Ecosystem-disturbing plant monitoring method, device and computer program

Abstract

An ecosystem-disturbing plant monitoring method according to one embodiment of the present application comprises the steps of: using a drone having a multispectral sensor, so as to acquire a multispectral image of a region expected to have distribution of ecosystem-disturbing plants; converting the resolution of the multispectral image; using a pre-trained deep learning model to analyze the multispectral image and detect disturbing plants from the multispectral image; and monitoring the detected disturbance plant. According to one embodiment of the present invention, ecosystem-disturbing plants can be effectively monitored.

Description

Method, device, and computer program for monitoring ecosystem-disturbing plants

[0001] The present application relates to a method, apparatus, and computer program for monitoring ecosystem-disturbing plants. Specifically, it relates to a method, apparatus, and computer program for monitoring ecosystem-disturbing plants that can remove ecosystem-disturbing plants and perform continuous monitoring to prevent recurrence after removal.

[0002] Existing monitoring systems for invasive species have primarily relied on direct field surveys or the analysis of low-resolution satellite images. Due to the low resolution of these satellite images, it was difficult to detect small invasive plants or early intrusions, and the long time between data collection and analysis made real-time response challenging. Furthermore, existing systems suffer from the problem of recurring invasive species in the same areas because they lack continuous monitoring and management to prevent recurrence after removal.

[0003] The problem that the present invention aims to solve is to provide a method, device, and computer program for monitoring ecosystem-disturbing plants that can accurately detect even small disturbing plants by providing high-resolution images.

[0004] Furthermore, the objective of the present invention is to provide a method, device, and computer program for monitoring ecosystem-disturbing plants that can immediately respond to the rapid spread of disturbing plants through real-time data transmission and analysis, enable efficient deployment and work instructions for field personnel, and provide accurate data without the need for repetitive field surveys.

[0005] Furthermore, the objective of the present invention is to provide a method, device, and computer program for monitoring ecosystem-disturbing plants that utilize multispectral data to accurately distinguish and rapidly identify disturbing plants from other vegetation, reduce the recurrence rate of disturbing plants through accurate monitoring and continuous management, and enable long-term ecosystem protection through efficient removal operations.

[0006] The problems that the present invention aims to solve are not limited to those described above, and problems not mentioned will be clearly understood by those skilled in the art from this specification and the attached drawings.

[0007] A method for monitoring ecosystem-disturbing plants according to one embodiment of the present invention includes the steps of: acquiring a multispectral image of an area expected to have a distribution of disturbing plants using a drone equipped with a multispectral sensor; converting the resolution of the multispectral image; analyzing the multispectral image using a pre-trained deep learning model to detect the disturbing plants from the multispectral image; and monitoring the detected disturbing plants.

[0008] An ecosystem-disturbing plant monitoring device according to one embodiment of the present invention includes a processor for monitoring detected disturbing plants based on a multispectral image acquired from a drone equipped with a multispectral sensor, and the processor includes an image acquisition module for acquiring a multispectral image of an area expected to have a disturbing plant distribution using the drone, a super-resolution module for converting the resolution of the multispectral image, a detection module for detecting disturbing plants from the multispectral image by analyzing the multispectral image using a pre-trained deep learning model, and a monitoring module for monitoring the detected disturbing plants.

[0009] The means for solving the problem of the present invention are not limited to the means for solving the problem described above, and unmentioned means for solving the problem will be clearly understood by those skilled in the art from this specification and the attached drawings.

[0010] According to one embodiment of the present invention, high-resolution images can be provided to accurately detect even small disturbing plants, and the rapid spread of disturbing plants can be immediately responded to through real-time data transmission and analysis.

[0011] In addition, according to one embodiment of the present invention, efficient deployment and work instructions for field personnel are possible, and accurate data can be provided without repetitive field surveys.

[0012] In addition, according to one embodiment of the present invention, by utilizing multispectral data, it is possible to accurately distinguish and rapidly identify disturbing plants and other vegetation, reduce the recurrence rate of disturbing plants through accurate monitoring and continuous management, and enable long-term ecosystem protection through efficient removal operations.

[0013] FIG. 1 is a diagram showing the schematic components of an ecosystem-disturbing plant monitoring system according to one embodiment of the present application.

[0014] FIG. 2 is a schematic block diagram of an ecosystem-disturbing plant monitoring device according to one embodiment of the present application.

[0015] FIG. 3 is a schematic block diagram of a processor equipped in an ecosystem-disturbing plant monitoring device according to one embodiment of the present application.

