KR102276549B1 - Method and system for identification and capturing harmful animals through ai-based monitoring camera - Google Patents

Abstract

The present disclosure relates to a method and a system for identifying and capturing harmful birds using an artificial intelligence-based detection camera. The system for identifying and capturing harmful birds using an artificial intelligence-based detection camera includes: a capturing frame including a first gate and a second gate that can be opened and closed; a first camera that detects the appearance of harmful birds within a scope of a first region of interest set in the first gate and detects the number of individuals; and a second camera that detects the appearance of harmful birds within a scope of a second region of interest set in the second gate and detects the number of individuals.

Description

Method and system for identifying and capturing harmful tides through an AI-based selective object detection camera {METHOD AND SYSTEM FOR IDENTIFICATION AND CAPTURING HARMFUL ANIMALS THROUGH AI-BASED MONITORING CAMERA}

The present disclosure relates to a method and system for identifying and capturing harmful tides through an AI-based selective object detection camera, and specifically, to a region of interest (ROI) through a camera with an AI-based harmful tide detection function. The present invention relates to a method and system for detecting the appearance and number of harmful birds within a range, and for identifying and capturing harmful birds based on the detected number of entering and exiting traps.

In general, harmful tidal waves refer to animals or birds that damage people's living space and damage crops, equipment, or the like, or threaten people to inflict harm, such as an accident risk. Recently, as the wild habitats of harmful tidal currents are gradually decreasing or domestic canines have been abandoned, they have adapted to the wild and intrusion of harmful tidal currents into urban dwellings, etc. is increasing. Therefore, the development of technology for catching harmful algae is required.

If harmful tides carry bacteria such as viruses that are at risk of infection, they can cause great harm to the hygiene of an unspecified number of people, such as humans and livestock, so the importance of quarantine activities and quarantine is increasing. For example, African swine fever virus is a type 1 livestock contagious disease that occurs in all age groups of wild boars and domestic pigs. Pigs infected with African swine fever virus show 100% mortality within 2 to 10 days. After the first case of African swine fever virus was announced in Kenya in 1921, the first infection started in Europe in 1957, and then it occurred in China and Southeast Asia. Recently in Korea, efforts are being made to prevent the movement and quarantine of wild boars, which are vectors, in order to block the transmission. The main cause of the spread of the African swine fever virus is through contact with infected individuals and spread through distribution and leftover feeding of infected individuals.

To solve this problem, it is necessary to capture and manage harmful algae that can become vectors of infectious diseases. The device for capturing such harmful algae basically includes a trapping frame for trapping the harmful algae, and a door for trapping the harmful algae attracted to the trapping frame through the entrance of the trapping frame to prevent escape. For example, looking at the conventional Korean Patent Registration No. 10-1625188, a technique for catching harmful tides using a wire mesh frame that connects a plurality of frames in a grid shape while covering each frame with a wire mesh is disclosed. . According to this prior art, wild animals can be continuously captured by using a door in which the end of one side of the wire mesh frame is connected by a hinge so that it is opened while rotating only inward.

However, in the prior art, the door for trapping noxious birds in the trapping device is usually manufactured in a size suitable for the adult, so the pups in the tangible age cannot be pushed in by the weight of the door made to fit the adult. There was a problem that it was difficult to capture. In addition, in the prior art, in order to reduce the vigilance of harmful tidal birds, a catching device is installed for a long period of time, and a person periodically inspects it directly, and a person directly checks whether or not the toxic tidal is captured. , it was difficult to confirm the capture, so there was a problem that the convenience was significantly reduced.

In order to solve the above problems, the present disclosure efficiently manages identification and capture of harmful birds through an artificial intelligence-based selective object detection camera, and simultaneously captures according to the habit of harmful birds moving in groups with adults and offspring. It provides a method and system for identifying and capturing harmful birds through an artificial intelligence-based detection camera.

According to the present disclosure, there is provided a method and system for detecting the appearance and number of harmful birds within a region of interest (ROI) range through an artificial intelligence camera, and detecting the number of entering and exiting a trap to identify and capture harmful birds. .

In addition, according to the present disclosure, a harmful tidal classification model configured to detect and classify a type of harmful tidal to be captured through deep learning learning may be mounted on an edge camera installed in a capture frame. Accordingly, at the capture frame site, the camera itself determines the appearance, number, and presence of harmful tidal objects, and transmits the determination result in the form of an alarm or push to the administrator's terminal device through the network for more efficient capture and capture. The maintenance of the frame is possible.

