Wild animal tracking system, and control method thereof

Abstract

A wild animal tracking system and a control method thereof are provided. The control method of a wild animal tracking system according to the present invention comprises the steps of: receiving a captured image from each image capturing device and storing same; receiving a wild animal discovery report; extracting wild animal type information and wild animal discovery place information from the wild animal discovery report; selecting an analysis target captured image captured by an image capturing device located within a radius preset with reference to the extracted wild animal discovery place information, and stored; and determining a discovery location of a wild animal by using the selected analysis target captured image and the extracted wild animal type information.

Description

Wildlife tracking system and control method

[0001] The present invention relates to a wild animal tracking system and a control method thereof, and more specifically, to providing a system and a control method thereof for tracking the location of wild animals that appear in urban parks and pose a threat to citizens.

[0002] Wild animals such as water deer, deer, Asiatic black bears, and wild boars are generally not found in urban areas, but sometimes they appear in parks connected to mountains and cause harm to people.

[0003] Of course, these wild animals do not always attack humans and sometimes run away on their own, but there are instances where they directly attack or threaten people, and recently, there was even a case where citizens were hospitalized after being attacked by deer.

[0004] It is desirable to identify wild animals that have come down to urban areas as quickly as possible, safely capture them, and return them to the mountains; however, analyzing video footage recorded by hundreds or thousands of cameras every moment to locate wild animals is not advisable as it consumes a significant amount of resources.

[0005] Therefore, it is necessary to track wild animals as quickly as possible, at least from the moment a report of a sighting is received from citizens; however, no system for tracking wild animals in conjunction with citizens' reports of sightings has been proposed in the past.

[0006] (Prior Art) Korean Published Patent No. 10-2023-0105786

[0007] Accordingly, the present invention has been devised to satisfy the aforementioned conventional needs, and its purpose is to provide a system and a control method for rapidly identifying and tracking the location of wild animals at minimal cost in conjunction with reports of wild animal sightings by citizens.

[0008] To achieve the above-mentioned purpose, a control method for a wild animal tracking system according to the present invention may include: receiving and storing images captured from each imaging device; receiving a report of a wild animal sighting; extracting wild animal type information and wild animal sighting location information from the wild animal sighting report; selecting an image to be analyzed, captured and stored by an imaging device within a preset radius based on the extracted wild animal sighting location information; and determining the location of the wild animal sighting using the selected image to be analyzed and the extracted wild animal type information.

[0009] Here, the wild animal sighting report is a voice signal via ARS, and information on the type of wild animal and the location where the wild animal was found can be extracted by analyzing the ARS voice signal.

[0010] Here, the method further includes the step of constructing an artificial intelligence module by performing machine learning that takes a captured image and wild animal type information as input and outputs whether the wild animal type information exists in the captured image, and inputs the selected captured image to be analyzed and the extracted wild animal type information into the artificial intelligence module, and can determine the location where the wild animal was found using the output of the artificial intelligence module.

[0011] Here, the method may further include a step of determining the movement path of a wild animal using the multiple locations where the wild animal was found, and then controlling the wild animal to be tracked using a drone camera equipped with a camera function.

[0012] Here, when movement is detected by a motion detection sensor attached to the above-mentioned shooting device, the captured image can be received.

[0013] In addition, to achieve the above-mentioned purpose, the wild animal tracking system according to the present invention may include: a captured image storage unit that receives and stores captured images from each capturing device; a report receiving unit that receives and stores wild animal discovery reports; an information extraction unit that extracts wild animal type information and wild animal discovery location information from wild animal discovery reports stored by the report receiving unit; an analysis target selection unit that selects an analysis target captured image stored by a capturing device within a preset radius based on the wild animal discovery location information extracted by the information extraction unit; and a wild animal location determination unit that determines the location of wild animal discovery using the analysis target captured image selected by the analysis target selection unit and the wild animal type information extracted by the information extraction unit.

[0014] Here, the wild animal sighting report stored by the above-mentioned report receiving unit is a voice signal via ARS, and the above-mentioned information extraction unit can analyze the above-mentioned ARS voice signal to extract information on the type of wild animal and the location where the wild animal was found.

[0015] Herein, the system further includes an artificial intelligence module generation unit that constructs an artificial intelligence module by performing machine learning that takes a captured image and wild animal type information as inputs and outputs whether the wild animal type information exists in the captured image, and the wild animal location determination unit inputs the captured image of the target of analysis selected by the analysis target selection unit and the wild animal type information extracted by the information extraction unit into the artificial intelligence module, and can determine the location where the wild animal was found using the output of the artificial intelligence module.

[0016] Here, the drone control unit may further include a drone control unit that determines the movement path of a wild animal using a plurality of wild animal discovery locations determined by the wild animal location determination unit, and controls the wild animal to track using a drone shooting device equipped with a shooting function.

