Learning-based trajectory tracking and analysis of bees

Abstract

The present application relates to a device and method for tracking and analyzing the trajectory of a bee, and more particularly to a device and method for tracking and analyzing the trajectory of a bee using learning data acquired by learning the features of the body structure and color of a bee through a deep learning model.

Description

Technical Field

[0001] This invention relates to a device and method for tracking and analyzing the trajectory of bees, and more particularly to a device and method for tracking and analyzing the trajectory of bees using learning data obtained by learning features such as the body structure and color of bees through a deep learning model. Background Technology

[0002] Honeybees are bees used for storing and producing honey, while bumblebees are bees used for pollinating plants, pollinating flowers through buzz pollination.

[0003] Bee pollination refers to the behavior of collecting pollen by vibrating the pectoral muscles to transfer pollen to the hairy tufts on the body. This vibration helps the pollen to fall onto the stigma of the pistil.

[0004] Therefore, bumblebees (also known as garden bumblebees or red-glowing bumblebees) are mostly used for pollination of plants in the Solanaceae family such as tomatoes, eggplants, and peppers, which are difficult to pollinate by bees due to the lack of nectar, or for the cultivation of crops such as peaches, plums, apples, apricots, strawberries, cantaloupes, pears, blueberries, raspberries, and mangoes.

[0005] For example, when cultivating tomatoes, bumblebees are released into the colony. Typically, 10% to 20% of the worker bees will collect pollen 5 to 12 times a day, and during each collection activity, they will fly to 50 to 220 flowers.

[0006] If the number of bees far exceeds the number of flowers, the excessive number can lead to deformed fruit or poor pollination. Therefore, using beehives with separate entrances and exits and a return port, and adjusting the switches according to the time period and bee activity, can control the bumblebee's foraging activities.

[0007] If a specified period of time passes after setting up a beehive, the number of individual bumblebees will decrease, and their activity will also decrease, thus requiring the beehive to be replaced. If the temperature exceeds 30 degrees Celsius, ventilation is needed to lower the temperature of the beehive, which will cause more bees to remain in the beehive, resulting in reduced pollination activity. If the temperature exceeds 33 degrees Celsius, the bees will enter survival mode, killing the larvae and ceasing pollination activities. To confirm this situation, there are already proposed bumblebee management methods that allow for remote monitoring of the bumblebee's status.

[0008] Korean Patent Publication No. 10-2016-0141224, "Bumblebee Management Device and Management System," discloses a technique that uses a sensor to detect the number of times bumblebees enter and exit through the entrance and exit of a bumblebee box to remind bumblebees of replacement time. Korean Patent No. 10-1963648, "Greenhouse Bumblebee Management System and Method, Pollination Bee Box," discloses a technique that uses multiple sensors to detect the direction and number of times bees enter and exit the pollination bee box to control the opening and closing of the entrance and exit.

[0009] However, the aforementioned existing technologies require the construction of an environment that may affect bees, such as installing sensors in the hive to detect the entry and exit of bumblebees. The effect is limited to counting the number of individual bees, and there are also problems such as failing to track the bees' trajectories or provide information related to the appropriateness of environmental changes or bee pollination activities by analyzing bumblebee movements, or providing the necessary information for hive and greenhouse management.

[0010] On the other hand, many bumblebee farms sell their hives by the hive unit, but even newly purchased hives may contain bees in poor health. In such cases, it is necessary to assess the bees' health and provide advance warnings to ensure farmers do not miss pollination opportunities, but the aforementioned existing technologies do not address this problem or offer solutions.

[0011] Existing technical documents

[0012] Patent documents

[0013] Patent Document 1: Korean Patent Publication No. 10-2016-0141224 (published on December 8, 2016)

[0014] Patent Document 2: Korean Patent No. 10-1963648 (Published on April 1, 2019) Summary of the Invention

[0015] To address the aforementioned problems, the present invention aims to provide an apparatus and method for tracking bee trajectories, determining whether bees have entered or exited, and counting the number of bees that have entered or exited by comparing the movement and learning data of bees captured near a beehive.

[0016] Furthermore, the present invention aims to provide an apparatus and method for analyzing, using captured images, the amount of pollen collected by bees, the appropriateness of bee pollination activities, and the health status of bees.

[0017] Furthermore, the present invention aims to provide an apparatus and method for determining the appropriateness of bee pollination activities by analyzing bee trajectories and pollen status in captured images, and thereby determining the hive replacement time when necessary.