In addition, according to the present disclosure, an artificial intelligence-based harmful tide classification model trained to classify or discriminate each of a plurality of types of harmful birds having different shapes and operation patterns, such as wild boar, elk, wild dog, and nutria, is appropriately applied as needed. By mounting it on an AI camera, it may be possible to efficiently capture certain types of harmful algae. In addition, according to the present disclosure, the implementation and installation of a trapping frame and a trapping method are different depending on the type of harmful algae, so that it is difficult to apply an appropriate trapping method when a new trapping frame is installed.

The system for identifying and capturing harmful birds through an artificial intelligence-based selective object detection camera according to an embodiment of the present disclosure includes a capture frame including an openable and openable first gate and a second gate, and a first set adjacent to the first gate. and a first camera for detecting the appearance and number of harmful birds within the region of interest, and a second camera for detecting the appearance and number of harmful birds within a second region of interest set adjacent to the second gate.

According to another embodiment of the present disclosure, there is provided a method for identifying and capturing harmful birds using an artificial intelligence-based selective object detection camera, in a capture frame including a first gate and a second gate that can be opened and closed by a first camera, Detecting the appearance of an adult of noxious birds within a range of a first region of interest set adjacent to the first gate, detecting, by a first camera, the number of adults of noxious birds that have passed through the first gate; Controlling, by the first camera, to close the first gate, when the number of adults of the passing noxious bird exceeds the first predetermined number; by the second camera, a second set adjacent to the second gate Detecting the appearance of offspring of harmful birds within the region of interest; detecting, by the second camera, the number of offspring of harmful birds that have passed through the second gate; and the number of offspring of harmful birds that have passed through the second gate and controlling, by the second camera, to close the second gate when the number of ? exceeds a first predetermined number of individuals.

According to some embodiments of the present disclosure, since the artificial intelligence-based harmful tide detection model learned for each harmful tide to be captured is installed in the camera installed in the capture frame, it is possible to efficiently control and utilize the capture frame according to the capture target or purpose Do.

According to some embodiments of the present disclosure, it is possible to check the status of the capture frame in real time using a user terminal or a smart phone that operates remotely through wireless communication with a surveillance camera installed on the capture frame, thereby simplifying the maintenance of the capture device. Convenience is greatly improved.

According to some embodiments of the present disclosure, as a plurality of gates are provided for each size of noxious tide in the capture frame, noxious birds of various sizes can be captured, and the range of captureable entities can be expanded in consideration of the habits of the target. can have an effect.

According to some embodiments of the present disclosure, by arranging an attractant such as excreta of harmful algae in the trapping frame, there is an effect of improving the capture probability by easily guiding harmful tides into the trapping frame.

According to some embodiments of the present disclosure, since the opening and closing of the gate is controlled according to the number of harmful birds entering and exiting the trapping frame, the effect of simultaneously capturing a plurality of harmful birds in consideration of the swarm behavioral characteristics of the capture target is obtained. have

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present disclosure will be described with reference to the accompanying drawings described below, wherein like reference numerals denote like elements, but are not limited thereto.
1 is an exemplary diagram illustrating the configuration of a system for identifying and capturing harmful birds using an artificial intelligence-based detection camera according to an embodiment of the present disclosure.
2 is a configuration in which the harmful bird identification and capture system is connected to communicate with a plurality of user terminals in order to share information of the harmful bird identification and capture system through an artificial intelligence-based detection camera according to an embodiment of the present disclosure; It is a schematic diagram showing
3 is a block diagram illustrating an internal configuration of an AI-based detection camera according to an embodiment of the present disclosure.
4 is a diagram illustrating a configuration of a harmful tide identification model including an artificial neural network trained to classify a harmful tide image or image into a harmful tide type according to an embodiment of the present disclosure.
5 is a flowchart illustrating a method for identifying and capturing harmful birds using an artificial intelligence-based detection camera according to an embodiment of the present disclosure.
6 is a diagram illustrating an example of a user interface of a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure.
7 is a view showing the appearance of identified harmful birds designated in a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure.
FIG. 8 is a view showing entry and exit of a trapping frame of identified harmful birds to a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure.
FIG. 9 is a view showing identification of harmful birds and capturing of a designated number of individuals in a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure.