[0016] FIG. 4 is a flowchart illustrating a method for monitoring ecosystem-disturbing plants according to one embodiment of the present application.

[0017] FIGS. 5(a) and FIGS. 5(b) are drawings for illustrating an example of converting the resolution of a multispectral image according to one embodiment of the present application.

[0018] FIGS. 6(a) to 6(e) are drawings for illustrating an example of a multispectral image according to an embodiment of the present application.

[0019] FIG. 7 is a drawing for explaining an example of an image stack according to one embodiment of the present application.

[0020] FIGS. 8 and 9 are drawings for explaining an example of analyzing drone images using the U-net algorithm according to an embodiment of the present application.

[0021] FIG. 10 is a drawing for illustrating an example of pre-training data according to one embodiment of the present application.

[0022] FIG. 11 is a drawing for illustrating an example of an image chip of pre-training data according to one embodiment of the present application.

[0023]

[0024] The aforementioned objectives, features, and advantages of the present application will become more apparent from the following detailed description in conjunction with the accompanying drawings. However, as the present application is subject to various modifications and may have various embodiments, specific embodiments are illustrated in the drawings and described in detail below.

[0025] Throughout the specification, identical reference numbers generally represent identical components. Additionally, components with identical functions within the same scope of concept appearing in the drawings of each embodiment are described using the same reference numeral, and redundant descriptions thereof are omitted.

[0026] If it is determined that a detailed description of known functions or configurations related to this application could unnecessarily obscure the essence of this application, such detailed description is omitted. Furthermore, numbers used in the description of this specification (e.g., First, Second, etc.) are merely identifiers to distinguish one component from another.

[0027] Furthermore, the suffixes "module" and "part" for components used in the following embodiments are assigned or used interchangeably solely for the ease of drafting the specification, and do not inherently possess distinct meanings or roles.

[0028] In the following examples, singular expressions include plural expressions unless the context clearly indicates otherwise.

[0029] In the following embodiments, terms such as "include" or "have" mean that the features or components described in the specification are present, and do not preclude the possibility that one or more other features or components may be added.

[0030] In the drawings, the size of components may be exaggerated or reduced for convenience of explanation. For example, the size and thickness of each component shown in the drawings are arbitrarily depicted for convenience of explanation, and the present invention is not necessarily limited to what is illustrated.

[0031] Where an embodiment can be implemented differently, the order of a particular process may be performed differently from the order described. For example, two processes described consecutively may be performed substantially simultaneously or proceed in the reverse order of the description.

[0032] In the following embodiments, when components are described as being connected, the case includes not only instances where the components are directly connected but also instances where components are indirectly connected by interposing them in between.

[0033] For example, when it is stated in this specification that components, etc. are electrically connected, it includes not only cases where the components, etc. are directly electrically connected, but also cases where components, etc. are interposed in between and are indirectly electrically connected.

[0034] Hereinafter, the method, apparatus, and system for monitoring ecosystem-disturbing plants of the present application will be described with reference to FIGS. 1 to 11.

[0035] FIG. 1 is a diagram showing the schematic components of an ecosystem-disturbing plant monitoring system according to one embodiment of the present application.

[0036] Referring to FIG. 1, an ecosystem-disturbing plant monitoring system (1) according to one embodiment of the present application may include a drone (200) and an ecosystem-disturbing plant monitoring device (100). However, the present invention is not limited thereto, and the ecosystem-disturbing plant monitoring system (1) may include additional components or some components may be omitted. Some components of the ecosystem-disturbing plant monitoring system (1) may be separated into a plurality of devices, or a plurality of components may be merged into a single device. For example, the ecosystem-disturbing plant monitoring device (100) may represent a server device.

[0037] The drone (200) may represent a drone device that acquires multispectral images of an area where invasive plants are expected to be distributed. For example, the drone (200) may be equipped with a multispectral sensor. For example, the drone (200) may be a DJI M3M drone equipped with a multispectral sensor. Additionally, the drone (200) may be equipped with an RGB sensor. The drone (200) may acquire multispectral images using the multispectral sensor and the RGB sensor. For example, although one drone (200) is shown in FIG. 1, the ecosystem invasive plant monitoring system (1) may include two or more multiple drones (200).

[0038] The drone (200) and the ecosystem disturbing plant monitoring device (100) can be connected to each other via a network to exchange data. For example, the ecosystem disturbing plant monitoring device (100) can acquire multispectral images from the drone (200).