[0018] Furthermore, the object of the present invention is to provide an apparatus and method for alerting an abnormal individual when such an individual approaches by utilizing pre-learned data.

[0019] An embodiment of the present invention for achieving the above objectives, a learning-based bee trajectory tracking and analysis apparatus and method, includes: a camera unit for capturing images of bees moving near the entrance / exit of a beehive; a storage unit for storing learning data obtained by learning the body characteristics of bees through a deep learning model; and a control unit for tracking the captured bee trajectory using the images captured by the camera unit and the learning data stored in the storage unit.

[0020] Furthermore, the aforementioned learning data also includes data related to the morphological characteristics of pollen, and the control unit uses the pollen image included in the aforementioned image and the morphological characteristics of pollen included in the aforementioned learning data to analyze the amount of pollen.

[0021] Furthermore, the aforementioned control unit defines multiple virtual areas of different sizes centered on the entrance and exit of the aforementioned beehive. When a bee is detected in the captured image, the area around the bee is defined as the beehive.

[0022] Furthermore, the aforementioned control unit distinguishes whether the bees have entered or exited by recording or tracking the sequence of at least one area that overlaps with the aforementioned beehive.

[0023] Furthermore, the aforementioned area includes area 1, which occupies a specified area centered on the aforementioned beehive entrance and exit, and area 2, which includes area 1 and occupies an area larger than area 1.

[0024] Furthermore, when pollen is detected in the captured images, the control unit defines the area surrounding the pollen as a pollen box.

[0025] Furthermore, the pollen boxes mentioned above are generated within the aforementioned beehives.

[0026] Furthermore, the control unit calculates the change in the relative distance from the center point of the entrance / exit to the beehive and the number or frequency of times the beehive appears in the virtual area.

[0027] Furthermore, the control unit distinguishes between bees entering the beehive and bees exiting the beehive by comparing the bee trajectories captured in the virtual area with the bee trajectories included in the learning data.

[0028] Furthermore, the control unit distinguishes between normal and abnormal bees by comparing the trajectories of bees captured in the virtual area with the trajectories of bees included in the learning data.

[0029] Furthermore, the aforementioned control unit determines the replacement time of the beehive by calculating the ratio or number of normal and abnormal bees.

[0030] Furthermore, the control unit counts and accumulates the images of bees captured in the virtual area in units of image frames.

[0031] Furthermore, the control unit calculates the trajectory of the bees around the entrance / exit of the beehive in a manner that corresponds to the distance on the X-axis and the distance on the Y-axis starting from the entrance / exit.

[0032] Furthermore, when the beehive overlaps with the area within a specified period in the order from area 1 to area 2, the control unit distinguishes bees exiting the beehive; when the beehive overlaps with the area in the order from area 2 to area 1, the control unit distinguishes bees entering the beehive.

[0033] Furthermore, the device also includes a display unit, and the control unit displays the number of bees entering and exiting the beehive by counting the number of bees in and out of the beehive.

[0034] Furthermore, the control unit displays the amount of pollen on the display unit.

[0035] Furthermore, the control unit displays the color of the pollen on the display unit.

[0036] Furthermore, the aforementioned learning data also includes data related to the morphological characteristics of anomalous individuals. The control unit analyzes whether anomalous individuals exist by utilizing the images of anomalous individuals included in the aforementioned images and the morphological characteristics of anomalous individuals included in the aforementioned learning data.

[0037] Furthermore, in the aforementioned images captured, the control unit uses interpolation to predict the location of bees in the regions where they are not detected.

[0038] Furthermore, in order to locate the pollen distribution area within the pollen box, the control unit performs image preprocessing to form a contour line at the boundary of the pollen distribution area and calculates the color and concentration of the area inside the contour line.

[0039] According to the present invention, an apparatus and method will be provided for tracking the trajectory of bees and determining whether bees have entered or exited by comparing the movement and learning data of bees photographed near the beehive, and counting the number of bees that have entered or exited.

[0040] Furthermore, according to the present invention, an apparatus and method will be provided for analyzing, using captured images, the amount of pollen collected by bees, the appropriateness of bee pollination activities, and the health status of bees.

[0041] Furthermore, according to the present invention, an apparatus and method will be provided for determining the appropriateness of bee pollination activities by analyzing the bee's trajectory and pollen status in captured images, and thereby determining the hive replacement time when necessary.