Hereinafter, specific contents for carrying out the present disclosure will be described in detail with reference to the accompanying drawings. However, in the following description, if there is a risk of unnecessarily obscuring the subject matter of the present disclosure, detailed descriptions of well-known functions or configurations will be omitted.

In the accompanying drawings, identical or corresponding components are assigned the same reference numerals. In addition, in the description of the embodiments below, overlapping description of the same or corresponding components may be omitted. However, even if descriptions regarding components are omitted, it is not intended that such components are not included in any embodiment.

Advantages and features of the disclosed embodiments, and methods of achieving them, will become apparent with reference to the embodiments described below in conjunction with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the present disclosure to be complete, and those of ordinary skill in the art to which the present disclosure pertains. It is only provided to fully inform the person of the scope of the invention.

Terms used in this specification will be briefly described, and the disclosed embodiments will be described in detail. The terms used in this specification are selected as currently widely used general terms as possible while considering the functions in the present disclosure, but may vary depending on the intention or precedent of a person skilled in the art, the emergence of new technology, and the like. In addition, in a specific case, there is a term arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the corresponding invention. Therefore, the terms used in the present disclosure should be defined based on the meaning of the term and the contents of the present disclosure, rather than the simple name of the term.

References in the singular herein include plural expressions unless the context clearly dictates that the singular is singular. Also, the plural expression includes the singular expression unless the context clearly dictates the plural. When a part "includes" a certain element throughout the specification, this means that other elements may be further included, rather than excluding other elements, unless otherwise stated.

In addition, the term 'module' or 'unit' used in the specification means a software or hardware component, and 'module' or 'unit' performs certain roles. However, 'module' or 'unit' is not meant to be limited to software or hardware. A 'module' or 'unit' may be configured to reside on an addressable storage medium or may be configured to refresh one or more processors. Thus, as an example, 'module' or 'unit' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, may include at least one of procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, database, data structures, tables, arrays or variables. Components and 'modules' or 'units' are the functions provided therein that are combined into a smaller number of components and 'modules' or 'units' or additional components and 'modules' or 'units' can be further separated.

According to an embodiment of the present disclosure, a 'module' or a 'unit' may be implemented with a processor and a memory. 'Processor' should be construed broadly to include general purpose processors, central processing units (CPUs), microprocessors, digital signal processors (DSPs), controllers, microcontrollers, state machines, and the like. In some circumstances, a 'processor' may refer to an application specific semiconductor (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), or the like. 'Processor' refers to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of a plurality of microprocessors, a combination of one or more microprocessors in combination with a DSP core, or any other such configurations. You may. Also, 'memory' should be construed broadly to include any electronic component capable of storing electronic information. 'Memory' means random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erase-programmable read-only memory (EPROM); may refer to various types of processor-readable media, such as electrically erasable PROM (EEPROM), flash memory, magnetic or optical data storage, registers, and the like. A memory is said to be in electronic communication with the processor if the processor is capable of reading information from and/or writing information to the memory. A memory integrated in the processor is in electronic communication with the processor.

1 is an exemplary diagram illustrating the configuration of a system for identifying and capturing harmful birds using an artificial intelligence-based detection camera according to an embodiment of the present disclosure.

The system 100 for identifying and capturing harmful birds using an artificial intelligence-based detection camera according to an embodiment of the present disclosure includes a capture frame 110 including a first gate 160 and a second gate 170 that can be opened and closed. , a first camera 120 that detects the appearance of harmful tides within the range of the first region of interest 180 set in the first gate 160 , and detects the number of animals, and a second region of interest set in the second gate 170 . (190) and a second camera 140 that detects the appearance of harmful tides within the range and detects the number of individuals.

According to an embodiment, the first and second cameras 120 and 140 store and execute an artificial intelligence-based harmful tide detection model (or harmful tide identification model) that has been learned to detect each type of harmful tide to be captured. It can be configured to be possible. For example, the first and second cameras 120 and 140 may store and execute an artificial neural network-based harmful bird detection model trained to detect only a specific harmful bird (eg, wild boar) as a capture target. As described above, each of the first and second cameras 120 and 140 is implemented to embed and execute a pre-trained harmful tidal detection model, so that the cameras 120 and 140 are installed in the capture frame and efficiently in real time during operation. can detect harmful tides to be captured. When one or more harmful tides are set as capture targets, the first and second cameras 120 and 140 may embed and execute one or more harmful tide detection models learned to detect each harmful tide.