[0017] Here, the image storage unit can receive the captured image when movement is detected by a motion detection sensor attached to the shooting device.

[0018] FIG. 1 is a schematic diagram of an overall system including a wild animal tracking system according to one embodiment of the present invention, and

[0019] FIG. 2 is a functional block diagram of the wildlife tracking system of FIG. 1, and

[0020] FIG. 3 is an overall control flow diagram of a wild animal tracking system according to one embodiment of the present invention.

[0021] The present invention will be described in detail below with reference to the attached drawings.

[0022] The following embodiments according to the present invention are merely examples to aid in understanding the invention, and the invention is not limited to these embodiments. In particular, the present invention may be composed of a combination of at least one of the individual components, individual functions, or individual steps included in each embodiment.

[0023] In particular, for convenience, some claims in the scope of claims have included alphabets such as '(a)', but these alphabets do not define the order of each step.

[0024] The schematic configuration of the entire system including a wild animal tracking system (100) according to one embodiment of the present invention is as shown in FIG. 1.

[0025] In the drawing, the reporter (400) is a person who reports that they have found a wild animal, and may report by calling the wild animal tracking system (100) and making a voice report or by sending a text message.

[0026] For example, when reporting using an ARS (Automatic Response Service), the reporter (400) can state the type of wild animal they found and the location where they found it, etc., according to the instructions given by the ARS.

[0027] In this case, the reporter (400) may explain the characteristics of the wild animal if the type of wild animal is not clear.

[0028] Likewise, when sending a text message, it is preferable for the reporter (400) to include the type of wild animal and the location where it was found.

[0029] To this end, the wild animal tracking system (100) must be equipped with the function of providing an ARS service or receiving and processing text messages.

[0030] Furthermore, the wild animal tracking system (100) may operate a web page to allow a reporter (400) to access it using a web browser and then input the location where the wild animal was found in the correct format.

[0031] Meanwhile, the shooting device (200) of Fig. 1 is a device that transmits a captured image taken at a specific point in time, and may be, for example, a CCTV operating in various parts of the city.

[0032] In particular, the shooting device (200) includes a motion detection sensor, so that it can take a picture and transmit the captured image when motion is detected.

[0033] As another example, the recording device (200) may continuously record video as a video recording device (200) and transmit the recorded video images to the wild animal tracking system (100).

[0034] This shooting device (200) can be controlled by the wild animal tracking system (100).

[0035] The drone shooting device (300) is equipped with a shooting device (200) attached to a drone capable of flying in the sky, and is provided for tracking wild animals while shooting them in the air, and can be controlled by a wild animal tracking system (100) as described below.

[0036] Meanwhile, the wild animal tracking system (100) performs the function of receiving a wild animal report by a reporter (400) as described above, and the function of determining and tracking the location of wild animals by controlling a filming device (200) and a drone filming device (300).

[0037] To this end, the wild animal tracking system (100) may be configured to include a captured image storage unit (110), a report receiving unit (120), an information extraction unit (130), an analysis target selection unit (140), a wild animal location determination unit (150), an artificial intelligence module generation unit (160), and a drone control unit (170), as shown in FIG. 2.

[0038] The image storage unit (110) performs the function of receiving and storing images from each shooting device (200).

[0039] At this time, the image storage unit (110) stores and manages the image as is when the shooting device (200) transmits a still image, and if the shooting device (200) transmits a video image, it selects and stores it based on its own judgment.

[0040] Specifically, the captured image storage unit (110) can analyze the received video captured image and, when a change occurs (e.g., the appearance of a new subject), generate and store a set number of captured video images (e.g., 50) at a set time interval (e.g., 5 seconds interval).

[0041] That is, the captured image storage unit (110) generates and stores a still image from a video in which a change has occurred.

[0042] The report receiving unit (120) performs the function of receiving and storing reports of wild animal sightings.

[0043] As mentioned above, the report receiving unit (120) can store (record) the voice report content of the reporter (400) by providing an ARS service, or it can store the report text by receiving a text message from the reporter (400) by having a text message receiving function.

[0044] Furthermore, the report receiving department (120) may operate a web server to provide a web page to the reporter (400) and allow the reporter (400) to write and register the report content in accordance with the format.

[0045] The information extraction unit (130) performs the function of extracting wild animal type information and wild animal discovery location information from wild animal discovery reports stored by the report receiving unit (120).

[0046] For example, if a report of a wild animal sighting is made in voice, the information extraction unit (130) can convert the voice into text and then extract the type of wild animal (e.g., wild boar, deer, water deer, Asiatic black bear, etc.) and the location of discovery information (address, area around major buildings, etc.) from the converted text.

[0047] Since the technology for converting voice signals into text is a known technology, a more detailed explanation is omitted.