[0042] Furthermore, according to the present invention, an apparatus and method will be provided for alerting an abnormal individual when such an individual approaches by utilizing pre-learned data. Attached Figure Description

[0043] Figure 1 This is a structural diagram of a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0044] Figures 2 to 15 This is an example image of a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0045] Figure 16 This is an analysis diagram of the moving distance and occurrence frequency of bees in a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0046] Figures 17 to 19 This is an example diagram illustrating trajectory tracking using a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0047] Figures 20 to 23 This diagram illustrates an example of bee state analysis using a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention. Detailed Implementation

[0048] The advantages and features of the present invention, and methods for implementing them, will become clear from the accompanying drawings and the various embodiments described in detail therewith. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in many different ways. The embodiments described herein are provided only to complete the disclosure of the invention and to give a full understanding of the scope of the invention to those skilled in the art. The invention is defined only by the scope of the claims. Throughout the specification, the same reference numerals denote the same structural elements.

[0049] The terminology used in this specification is for illustrative purposes only and is not intended to limit the invention. Unless otherwise expressly indicated in the context, singular expressions include plural expressions. It should be understood that, in this specification, terms such as "comprising" or "possessing" are used to indicate the presence of features, numbers, steps, actions, structural elements, components, or combinations thereof described in this specification, rather than precluding the presence or additional possibilities of one or more other features or numbers, steps, actions, structural elements, components, or combinations thereof.

[0050] In this specification, terms such as “part,” “module,” “device,” “terminal,” “server,” or “system” are used to refer to a combination of hardware and software driven by that hardware. For example, hardware may be a central processing unit (CPU) or a data processing device that includes other processors. Furthermore, software driven by hardware may be a running program, object, executable file, thread of execution, program, etc.

[0051] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0052] Figure 1 This is a structural diagram of a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0053] The device 100 includes: a control unit 110, which tracks the trajectory of bees and analyzes the amount of pollen collected through a trajectory tracking module 120 and a pollen analysis module 130; and a storage unit 150, which stores the learning data obtained by learning various information about bees and pollen through a deep learning model.

[0054] In addition, it may include: a camera unit 140 for capturing images of bees moving near the beehive; a display unit 160 for displaying the captured images; and a communication unit 170 for performing wired or wireless communication with external devices.

[0055] When using an external camera 200 instead of the built-in camera unit 140, the device 100 can perform data transmission and reception with the external camera 200 through the communication unit 170. An external display device (not shown) can be used instead of the display unit 160 built into the device 100, or the external display device can be used in conjunction with the display unit 160.

[0056] The device of the present invention can learn the characteristics of bumblebees, such as body structure or color distribution, through a deep learning model, and store them as learning data in the storage unit 150. It compares the video received from the camera unit 140 and stored in the storage unit, as well as the streaming video data received from the camera unit 140 and the learning data frame by frame, to find bumblebees in the images, track their movement paths, find pollen, and analyze the amount of pollen.

[0057] Figures 2 to 15 This is an example image of a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0058] The display screen provided by the device of the present invention can have the following characteristics: Figure 2 The structure shown.

[0059] Basically, bumblebees are identified by calculating the intersection between multiple quadrilaterals (hereinafter referred to as "boxes") virtually drawn near the entrance of the beehive and the quadrilaterals (hereinafter referred to as "beehives") automatically drawn when identifying bees.

[0060] exist Figure 2 In the diagram, the entrance and exit of the beehive are used as the center, and the area outside is divided into three regions: Area 1 (inside the red box), Area 2 (inside the blue box), and Area 3 (outside the blue box). The area around the bees is defined as the beehive (green box).

[0061] The device of the present invention distinguishes whether a bee is entering or leaving the beehive by recording or tracking the order in which the area inside the box intersects with the beehive.

[0062] For example, if the movement of the beehive (or the intersection of the beehive and the frame) is recorded in the order of area 1 → area 2 → area 3, it is considered as leaving the beehive; if it is recorded in the order of area 3 → area 2 → area 1, it is considered as entering the beehive; if it stays in area 1 for a certain period of time and then disappears, it is considered as entering the beehive.

[0063] exist Figure 2 In Chinese, "hive" refers to the number of bees present in a hive, which is a value obtained by counting the number of times multiple bees enter and exit.

[0064] "Pollination" refers to the amount of pollen collected by bees, "heavy" indicates a good state of pollen collection, and "light" indicates a relatively insufficient state of pollen collection.

[0065] "LoC" indicates the region where the bees (or beehives) are located.

[0066] On the other hand, in this invention, the IoU (intersection over union) metric, used as an indicator to evaluate the accuracy of object detection, is defined as follows.