In an embodiment, the first and second cameras 120 and 140 may include lenses configured to photograph the first ROI 180 and the second ROI 190 , respectively. In addition, each of the first and second cameras 120 and 140 inputs a photographed photo image or video to a harmful tide detection model built in each frame to detect the type of harmful tide included in the input image or video. can The harmful tide detection results by the first and second cameras 120 and 140 are transmitted to a remote user terminal through the communication unit 130, so that the user can be provided with various information and services based on the harmful tide detection results. .

More specifically, the first camera 120 detects the appearance of the harmful tide in the first region of interest 180 including the first gate 160 or the first region of interest using the built-in harmful tide detection model. (180) It is possible to determine the number of harmful tides contained within. In addition, the second camera 140 detects the appearance of harmful tides in the second region of interest 190 including the second gate 170 or the second region of interest 190 using the built-in harmful tide detection model. It is possible to determine the number of harmful algae contained within.

In one embodiment, the first gate 160 may have a height and a width that allow adults to enter and exit from among the harmful birds to be captured. In this case, the first camera 120 may have a built-in harmful tide detection model that has been trained to detect an adult noxious tide. For example, the first camera 120 may be configured to recognize an adult larger than a predetermined size among harmful birds to be captured. In addition, the second gate 170 may have a height and a width that allow a baby to enter and exit from among the harmful birds to be captured. That is, the size of the second gate 170 may be configured to be smaller than the size of the first gate 160 . In this case, the second camera 140 may have a built-in harmful bird detection model learned to detect harmful birds that are babies. For example, the second camera 140 may be configured to recognize a pup smaller than a predetermined size among harmful birds to be captured.

In an embodiment, the first camera 120 is configured to determine whether the number of objects that have passed through the first gate 160 among the harmful birds detected in the first region of interest 180 exceeds a predetermined number of the first gate. It may be configured to control the opening and closing of 160 . In addition, the second camera 140 detects the second gate 170 according to whether the number of harmful birds detected in the second region of interest 190 that has passed through the second gate 170 exceeds a predetermined number. It may be configured to control the opening and closing of the.

In this way, by configuring the first and second gates 160 and 170 and the first and second cameras 120 and 140, the adult among the harmful birds to be captured is first guided into the capture frame through the first gate and captured. Thereafter, the pups of the harmful bird can be captured by guiding them into the capture frame through the second gate. Such a configuration of the system 100, in general, harmful birds such as wild boars move or act in groups, and utilize the characteristic that young children are likely to exist in the vicinity of adults.

As shown, the first gate 160 may be formed on one side of the capture frame 110 , and the second gate 170 may be formed on the other side of the capture frame 110 . However, the configuration of the capture frame 110 of the present disclosure is not limited thereto, and the first and second gates 160 and 170 may be formed together on one side of the capture frame 110 .

In one embodiment, the capture frame 110 may be configured by coupling a plurality of capture frame parts of a predetermined unit size to each other. In this way, by manufacturing the capture frame 110 with small units of the capture frame parts, it can be easily transported and installed to the site where the harmful algae to be captured appear. In addition, in the capture frame 110, an inducing material (eg, secretions or food of a female noxious bird) that can induce access of harmful birds may be pre-arranged.

2 is a configuration in which the harmful bird identification and capture system is connected to communicate with a plurality of user terminals in order to share information of the harmful bird identification and capture system through an artificial intelligence-based detection camera according to an embodiment of the present disclosure; It is a schematic diagram showing

The system 100 for identifying and capturing harmful tides through an AI-based detection camera according to an embodiment of the present disclosure detects harmful tides through the AI-based detection cameras 120 and 140, and the communication network 150 ) may be configured to communicate with an external device (eg, an external user terminal device). The system 100 for identifying and capturing harmful birds through an artificial intelligence-based detection camera communicates with an external device through the communication network 150 to transmit information detected within the area of interest around the capture frame or data on control of the capture frame. can receive Here, the information detected within the region of interest around the capture frame and the data on the capture frame control are the detection result of the type of harmful tide, the human detection result, whether the harmful tide enters the capture frame, and the harmful tide passing through the gate of the capture frame. It may include, but is not limited to, data on the number of individuals and the opening/closing state of the capture frame gate.

