrences to the total number of outbound equivalent times; The predicted scale is multiplied by the ratio as the likelihood scale, and the proximity of the ratio to 1 is calculated as the confidence.

6. A method for regional biodiversity monitoring based on drone technology according to claim 1 or 2, characterized in that: According to the food chain relationship of the regional ecological model and the confidence of each occurrence species, the method of correcting the likelihood scale of each occurrence species includes: Obtaining the size ratios among all occurring species according to the food chain relationships of the regional ecological model; Select an occurrence category, and calculate the corrected scales of other occurrence categories based on the likelihood scale of the selected occurrence category and the scale ratio; Calculate the closeness between the corrected scale and the likelihood scale of each occurrence category, and use the closeness to correct the confidence of the occurrence category; The sum of the confidences of all occurring species is used as the confidence of the regional ecological model; The likelihood scale of each occurrence species was revised based on the regional ecological model with the highest confidence.

7. A system for regional biodiversity monitoring based on drone technology, characterized in that: include: A shooting module is used to read multiple aerial images in the area, wherein the aerial images include visible light aerial images and infrared aerial images; A first recognition module is configured to recognize plant populations in an area according to the visible light aerial image, and to obtain predicted animal species and predicted scales according to the plant populations; A second recognition module is used to recognize individual animals that appear in the visible light aerial image and the infrared aerial image, and obtain the type and number of the individual animals that appear; A prediction module, which obtains each occurrence type, likelihood scale and confidence level according to the occurrence type, occurrence quantity, predicted animal type and predicted scale; Ecological module, which builds a regional ecological model based on the likelihood scale of all the described species; A correction module, which corrects the likelihood scale of each occurrence species according to the food chain relationship of the regional ecological model and the confidence of each occurrence species; The result module obtains the regional biodiversity monitoring result according to the plant population and the regional ecological model after the likelihood scale is corrected.

8. An electronic device, characterized in that including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. Computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images