[0048] Extracting information on the types of wild animals and their locations from text can be accomplished by comparing words corresponding to the types of wild animals with a proprietary database to determine whether words for addresses or major landmarks are included.

[0049] For example, the type of wild animal can be identified by sequentially extracting wild animal names from a database containing wild animal names living in Korea and checking whether the wild animal name is included in the report made by the reporter (400).

[0050] If the type of wild animal is not specified in the report from the reporter (400), it can be processed using artificial intelligence.

[0051] For example, you can input text into ChatGPT (a service by OpenAI), which is widely used these days, to identify the type of wild animal.

[0052] For example, even if the reporter (400) does not mention a specific animal name and only describes its appearance, the type of wild animal can be identified through a query to ChatGPT.

[0053] This can be applied equally even when specifying the location of discovery.

[0054] The analysis target selection unit (140) performs the function of selecting an analysis target image that is captured and stored by a shooting device (200) within a preset radius based on wild animal discovery location information extracted by the information extraction unit (130).

[0055] That is, when selecting a target image for analysis from among the countless images stored by the image storage unit (110), the wild animal discovery location information is used as a standard.

[0056] For example, the analysis target selection unit (140) may select a captured image received from a shooting device (200) located within a radius of 3 km from the wild animal discovery location information based on a report as the captured image to be analyzed. To this end, each captured image stored by the captured image storage unit (110) may have information about the location installed in the shooting device (200) where the shooting took place stored in a matched manner.

[0057] Meanwhile, the artificial intelligence module generation unit (160) performs the function of constructing an artificial intelligence module by taking a captured image and wild animal type information as inputs and performing machine learning to determine whether the wild animal type information exists in the captured image as output.

[0058] For example, supervised learning can be performed with a picture containing a cat and the text 'cat' as input, so that 'correct' appears in the output (in this case, 'correct' can be said to be labeled), and as another example, supervised learning can be performed with a picture without a dog and the text 'dog' as input, so that 'incorrect' appears in the output (in this case, 'incorrect' can be said to be labeled).

[0059] The wild animal location determination unit (150) performs the function of determining the location where a wild animal is found using the image of the target selected by the analysis target selection unit (140) and the wild animal type information extracted by the information extraction unit (130).

[0060] For example, the wild animal location determination unit (150) inputs the wild animal type information extracted from the information extraction unit (130) and the analysis target image selected by the analysis target selection unit (140) into the artificial intelligence module, and if the output of the artificial intelligence module is 'correct', the wild animal is included in the input image, and therefore the location of the shooting device (200) that took the image can be determined as the location where the wild animal was discovered.

[0061] As another example, the wild animal location determination unit (150) inputs the wild animal type information extracted from the information extraction unit (130) and the analysis target image selected by the analysis target selection unit (140) into the artificial intelligence module, and if the output of the artificial intelligence module is 'incorrect', it means that the wild animal is not included in the input image, so the image can be ignored.

[0062] Meanwhile, the drone control unit (170) determines the movement path of the wild animal using the multiple wild animal discovery locations determined by the wild animal location determination unit (150), and then performs the function of controlling the wild animal to be tracked using a drone shooting device (300) equipped with a shooting function.

[0063] That is, since the direction and speed of movement of the wild animal can be determined using the discovery location and discovery time of multiple wild animals determined by the wild animal location determination unit (150) (i.e., the discovery time can be determined using the time of reporting), the drone control unit (170) can move the drone camera device (300) along the movement path of the wild animal and analyze the wild animal in the video captured by the drone camera device (300).

[0064] Here, the drone shooting device (300) continuously transmits video, and the drone control unit (170) can determine whether the wild animal is found by analyzing the video received from the drone shooting device (300).

[0065] That is, in the analysis process by the analysis target selection unit (140), still images are used to prevent unnecessary resource waste, but since the drone shooting device (300) is used during the process of tracking wild animals, the entire video is continuously analyzed to facilitate the early detection of wild animals.

[0066] Hereinafter, the overall control flow of a wild animal tracking system (100) according to one embodiment of the present invention will be explained with reference to FIG. 3.

[0067] First, the wild animal tracking system (100) performs the function of receiving and storing images captured by the shooting device (200) (step S1).

[0068] For example, when a video is captured and received by a shooting device (200), the wild animal tracking system (100) may generate and store a captured image considering whether movement has occurred, and may use it to determine the location of the wild animal.

[0069] When a report of a wild animal sighting is received from a reporter (400) (step S3), the wild animal tracking system (100) extracts information on the type of wild animal and the location of discovery from the received report (step S5).

[0070] Next, the wild animal tracking system (100) selects a captured image to be analyzed based on the extracted discovery location information (step S7), and finally determines the wild animal discovery location by checking whether the selected captured image to be analyzed includes an image corresponding to a type of wild animal (step S9).