According to an embodiment, at least one of the first camera 120 and the second camera 140 may receive information about at least one of the type and number of harmful birds to be captured from the external user terminal device. In addition, at least one of the first camera 120 and the second camera 140 may determine whether or not the appearance of the harmful tide in the first ROI or the second ROI is determined based on the received type and number of the harmful tidal current. It may be configured to detect a population. In addition, at least one of the first camera 120 and the second camera 140 transmits, to the external user terminal device, information on the appearance or number of harmful birds detected in the first region of interest or the second region of interest. can be configured to

According to an embodiment, the harmful tide identification and capture system 100 may periodically or aperiodically transmit/receive harmful tide information and capture frame control information to and from an external device through the network 150 . In FIG. 2 , the AI-based detection cameras 120 and 140 are illustrated as an example of communicating with an external device through the network 150 , but the present invention is not limited thereto. The detection cameras 120 and 140 may include any computing device capable of wired and/or wireless communication with an external user terminal device and capable of receiving data about the capture frame. That is, in the field where the capture frame is installed, the first camera 120 or the second camera 140 determines the appearance and number of harmful algae (eg, the total number of harmful algae in the first or second area of interest around the capture frame). 5) is detected and transmitted to an external user terminal device (eg, a smartphone). In the external user terminal device that has received such information, for example, a user who has confirmed that the number of harmful birds (ie, 5) transmitted by the first camera 120 or the second camera 140 is the target of the capture. The number of harmful birds (ie, 5) is input, and the user terminal transmits the corresponding information to the first camera 120 or the second camera 140 . Accordingly, the first camera 120 or the second camera 140 may set a predetermined number of adults or pups (eg, four) based on the received number of harmful birds. In addition, the first camera 120 or the second camera 140 detects the number of noxious birds within the range of the first region of interest or the second region of interest, and determines the number of individuals passing through the first gate or the second gate. In case of exceeding the population (that is, when the number of individuals passing through the gate exceeds the predetermined number of 4 and becomes 5), it is confirmed that all harmful algae existing near the trap enter the trap, and It can be controlled to close the first gate or the second gate.

The network 150 may be configured to enable communication between the AI-based detection cameras 120 and 140 and an external user terminal device. Network 150 according to the installation environment, for example, Ethernet (Ethernet), wired home network (Power Line Communication), telephone line communication device and wired networks such as RS-serial communication, mobile communication network, WLAN (Wireless LAN), It may consist of wireless networks such as Wi-Fi, Bluetooth and ZigBee, or a combination thereof.

According to an embodiment, the system 100 for identifying and capturing harmful birds provides computer-executable programs and data related to the control of the trapping frame and the provision of information on the state of capturing harmful birds based on the type, number, and entry of harmful birds. It may include one or more server devices and/or databases capable of storing, providing and executing, or one or more distributed computing devices and/or distributed databases based on cloud computing services. For example, the harmful tide identification and capture system 100 or AI-based detection cameras 120 and 140 are computing devices that can communicate with other devices through a wired or wireless network, and include a central processing unit (CPU). , a computing device capable of performing an arithmetic operation using a processing device such as a graphic processing unit (GPU), a digital signal processor (DSP), etc., but is not limited thereto.

The artificial intelligence-based detection cameras 120 and 140, based on the image or image taken of the capture frame installed in the system 100 for identifying and capturing harmful algae, or the region of interest around it, detects the appearance of harmful tides and the detection of harmful tides. It may be configured to embed and execute one or more artificial intelligence-based harmful tide detection models that have been trained to detect the number of individuals and whether or not the harmful tide enters the trapping frame. The AI-based detection cameras 120 and 140 receive information on the type of the target harmful tide received from the user terminals 210 to 230, and based on the received type of harmful tide, a plurality of harmful tide detection models By selecting and executing any one of them, it is possible to more accurately determine the appearance of the harmful tide set by the user as a capture target. The configuration and process of detecting harmful tidal water and controlling the trapping frame through communication with an external user terminal device will be described in detail below with reference to FIG. 3 .

3 is a block diagram illustrating an internal configuration of an artificial intelligence-based detection camera according to an embodiment of the present disclosure.