[0039] An ecosystem disturbing plant monitoring device (100) can store location data by mapping multispectral images. Here, the location data may represent geographical information where the multispectral image was captured. Additionally, the ecosystem disturbing plant monitoring device (100) can convert the multispectral image mapped with location data into high resolution. Additionally, the ecosystem disturbing plant monitoring device (100) can detect disturbing plants by analyzing the high-resolution converted multispectral image using a pre-trained deep learning algorithm (e.g., U-Net model). Additionally, the ecosystem disturbing plant monitoring device (100) can direct the removal of disturbing plants based on information about the disturbing plants detected in real time. Additionally, the ecosystem disturbing plant monitoring device (100) can generate a work completion report after the removal of disturbing plants is completed. Additionally, the ecosystem disturbing plant monitoring device (100) can monitor information about the disturbing plants before and after the removal of disturbing plants.

[0040] FIG. 2 is a schematic block diagram of an ecosystem-disturbing plant monitoring device according to one embodiment of the present application. FIG. 3 is also a schematic block diagram of a processor equipped in an ecosystem-disturbing plant monitoring device according to one embodiment of the present application.

[0041] First, referring to FIG. 2, an ecosystem disturbing plant monitoring device (100) according to one embodiment of the present application may include a memory (110), a processor (120), and a communication module (130). The ecosystem disturbing plant monitoring device (100) may be a server device provided in an ecosystem disturbing plant monitoring system (1).

[0042] The communication module (130) can provide a function for communicating with an external device through a network. For example, the ecosystem-disturbing plant monitoring device (100) can connect to a network through the communication module (130) to transmit and receive various data. The transmitting and receiving unit may include a wired type and a wireless type. Since the wired type and the wireless type each have their own advantages and disadvantages, the ecosystem-disturbing plant monitoring device (100) may be equipped with both wired and wireless types depending on the case. Here, in the case of the wireless type, communication methods of the WLAN (Wireless Local Area Network) series, such as Wi-Fi, can be mainly used. Alternatively, in the case of the wireless type, cellular communication, such as LTE or 5G series communication methods, can be used. However, wireless communication protocols are not limited to the examples described above, and it is possible to use any appropriate wireless type of communication method. In the case of the wired type, LAN (Local Area Network) or USB (Universal Serial Bus) communication are representative examples, and other methods are also possible.

[0043] Memory (110) is a computer-readable recording medium, and memory (110) can store various information. Various data can be stored in memory (110) temporarily or semi-permanently. Examples of memory (110) may include a hard disk drive (HDD), a solid state drive (SSD), flash memory, ROM (Read-Only Memory), and RAM (Random Access Memory). Memory (110) may be provided in a form that is embedded in the ecosystem-disturbing plant monitoring device (100) or in a detachable form. Memory (110) may store various data necessary for the operation of the ecosystem-disturbing plant monitoring device (100), including an operating system (OS) for operating the ecosystem-disturbing plant monitoring device (100) and a program for operating each component of the ecosystem-disturbing plant monitoring device (100).

[0044] Referring to FIG. 2 and FIG. 3 together, a processor (120) according to one embodiment of the present application can control the overall operation of an ecosystem disturbing plant monitoring device (100). For example, the processor (120) can monitor the detected disturbing plants based on multispectral images acquired from a drone (200). More specifically, the processor (120) may include an image acquisition module (121), a super-resolution module (122), a detection module (123), and a monitoring module (124).

[0045] The processor (120) can load and execute a program for the overall operation of the ecosystem-disturbing plant monitoring device (100) from memory (110). The processor (120) can be implemented as an Application Processor (AP), a Central Processing Unit (CPU), a Microcontroller Unit (MCU), or a similar device depending on hardware, software, or a combination thereof. In this case, hardware-wise, it may be provided in the form of an electronic circuit that processes electrical signals to perform control functions, and software-wise, it may be provided in the form of a program or code that drives the hardware circuit.

[0046] The image acquisition module (121) can acquire multispectral images of the expected distribution area of ​​disturbed plants using a drone. Additionally, the super-resolution module (122) can convert the resolution of the multispectral images acquired from the drone.

[0047] The detection module (123) can detect disturbing plants from multispectral images by analyzing multispectral images using a pre-trained deep learning model. Additionally, the monitoring module (124) can monitor the detected disturbing plants.

[0048] Hereinafter, with reference to FIGS. 4 to 11, a method for monitoring ecosystem-disturbing plants according to one embodiment of the present application will be described. The method for monitoring ecosystem-disturbing plants according to one embodiment of the present application can be performed by an ecosystem-disturbing plant monitoring device (100) shown in FIG. 2.