[0067] The intersection value calculated as IoU = (intersection area of ​​the box and the beehive) / (sum of the areas of the box and the beehive) will be displayed in [the image / image / etc.]. Figure 2 In the image, the "pIoU1", "pIoU2", "cIoU1", and "cIoU2" at the top right are the values. The first letter "p" means the previous value, and the first letter "c" means the current value.

[0068] “pIoU1” represents the value measured in the previous frame in region 1 (red box), “pIoU2” represents the value measured in the previous frame in region 2 (blue box), “cIoU1” represents the value measured in the current frame in region 1, and “cIoU2” represents the value measured in the current frame in region 2.

[0069] Not only "IoU", but also the values ​​of "Hive" and "Pollination" are updated in each frame of the image.

[0070] On the other hand, bee and pollen identification is performed by comparing stored learning data with captured images. When data on individuals such as bees or pollen grains are applied to captured images, the probability that an individual in an image acquired at any given time period is a bee or pollen grain can be numerically displayed near the beehive or pollen box. For example, if "0.99" is displayed next to the beehive, it means that the probability that the individual is a "bee" is 99%.

[0071] Pollen boxes are generated inside beehives. The value displayed next to the pollen box represents the probability that an image that intersects with the beehive is predicted to be pollen. In other words, it represents the probability that an overlapping pollen image is predicted to be pollen.

[0072] To locate regions predicted to be pollen, image preprocessing is used to draw outlines in the corresponding regions, and the color and density of the regions inside the outlines (i.e., the ratio of pixels to a specific color) are calculated.

[0073] It can be confirmed that the distribution pattern of pollen adhering to a bee's body has a prescribed correlation with the degree of pollination activity of the bee. The shape and radius of curvature of the outline can be used as a benchmark to distinguish the degree of pollination activity of the bee. For example, if the outline is elliptical or the pollen distribution area is wide, it can be judged that the bee has little or no pollination activity, while if the pollen distribution area is small and close to a circle, it can be judged that the bee has vigorous and smooth pollination activity.

[0074] Figure 3 This shows the amount of pollen expressed numerically when the amount collected by bees exceeds a specified baseline and is therefore relatively abundant (heavy). Figure 4 The image shown here is characterized by pollen being represented by a relatively clear and bright color, and its boundary lines are also very distinct.

[0075] Figure 5 This illustrates the numerical representation of pollen quantity when the amount collected by bees is relatively insufficient (light) due to a lack of a specified baseline. Figure 6 The pollen image shown in this case is characterized by pollen being represented by a relatively blurry or dark color, and its boundary lines being blurred or scattered across multiple regions or a wide area.

[0076] Figures 7 to 11 The image shows the trajectory of bees entering the hive.

[0077] exist Figure 7 In the footage, after filming began, four bees entered the hive (In: 4), and one bee left the hive (Out: 1), leaving a total of three bees in the hive (Hive: 3). Furthermore, the amount of pollen collected by the four bees entering the hive ranged from one (Pollination-heavy: 1) to three (Pollination-lihgt: 3).

[0078] Furthermore, in both the previous and current frames, no bee movement was detected in any region (pIoU1, cIou1, pIoU2, cIoU3 = '0'). In the image, there is an image or device of a bee at the upper left of region 2, but it has not yet been identified as a bee.

[0079] Next, in Figure 8 In the image, within region 2, the IoU value increases in the manner of pIoU2: 0.009 → cIoU: 0.014, which is used to identify bees entering region 2. Figure 9 In the image, as the bees enter region 1, the IoU value in region 1 changes in the manner of pIoU1: 0.048 → cIoU: 0.036, thus creating an intersection with the beehive. Figure 10 In the video, as the bee enters the entrance / exit, both the IoU1 and IoU2 values ​​decrease, but... Figure 11 In the image, because the bee completely enters the hive and disappears from the image, the "Hive" value (IoU1=Iou2='0') increases to 4, and the "Pollination-light" value increases to 4 due to the amount of pollen collected by the bee.

[0080] on the other hand, Figures 12 to 15 The image shows the trajectory of bees leaving the hive.

[0081] Figure 12 The image in the middle represents from Figure 11 After a certain period of time, two bees re-enter the hive, and one bee comes out of the hive. One bee has "heavy" pollen, while the remaining bee has "light" pollen.