Hereinafter, the artificial intelligence-based detection camera 300 stores and executes the learned harmful tide identification model to be able to identify harmful tides, so that the region of interest set near the capture frame or the number or types of harmful tides existing in the capture frame. The configuration and process for determining the will be described in detail. In one embodiment, as shown in FIG. 3 , the AI-based detection camera 300 includes a communication unit 380 , a power supply unit 390 , an image sensor 320 , a processor 340 and a storage unit 360 . can be configured to include.

The communication unit 380 may be configured to transmit the harmful tide determination result to an external user terminal device through the network 150 or to receive data on whether to control the harmful tide and capture frame from the user terminal device. According to an embodiment, the communication unit 380 may receive preset data on whether to control the harmful tide and the trapping frame from the external user terminal device. Here, the data on whether the harmful tide and the capture frame are controlled include data related to the type of harmful tide to be determined by the camera 300, whether a person is detected, whether or not the harmful tide enters the capture frame, and the opening and closing state of the capture frame gate can, but is not limited thereto. For example, the data on whether harmful tides and the capture frame are controlled may further include various information related to the capture frame and the range of the region of interest, such as a warning notification when a person enters the capture frame. In addition, the communication unit 380 may provide the received data to the storage unit 360 , and transmit information on harmful tides determined by the processor 340 or information analyzed or processed such information to the network 150 . It may be configured to transmit to an external user terminal device through the

The power supply unit 390 converts solar thermal energy, which is natural light, into electrical energy to charge the battery of the camera 300, and boosts the voltage (Vcc) output from the power supply unit 390 to the image sensor 320 of the camera 300, It is supplied to the processor 340 , the storage unit 360 , and the communication unit 380 .

The power supply unit 390 is composed of a plurality of solar cells SOLAR that converts solar thermal energy into electrical energy, and the solar thermal energy converted into electrical energy by the solar cells SOLAR is charged to the camera 300 battery CHARGE. do. As described above, the electric energy charged in the battery CHARGE is used as driving power for the image sensor 320 , the processor 340 , the storage unit 360 , and the communication unit 380 .

The processor 340 includes at least one of a central processing unit (CPU), a graphic processing unit (GPU), a digital signal processor (DSP), a field programmable gate array (FPGA), and an application specific integrated circuit (ASIC). It may be configured to perform an operation and manage, process, and/or store data received from an external device and/or data obtained by analyzing or processing the received data in the storage unit 360 .

The processor 340 may store data on the harmful tide determination model received from the external device in the storage unit 360 . One or a plurality of harmful tide identification models stored in the storage unit 360 may be stored according to the type of harmful tide determined by the camera 300 . For example, when the camera 300 is set to determine one type of harmful tide (eg, wild boar), the storage unit 360 may store the harmful tide identification model learned to identify the wild boar. In another example, when the camera is set to discriminate a plurality of types of harmful tides (eg, wild boar, elk, etc.), the storage unit 360 uses the harmful tide identification models learned to identify each of the plurality of types of harmful tides. You can also save them all.

The processor 340 determines the image or image captured by the image sensor 320 using the harmful tide determination model stored in the storage unit 380 , and transmits the determination result to an external user terminal device through the communication unit 380 . can be transmitted

The harmful tide determination model stored in the storage unit 360 and executed by the processor 340 may be previously learned through a machine learning algorithm. The machine learning algorithm used to learn the harmful tidal identification model is one of machine learning or artificial neural network-based image identification models such as artificial neural networks, deep learning, and decision trees. It may be one, but is not limited thereto.

4 is a diagram illustrating a configuration of a harmful tide identification model including an artificial neural network trained to classify a harmful tide image or image into a harmful tide type according to an embodiment of the present disclosure.

The harmful tide determination model 440 may be generated, stored, or updated by an external computing device, and may be executed by the processor 340 illustrated in FIG. 3 . As shown, the harmful tide identification model 430 may include an artificial neural network model 432 that classifies the input harmful tide images 410 and 420 into adults or offspring 440 of the corresponding harmful tide.

The artificial neural network model 432 receives the harmful tide images 410 and 420 or its visual features as input, and the association score for a plurality of types of harmful birds (eg, adults or offspring of specific harmful birds) 440 . (450) can be calculated. The artificial neural network model 432 uses the learning data extracted from the harmful tide image and image database (not shown), and the association score 450 of the plurality of harmful tide types 440 for the harmful tide images 410 and 420. can be learned to yield . The learning data extracted from the harmful tide image and image database may include, for example, pairs of a plurality of harmful tide types associated with each of the plurality of harmful tide images.