[0071] Afterwards, the wild animal tracking system (100) can be controlled to track the location of the wild animal using a drone camera (300) (step S11).

[0072] Meanwhile, it goes without saying that the process of carrying out each of the above-described embodiments may be performed by a program or application stored on a predetermined recording medium (e.g., computer-readable). Here, the recording medium includes all types of electronic recording media such as RAM (Random Access Memory), magnetic recording media such as hard disks, and optical recording media such as CDs (Compact Disks).

[0073] At this time, the program stored on the recording medium can be executed on hardware such as a computer or a smartphone to perform each of the embodiments described above. In particular, at least one of the functional blocks of the wild animal tracking system according to the present invention described above can be implemented by such a program or application.

[0074] Furthermore, the present invention is not limited to the specific embodiments described above, but can be implemented with various modifications and variations within the scope of the essence of the invention. It will be obvious that such modifications and variations are included in the present invention if they fall within the scope of the appended claims.

[0075] As explained above, according to the present invention, the location of wild animals can be determined and tracked based on reports of wild animal sightings received from citizens at minimal cost.

[0076] In particular, storing video footage received from numerous recording devices (e.g., CCTV) is not only costly but can also take a long time to analyze. However, the present invention allows for the rapid identification of wild animal locations at minimal cost by managing still images and selecting images for analysis based on reports from citizens.

Claims

1. (a) a step of receiving and storing images captured from each capturing device; and (b) the step of receiving reports of wild animal sightings; (c) a step of extracting wild animal type information and wild animal discovery location information from the wild animal discovery report of step (b) above; and (d) a step of selecting an image to be analyzed, which is captured and stored by a camera within a preset radius based on the wild animal discovery location information extracted in step (c); and (e) A method for controlling a wild animal tracking system characterized by including a step of determining the location of discovery of a wild animal using the captured image of the target for analysis selected in step (d) and the wild animal type information extracted in step (c).

2. In Paragraph 1, The wild animal sighting report in step (b) above is a voice signal via ARS, and A control method for a wild animal tracking system characterized by analyzing the ARS voice signal in step (c) above to extract information on the type of wild animal and the location where the wild animal was found.

3. In Paragraph 1, It further includes a step of constructing an artificial intelligence module by performing machine learning that takes captured images and wild animal type information as input and outputs whether the wild animal type information exists in the captured images, and A control method for a wild animal tracking system characterized by inputting the captured image of the target for analysis selected in step (d) and the wild animal type information extracted in step (c) into the artificial intelligence module in step (e), and determining the location where the wild animal was found using the output of the artificial intelligence module.

4. In Paragraph 1, A method for controlling a wild animal tracking system, characterized by further including the step of determining the movement path of a wild animal using the multiple locations of wild animals determined in step (e) above, and then controlling the wild animal to be tracked using a drone camera equipped with a camera function.

5. In Paragraph 1, A control method for a wild animal tracking system characterized by receiving a captured image when movement is detected by a motion detection sensor attached to the above-mentioned shooting device in step (a).

6. A computer-readable recording medium storing a program for executing the method of any one of paragraphs 1 through 5.

7. A shooting image storage unit that receives and stores shooting images from each shooting device; A report reception desk that receives and stores reports of wild animal sightings; An information extraction unit that extracts wild animal type information and wild animal discovery location information from wild animal discovery reports stored by the above-mentioned report receiving unit; An analysis target selection unit that selects an analysis target image captured and stored by a shooting device within a preset radius based on wild animal discovery location information extracted by the above information extraction unit; A wild animal tracking system characterized by including a wild animal location determination unit that determines the location where a wild animal is found using an image of a target selected by the analysis target selection unit and wild animal type information extracted by the information extraction unit.

8. In Paragraph 7, The wild animal sighting reports stored by the above-mentioned reporting department are voice signals via ARS, and A wild animal tracking system characterized by the above information extraction unit analyzing the above ARS voice signal to extract wild animal type information and wild animal discovery location information.

9. In Paragraph 7, It further includes an AI module generation unit that constructs an AI module by performing machine learning that takes captured images and wild animal type information as inputs and outputs whether the wild animal type information exists in the captured images. A wild animal tracking system characterized by the above wild animal location determination unit inputting an image of a target selected by the above analysis target selection unit and wild animal type information extracted by the above information extraction unit into the above artificial intelligence module, and determining the location where the wild animal was discovered using the output of the said artificial intelligence module.

10. In Paragraph 7, A wild animal tracking system characterized by further including a drone control unit that determines the movement path of a wild animal using a plurality of wild animal discovery locations determined by the wild animal location determination unit, and controls the wild animal to be tracked using a drone shooting device equipped with a shooting function.

11. In Paragraph 7, A wild animal tracking system characterized by the above-mentioned image storage unit receiving a captured image when movement is detected by a motion detection sensor attached to the above-mentioned shooting device.

Images