[0082] Next, in Figure 13 In the images, as bees are captured emerging from the hive, the IoU value increases simultaneously in both region 1 and region 2. Figure 14In the image, as the bee moves from region 2 to region 3, the IoU value does not change in region 1, and the IoU value in region 2 decreases in the manner of pIoU: 0.018 → cIoU: 0.0, which is regarded as the bee leaving the hive. The "Hive" value decreases to 4, and the "Out" value increases to 3.

[0083] Finally, the bee completely disappeared from the image. Figure 15 In this case, other values ​​remain unchanged, and all IoU values ​​remain at "0".

[0084] On the other hand, simply calculating the IoU (Intersection over Union) based on whether there is an intersection is insufficient to locate the bee's trajectory. For example, methods are needed to address situations where individual recognition ability declines due to insufficient learning or where individuals suddenly disappear from the image.

[0085] Therefore, after finding the initial trajectory using an algorithm that cross-analyzes the intersections (pIoU1, pIoU2) in previous images and the current intersections (cIoU1, cIoU2), interpolation is used to supplement the movement distance. In other words, in cases where the recognition rate drops due to insufficient learning of individuals or poor image learning (i.e., learning from blurry photos), or when bees suddenly disappear from the image, an interpolation method will be used to predict where the bees will be.

[0086] like Figure 16 As shown, the device of the present invention can generate a chart of the trajectory of an analytical bee by analyzing abnormal images.

[0087] Figure 16 This is an analysis diagram of the moving distance and occurrence frequency of bees in a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0088] Figure 16 To display both the bee's traveling distance and the number of appearances in the trajectory chart, the movement and status of the bees can be understood by comprehensively analyzing the bee's trajectory line (blue) representing the distance from the beehive entrance to the beehive and the appearance line (red) representing the number of times the beehive appears.

[0089] Figures 17 to 19 This is an example diagram illustrating trajectory tracking using a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0090] For example, with Figure 17 The trajectory of a normal bee entering the hive, as shown, or as... Figure 18 Compared to the trajectory of a normal bee emerging from the hive, the following is a comparison: Figure 19 The bees stopped moving around the entrance and exit, indicating that they had fallen to their deaths.

[0091] Figures 20 to 23 This diagram illustrates an example of bee state analysis using a learning-based bee trajectory tracking and analysis device according to an embodiment of the present invention.

[0092] The device of the present invention can provide state analysis diagrams of bees in various morphologies. For example, Figure 20 The example images include a flight trajectory chart representing the bee's flight path on the XY coordinate system, and an observation of the bee's staying status chart representing the cumulative frequency of bees in the image. Figures 21 to 23 The example diagram includes the traveling distance of the bee starting from the reference point and a graph showing the bee's presence status.

[0093] It is possible Figure 20 The chart in the diagram can be used to confirm the normal movement of bees exiting the hive. Figures 21 to 23 The movement of bumblebees that had been wandering around the hive and fallen to their deaths was confirmed.

[0094] In particular, it is possible Figure 21 and Figure 22 The diagram shows the trajectory of bees continuously rotating around the entrance and exit, without increasing or decreasing the distance traveled from the reference point, maintaining a constant state within a specified time. On the other hand, this movement of bees also occurs when observing healthy bees immediately after setting up the hive, because they need time to adapt to the unfamiliar environment; this situation is judged as "normal."

[0095] Observations show that normal bees take about 1 hour to enter the hive and even less time to leave. However, this entry and exit time gradually increases over time after the initial hive replacement, indicating that the bees' activity level is gradually decreasing due to aging and energy depletion.

[0096] When shooting video at 30 frames per second, if the tracked bee's entry and exit trajectory is within 100 steps (Note: the baseline for the number of steps can be set arbitrarily and can be learned through the initialization program), it is considered normal. If it exceeds 200 steps, it is considered to have a problem. If it exceeds 300 steps, it is considered to be time to replace the beehive.

[0097] Furthermore, the device of the present invention can use pre-learned data to alert the user when an abnormal individual approaches. For example, it can automatically alert the user when an abnormal individual approaches by analyzing captured images based on the characteristics (size, pattern, movement) of individuals harmful to bumblebees (such as wasps or lice).

[0098] Furthermore, the device of this invention can learn data such as pollen morphology, color, and normal and abnormal bee movement. Therefore, since pollen color can be distinguished in the captured images, it can be determined whether bees are seeking other plants of a different species than the target plant, and whether bee movement is normal or abnormal. In this way, by analyzing bee movement, information related to environmental changes or the appropriateness of bee pollination activities can be provided, or information needed for beehive and greenhouse management can be provided.