The artificial neural network model 432 may output information on one or more harmful tide types associated with the harmful tide images 410 and 420 based on the calculated association scores 450 . In an embodiment, the artificial neural network model 432 may select and output information on the types of one or more harmful tides exceeding a predetermined threshold (eg, 0.5) from the harmful tide types 440 .

5 is a flowchart illustrating a method for identifying and capturing harmful birds using an artificial intelligence-based detection camera according to an embodiment of the present disclosure.

The method for identifying and capturing harmful birds using an artificial intelligence-based detection camera according to an embodiment of the present disclosure may begin with the step ( S510 ) of detecting the appearance of a specific type of harmful birds. According to an embodiment, in a capture frame including a first gate and a second gate that can be opened and closed by the first camera, the appearance of an adult of harmful birds within a range of a first region of interest set adjacent to the first gate is detected. can do. In addition, the second camera may detect the appearance of offspring of harmful birds within the range of the second region of interest set adjacent to the second gate.

Next, a step ( S520 ) of determining a designated identification target within a preset region of interest (ROI) range set by the user may be performed. When it is determined as the designated identification target, the steps of determining the adult or offspring of the identification target (S532, S534), determining the adult or offspring, and determining the number of individuals according to the determined individual (S542, S544), inside the capture frame In step S560 of determining the number of individuals designated by the user, and when the number of individuals designated inside the capture frame is confirmed, the step of capturing by closing the gate of the capture frame (S570) may be executed.

As described above, by actively utilizing the user-specified information on harmful tidal currents and learning through deep learning to process the trapping frame control, the user can efficiently capture and manage a specific type of harmful tidal target to be captured.

6 is a diagram illustrating an example of a user interface of a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure.

As shown, the user interface 600 of the user terminal device includes a user interface 620 for specifying the type of the harmful bird to be identified and a user interface 640 for specifying the number of the target harmful bird to be captured.

The user may select a specific type of harmful tide (eg, wild boar, elk, mole, etc.) through the user interface 620 . The information on the selected type of harmful tidal water may be transmitted from the user terminal device to at least one of the cameras installed in the capture frame. The camera, receiving the information on the type of harmful tide, uses an artificial intelligence-based harmful tide identification model that has been trained to detect or identify the harmful tide, and the occurrence and capture of the harmful tide within the designated area of interest around the capture frame. It is possible to detect and determine the number of harmful tides that have passed through the gate installed in the frame.

Also, the user may select the number of seaweeds to be captured through the user interface 640 . The information on the number of the selected noxious birds may be transmitted from the user terminal device to at least one of the cameras installed in the capture frame. The camera, which received the information on the number of harmful tides, appeared in the designated area of interest around the capture frame and then installed in the capture frame using an artificial intelligence-based harmful tide identification model learned to detect or identify the harmful tide. It is possible to detect and determine the number of harmful tides that have passed through. Based on the number of detected harmful tides, the camera may control opening and closing of the corresponding gate.

As described above, the user can effectively capture a desired type of harmful tide by selectively setting the type and number of harmful birds to be captured through the user terminal device.

7 is a view showing the appearance of identified harmful birds designated in a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure. FIG. 8 is a view showing entry and exit of a trapping frame of identified harmful birds to a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure. 9 is a diagram illustrating identification and capture of a number of individuals designated in a user terminal device for controlling identification and capture of harmful birds through an artificial intelligence-based detection camera according to an embodiment of the present disclosure.

As shown in FIGS. 7 to 9 , the operation of the user interface of the user terminal device for controlling the identification and capture of harmful birds through the AI-based detection camera according to an embodiment of the present disclosure is an identification target designated by the user. Detects the appearance of harmful birds within the ROI range of the harmful bird (700), checks the entry and exits from the harmful bird capture frame of the identification target (800), and when a preset number of individuals is captured inside the capture frame, close the capture frame gate to complete the capture (900) at the user's discretion It can be checked through a terminal device and controlled through a user interface.

Although the present disclosure has been described in connection with some embodiments herein, various modifications and changes may be made without departing from the scope of the present disclosure that can be understood by those skilled in the art to which the present disclosure pertains. Further, such modifications and variations are intended to fall within the scope of the claims appended hereto.