[0049] First, FIG. 4 is a flowchart illustrating a method for monitoring ecosystem-disturbing plants according to one embodiment of the present application. Referring to FIG. 4, in step S110, an ecosystem-disturbing plant monitoring device according to one embodiment of the present application can acquire a multispectral image of an area where the distribution of disturbing plants is expected using a drone equipped with a multispectral sensor.

[0050] An ecosystem-disturbing plant monitoring device according to one embodiment of the present application can map location data of an area expected to have a disturbing plant distribution with a multispectral image. For example, the location data may include latitude coordinates and longitude coordinates, etc.

[0051] An ecosystem disturbing plant monitoring device according to one embodiment of the present application can acquire RGB image bands using an RGB sensor equipped on a drone for the same regional range of the area where the disturbing plant is expected to be distributed. Additionally, the ecosystem disturbing plant monitoring device can acquire multispectral image bands using a multispectral sensor for the same regional range of the area where the disturbing plant is expected to be distributed. Furthermore, the ecosystem disturbing plant monitoring device can acquire vegetation index bands calculated based on the multispectral image bands for the same regional range of the area where the disturbing plant is expected to be distributed.

[0052] An ecosystem disturbing plant monitoring device according to one embodiment of the present application can generate an image stack in which images of an RGB image band, a multispectral image band, and a vegetation index band are stacked for the same regional range of an area where the distribution of disturbing plants is expected, based on an RGB image band, a multispectral image band, and a vegetation index band. For example, the image stack may represent an image in which images of an RGB image band, a multispectral image band, and a vegetation index band are superimposed.

[0053] In step S120, the ecosystem disturbing plant monitoring device according to one embodiment of the present application can convert the resolution of a multispectral image. For example, the ecosystem disturbing plant monitoring device according to one embodiment of the present application can convert the resolution of the multispectral image to four times higher using a Super-Resolution Convolutional Neural Network (SRCNN) algorithm.

[0054] In step S130, an ecosystem disturbing plant monitoring device according to one embodiment of the present application can detect disturbing plants from multispectral images by analyzing multispectral images using a pre-trained deep learning model. For example, an ecosystem disturbing plant monitoring device according to one embodiment of the present application can detect the location and status of disturbing plants from an image stack using a pre-trained U-Net model. For example, the U-Net model can be trained based on captured images of disturbing plants.

[0055] In step S140, an ecosystem disturbing plant monitoring device according to one embodiment of the present application can monitor detected disturbing plants. For example, an ecosystem disturbing plant monitoring device according to one embodiment of the present application can direct a disturbing plant removal operation based on the location and condition of the detected disturbing plants. Additionally, the ecosystem disturbing plant monitoring device can evaluate the disturbing plant removal operation using multispectral images before and after the disturbing plant removal operation. Furthermore, the ecosystem disturbing plant monitoring device can recognize the pattern of the detected disturbing plants by analyzing the trend of change over time of the detected disturbing plants.

[0056] FIGS. 5(a) and FIGS. 5(b) are drawings for illustrating an example of converting the resolution of a multispectral image according to one embodiment of the present application.

[0057] Referring to FIG. 5(a), an example of a multispectral image acquired from a drone is shown. Also, referring to FIG. 5(b), an example of a multispectral image acquired from a drone is shown, in which the resolution of the multispectral image is upscaled by 4x horizontally and 4x vertically using a Super-Resolution Convolutional Neural Network (SRCNN).

[0058] An ecosystem-disturbing plant monitoring device according to one embodiment of the present application can convert multispectral images acquired from a drone into high resolution using a Super-Resolution Convolutional Neural Network (SRCNN) algorithm. This method is effective for restoring image details and increasing pixel density. For example, SRCNN can take a low-resolution image as input and convert it into a high-resolution image through a multi-layer convolutional network.

[0059] FIGS. 6(a) to 6(e) are drawings for illustrating an example of a multispectral image according to an embodiment of the present application.

[0060] Referring to FIGS. 6(a) through 6(e), an example of a captured image from a drone is illustrated. The drone can capture five images from FIGS. 6(a) through 6(e) as a set for the same location. For example, FIG. 6(a) shows RGB image bands acquired using the drone's RGB sensor. For instance, three RGB image bands can be acquired using the drone's RGB sensor. Additionally, FIGS. 6(b) through 6(e) show four multispectral image bands acquired using the drone's multispectral sensor. Here, FIG. 6(b) may represent Green, FIG. 6(c) may represent Red, FIG. 6(d) may represent RedEdge, and FIG. 6(e) may represent NearInfrared.