[0099] The present invention has been described above through specific structural elements and other specific matters, limiting embodiments and accompanying drawings. However, this is only for a more comprehensive understanding of the present invention. The present invention is not limited to the above-described multiple embodiments. Those skilled in the art to which this invention pertains can make various modifications and variations based on this description.

[0100] Therefore, the concept of the present invention is not limited to the embodiments described above. All technical solutions that are modified in a manner equivalent to or equivalent to the scope of the invention claims are within the scope of the present invention.

Claims

1. A learning-based bee trajectory tracking and analysis device, characterized in that, include: The camera unit is used to capture the movement of bees near the hive entrance and exit; The storage unit stores the learning data acquired through deep learning models of bee body characteristics; and The control unit uses the images captured by the aforementioned camera unit and the learning data stored in the aforementioned storage unit to track the trajectory of the captured bees. The aforementioned control unit defines multiple virtual areas of varying sizes centered on the entrance and exit of the aforementioned beehive. When a bee is detected in the captured image, the area surrounding the bee is defined as the beehive. The aforementioned area includes area 1, which occupies a specified area centered on the entrance / exit of the aforementioned beehive, and area 2, which includes area 1 but occupies a larger area than area 1. The control unit calculates the cross-connection-to-union (CTU) ratio between the beehive and each region for each image frame, and tracks the CTU ratios pIoU1 and pIoU2 for region 1 and region 2 in the previous frame, respectively, as well as the CTU ratios cIoU1 for region 1 and cIoU2 for region 2 in the current frame. The control unit records the overlap order of the beehive with each region based on the change in the intersection-over-parallel ratio of the previous frame and the current frame, thereby distinguishing whether the bee enters the beehive or flies out of the beehive.

2. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The aforementioned learning data also includes data related to the morphological characteristics of pollen. The control unit analyzes the amount of pollen using pollen images included in the above-mentioned images and morphological characteristics of pollen included in the above-mentioned learning data.

3. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, When pollen is detected in the captured images, the control unit defines the area surrounding the pollen as a pollen box.

4. The learning-based bee trajectory tracking and analysis device according to claim 3, characterized in that, The pollen boxes mentioned above are generated within the beehives mentioned above.

5. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The control unit calculates the change in the relative distance from the center point of the beehive entrance / exit to the beehive and the number or frequency of the beehive appearing in the virtual area.

6. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The control unit distinguishes between bees entering the beehive and bees leaving the beehive by comparing the bee trajectories captured in the virtual area with the bee trajectories included in the learning data.

7. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The control unit distinguishes between normal and abnormal bees by comparing the trajectories of bees captured in the virtual area with the trajectories of bees included in the learning data.

8. The learning-based bee trajectory tracking and analysis device according to claim 7, characterized in that, The aforementioned control unit determines the replacement time of the aforementioned beehives by calculating the ratio or number of normal bees and abnormal bees.

9. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The control unit counts and accumulates the images of bees captured in the virtual area in units of image frames.

10. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The control unit calculates the trajectory of the bees around the beehive entrance / exit in a manner that corresponds to the distance on the X-axis and the distance on the Y-axis starting from the beehive entrance / exit.

11. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, When the beehive overlaps with the area in the order from area 1 to area 2 within a specified period, the control unit distinguishes bees exiting the beehive; when the beehive overlaps with the area in the order from area 2 to area 1, the control unit distinguishes bees entering the beehive.

12. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, It also includes the display section, The control unit displays the number of bees entering and exiting the beehive on the display unit.

13. The learning-based bee trajectory tracking and analysis device according to claim 2, characterized in that, It also includes the display section, The control unit displays the amount of pollen on the display unit.

14. The learning-based bee trajectory tracking and analysis device according to claim 13, characterized in that, The control unit displays the color of the pollen on the display unit.

15. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, The aforementioned learning data also includes data related to the morphological characteristics of anomalous individuals. The control unit analyzes whether there are abnormal individuals by using the images of abnormal individuals included in the above images and the morphological features of abnormal individuals included in the above learning data.

16. The learning-based bee trajectory tracking and analysis device according to claim 1, characterized in that, In the aforementioned images, the control unit uses interpolation to predict the location of bees in areas where they are not detected.

17. The learning-based bee trajectory tracking and analysis device according to claim 3, characterized in that, In order to locate the pollen distribution area within the pollen box, the control unit performs image preprocessing to form a contour line at the boundary of the pollen distribution area and calculates the color and concentration of the area inside the contour line.

Images