100: Harmful tide identification and capture system through an artificial intelligence-based detection camera
110: capture frame
120: first camera
130: communication department
140: second camera
150: network
160: first gate
170: second gate
180: first ROI
190: second ROI
210: Smartphone
220: tablet
230: PC
300: AI-based detection camera
320: image sensor
340: processor
360: storage
380: communication department
390: power supply
410, 420: Harmful tide image
430: harmful tide identification model
432: artificial neural network model
440: noxious algae adult or offspring classification model
450: calculated score
500: Harmful tide identification and capture method through AI-based detection camera
600 : user interface
620: Items designated as hazardous tides for identification
640: Items for specifying the number of harmful birds to be captured

Claims (10)

In the system for identifying and capturing harmful tides through an artificial intelligence-based selective object detection camera,
a trapping frame including an openable and openable first gate and a second gate;
When detecting the appearance and number of adults of noxious birds within a range of a first region of interest set adjacent to the first gate, and the number of adults of noxious birds passing through the first gate is less than or equal to a predetermined number of adults, a first camera that maintains the first gate in an open state and controls to close the first gate when the number of individuals passing through the first gate exceeds the predetermined number of adults; and
If the number of offspring of harmful birds is detected within a range of a second region of interest set adjacent to the second gate, and the number of offspring of the harmful birds that have passed through the second gate is less than or equal to a predetermined number of offspring, a second camera that maintains the second gate in an open state, and controls to close the second gate when the number of individuals passing through the second gate exceeds the predetermined number of offspring;
At least one of the first camera and the second camera is configured to transmit, to an external user terminal device, information on the appearance and number of noxious birds detected in the first region of interest or the second region of interest,
At least one of the first camera and the second camera receives information on the type and number of the noxious bird to be captured from the external user terminal device, and determines the type of the received noxious bird as an adult of the noxious bird or the Set as the type of the offspring of harmful birds, and set the predetermined number of adults or the predetermined number of offspring based on the received number of harmful birds,
At least one of the first camera and the second camera is configured to detect the appearance or number of noxious birds in the first region of interest or the second region of interest, based on the received type and number of harmful birds. , Hazardous Tide Identification and Capture Systems.
According to claim 1,
and the first gate is configured to have a larger area than the second gate.
According to claim 1,
at least one of the first camera and the second camera is configured to store and execute an artificial intelligence-based harmful tide identification model trained to detect the harmful tide.
delete delete delete delete delete delete In a method for identifying and capturing harmful tides through an artificial intelligence-based selective object detection camera,
A trapping frame including a first gate and a second gate that can be opened and closed by a first camera, the capture frame comprising: detecting the appearance of an adult noxious algae within a range of a first region of interest set adjacent to the first gate;
transmitting, by the first camera, to an external user terminal device, information on the appearance and number of adults of the harmful bird detected in the first region of interest;
By the first camera, information on the type and number of noxious birds to be captured is received from the external user terminal device, the received type of noxious tide is set as the type of the adult of the noxious bird, and the received pre-determining the number of adults based on the number of harmful tides;
detecting, by the first camera, the appearance or number of harmful birds in the first region of interest, based on the received type and number of harmful birds;
detecting, by the first camera, the number of adults of the noxious birds that have passed through the first gate;
When the number of adults of the noxious bird that has passed through the first gate by the first camera is less than or equal to the predetermined number of adults, the first gate is controlled to an open state, and the predetermined number of adults is exceeded. if so, controlling the first gate to close;
detecting, by a second camera, the appearance of offspring of harmful birds within a range of a second region of interest set adjacent to the second gate;
transmitting, by the second camera, information on the appearance and number of offspring of the harmful bird detected within the second region of interest to the external user terminal device;
By the second camera, information on the type and number of the noxious bird to be captured is received from the external user terminal device, the received type of noxious bird is set as the type of the offspring of the noxious bird, and the received pre-determining the number of offspring based on the number of harmful tides;
detecting, by the second camera, the appearance or number of harmful birds in the second region of interest, based on the received type and number of harmful birds;
detecting, by the second camera, the number of offspring of the harmful bird passing through the second gate; and
By the second camera, when the number of offspring of the harmful bird that has passed through the second gate is less than or equal to a predetermined number of offspring, the second gate is controlled in an open state, and the predetermined number of offspring is exceeded If so, controlling to close the second gate
A method for identifying and capturing harmful algae, including.

Images