[0061] An ecosystem disturbing plant monitoring device according to one embodiment of the present application can generate vegetation index bands based on multispectral imaging bands. For example, five vegetation index bands can be calculated based on multispectral imaging bands.

[0062] For example, an ecosystem-disturbing plant monitoring device according to one embodiment of the present application can acquire a total of 12 bands for the same location. For example, the RGB bands can represent three bands: R, G, and B bands (e.g., 4 / 3 CMOS, 20MP, FOV: 84). In addition, the multispectral imaging bands can represent four bands in the Green, Red, RedEdge, and NIR regions, with wavelengths of Green (G) being 560 ±16 nm, Red (R) being 650 ±16 nm, RedEdge (RE) being 730 ±16 nm, and Near-Infrared (NIR) being 860 ±26 nm. In addition, the vegetation index bands can represent vegetation indices NDVI (Normalized Difference Vegetation Index), GNDVI (Green Normalized Difference Vegetation Index), LCI (Leaf Chlorophyll Index), NDRE (Normalized Difference Red Edge), and OSAVI (Optimized Soil Adjusted Vegetation Index), which are generated by calculating the multispectral image bands acquired by the multispectral sensor.

[0063] FIG. 7 is a drawing for explaining an example of an image stack according to one embodiment of the present application.

[0064] Referring to FIG. 7, an example of an image stack in which 12 bands are superimposed at the same location is illustrated according to one embodiment of the present application.

[0065] An image stack according to one embodiment of the present application may have RGB image bands, multispectral image bands, and vegetation index bands superimposed.

[0066] FIGS. 8 and 9 are drawings for explaining an example of analyzing drone images using the U-net algorithm according to an embodiment of the present application.

[0067] Referring to Fig. 8, an example of a video captured by a drone is illustrated.

[0068] Referring to Fig. 9, an example of an image in which vegetation is classified by the U-net algorithm is shown in the image of Fig. 8.

[0069] An ecosystem-disturbing plant monitoring device according to one embodiment of the present application can automatically detect and classify the location and status of disturbing plants in an image by utilizing a pre-trained U-Net architecture. For example, the convolutional network of U-Net can enable accurate classification by extracting features at various scales. For instance, the U-Net algorithm according to the present invention can learn disturbing plants based on the image features and variance values ​​by extracting image features from three RGB bands and calculating the variance value of each pixel in the image from the remaining nine bands. Additionally, the U-Net algorithm according to the present invention can classify disturbing plants based on the image features and variance values.

[0070] An ecosystem-disturbing plant monitoring device according to one embodiment of the present application can issue work instructions using a Raster format tif file containing a work area so that it can be viewed in a program such as Qfield of QGIS, an open GIS program that can be activated on a mobile device.

[0071] An ecosystem-disturbing plant monitoring device according to one embodiment of the present application can learn and detect data on ecosystem-disturbing plants such as prickly pear, Japanese knotweed, Japanese ragweed, goldenrod, aster, fleabane, cordgrass, and goldenrod, and direct removal operations.

[0072] The ecosystem-disturbing plant monitoring device according to the present invention is capable of classifying more than 90% of species with a leaf length of about 15 cm or those that bloom at a specific time, and the deep learning model currently possessed has a classification accuracy of 91.5%.

[0073] FIGS. 10(a) and FIGS. 10(b) are drawings for illustrating an example of pre-training data according to one embodiment of the present application. Also, FIGS. 11(a) and FIGS. 11(b) are drawings for illustrating an example of an image chip of pre-training data according to one embodiment of the present application.

[0074] The ecosystem-disturbing plant monitoring device according to the present invention can be pre-trained based on an image chip as shown in FIG. 11(a) and FIG. 11(b) after (X, Y) labeling work as shown in FIG. 10(a) and FIG. 10(b).

[0075] For example, as illustrated in FIG. 10(a), the polygon (10) representing the labeling area may be composed of data in the form of a shape file, geographic coordinates, and an attribute table.

[0076] For example, as shown in FIG. 10(b), the attribute data may consist of an integer labeling number Classvalue, a scientific name of the species to be classified Classname, a latitude coordinate X column, and a longitude coordinate Y column.

[0077] A preliminary dataset for deep learning training can be selected within the polygon (10) area. For example, as shown in FIG. 11(a) and FIG. 11(b), the image chip can be set with a number of pixels such as 64, 64 or 128, 128, taking into account the size of the plant species.

[0078] The features, structures, effects, etc. described in the embodiments above are included in at least one embodiment of the present invention and are not necessarily limited to only one embodiment. Furthermore, the features, structures, effects, etc. exemplified in each embodiment may be combined or modified and implemented in other embodiments by a person skilled in the art to which the embodiments belong. Accordingly, details regarding such combinations and modifications should be interpreted as being included within the scope of the present invention.

[0079] Furthermore, although the embodiments have been described above, this is merely illustrative and does not limit the invention. Those skilled in the art will understand that various modifications and applications not exemplified above are possible within the scope of the essential characteristics of the embodiments. In other words, each component specifically shown in the embodiments may be modified and implemented. Differences related to such modifications and applications should be interpreted as being included within the scope of the invention as defined in the appended claims.

Claims

1. Regarding methods for monitoring ecosystem-disturbing plants, A step of acquiring multispectral images of an area expected to have a disturbed plant distribution using a drone equipped with a multispectral sensor; A step of converting the resolution of the above multispectral image; A step of detecting disturbing plants from the multispectral image by analyzing the multispectral image using a pre-trained deep learning model; and Step of monitoring detected disturbing plants; A method for monitoring ecosystem-disturbing plants, including 2. In Paragraph 1, A method for monitoring ecosystem-disturbing plants, wherein the step of acquiring the multispectral image includes the step of mapping the multispectral image to location data of the area where the disturbed plants are expected to be distributed.

3. In Paragraph 1, The step of acquiring the above multispectral image is, A step of acquiring RGB image bands using an RGB sensor equipped on the drone for the same regional range of the above-mentioned disturbed plant distribution area; A step of acquiring multispectral image bands using the multispectral sensor for the same regional range; and A method for monitoring ecosystem-disturbing plants, comprising the step of obtaining a vegetation index band calculated based on the above multispectral image band.

4. In Paragraph 3, A method for monitoring ecosystem-disturbing plants, comprising the step of acquiring the multispectral image, which includes generating an image stack in which images of the RGB image band, the multispectral image band, and the vegetation index band are stacked for the same regional range based on the RGB image band, the multispectral image band, and the vegetation index band.

5. In Paragraph 4, A method for monitoring ecosystem-disturbing plants, wherein the step of converting the resolution includes the step of converting the resolution of the multispectral image to a 4-fold higher resolution using an SRCNN (Super-Resolution Convolutional Neural Network) algorithm.

6. In Paragraph 5, A method for monitoring ecosystem-disturbing plants, wherein the step of detecting the above-mentioned disturbing plants includes the step of detecting the location and status of the disturbing plants from the image stack using a pre-trained U-Net model.

7. In Paragraph 6, The step of monitoring the above-mentioned disturbed plants is, A step of directing the removal of disturbing plants based on the location and condition of the detected disturbing plants; A step of evaluating a disturbing plant removal operation using the above multispectral images before and after the disturbing plant removal operation; and A method for monitoring ecosystem-disturbing plants, comprising the step of recognizing patterns of detected disturbing plants by analyzing the trend of change over time of the detected disturbing plants.

8. A computer-readable recording medium storing a program for executing a method according to any one of claims 1 through 7.

9. In a device for monitoring ecosystem-disturbing plants, A processor for monitoring detected disturbing plants based on multispectral images acquired from a drone equipped with a multispectral sensor; comprising The above processor is, An image acquisition module that acquires multispectral images of an area expected to have disturbed plant distribution using the above drone; A super-resolution module for converting the resolution of the above multispectral image; A detection module that detects disturbing plants from the multispectral image by analyzing the multispectral image using a pre-trained deep learning model; and Monitoring module for monitoring detected disturbing plants; An ecosystem-disturbing plant monitoring device including 10. In a system for monitoring ecosystem-disturbing plants, A drone equipped with a multispectral sensor to acquire multispectral images of an area expected to have a disturbed plant distribution; and A processor for monitoring detected disturbing plants based on multispectral images acquired from the above drone; comprising The above processor is, An image acquisition module that acquires multispectral images of an area expected to have disturbed plant distribution using the above drone; A super-resolution module for converting the resolution of the above multispectral image; A detection module that detects disturbing plants from the multispectral image by analyzing the multispectral image using a pre-trained deep learning model; and Monitoring module for monitoring detected disturbing plants; An ecosystem-disturbing plant monitoring system including

Images