A method for analyzing the foraging behavior of pollinating bees based on RFID data

Abstract

The present invention relates to the field of agricultural modernization monitoring technologies, and particularly relates to a method for analyzing the foraging behavior of pollination bees based on RFID data, which is used to solve problems such as the complexity, redundancy, and difficulty in processing the behavior data of pollination bees. Specifically, it includes: obtaining data such as a set of reader numbers; dividing the set of reader numbers into several data segments; traversing the data segments and marking data tags and data types, where the data tags include: potential out-of-hive and potential return-to-hive; and the data types include: the first type of data, the second type of data, the third type of data, and the fourth type of data; sequentially reading the data segments of each pollination bee and finding the data segments that can be marked as out-of-hive according to the data type and the data type or data tag of the next data segment, then reading the next data segment, and finding the data segments that can be marked as return-to-hive according to the tag type or data type, until all data segments are read, so as to judge the behavior state of each pollination bee and complete the analysis of the foraging behavior of the pollination bee population.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural modernization monitoring, and particularly to a method for analyzing the foraging behavior of pollinating bees based on RFID data. Background Art

[0002] In the field of pollinating bee rearing and pollination behavior research, it is necessary to detect and record the number of times pollinating bees leave and return to the hive, so as to analyze and study the in-and-out hive activity habits, foraging behaviors, etc. of the pollinating bee colony. In early research, the identification of pollinating bee foraging behaviors mostly relied on manual annotation, resulting in a cumbersome and inefficient data processing process. In addition, technologies such as machine learning can also be used, but currently, machine learning technologies lack an in-depth understanding of pollinating bee foraging behaviors and fail to fully utilize the prior knowledge in the field. At the same time, although existing machine learning-based technologies have high recognition accuracy, they are often "black box" models and are difficult to provide researchers with interpretable behavior rules. For example, deep learning models such as neural networks may be able to identify the behaviors of bees, but it is difficult for researchers to trace how these behaviors are identified.

[0003] In summary, a rule-based identification method is needed to clearly define the criteria for foraging behaviors, making the foraging behavior identification results more transparent and understandable. Currently, RFID (Radio Frequency Identification) technology can be used to collect the behavior data of pollinating bees, but the behavior data is very complex and there is a large amount of redundant data, and a method for analyzing the foraging behavior of pollinating bees based on RFID data is needed for processing. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for analyzing the foraging behavior of pollinating bees based on RFID data, which can analyze the behavior data of pollinating bees.

[0005] To solve the above problems, the present invention provides a method for analyzing the foraging behavior of pollinating bees based on RFID data, including the following steps:

[0006] Step 1000: Obtain the types of pollinating bees, the pollination environment, and the RFID scan data of each pollinating bee, where the RFID scan data includes the ID information of the pollinating bee, the scan timestamp, and the set of reader numbers passed through;

[0007] Step 2000: Set a foraging behavior threshold for each pollinating bee according to the types of pollinating bees, the pollination environment, and the preset standard foraging behavior threshold based on historical data, and divide the set of reader numbers into several data segments according to the scan timestamp and the foraging behavior threshold;

[0008] Step 3000: Traverse the data segments of each of the pollinating bees and mark data tags, where the data tags include: potential leaving the hive and potential returning to the hive. The rules for marking the data tags are as follows: Combine consecutive identical reader numbers within each data segment, and then sequentially determine whether the first two reader numbers in the combined data segment are the first reader and the second reader respectively. If so, mark this data segment as the potential leaving the hive, and then continue to determine whether the last two reader numbers in the combined data segment are the second reader and the first reader respectively. If so, mark this data segment as the potential returning to the hive;

[0009] The first reader is located on the side of the access passage close to the beehive, and the second reader is located on the side of the access passage far from the beehive;

[0010] Step 4000: Traverse the data segments of each of the pollinating bees and mark data types, where the data types include: first type of data, second type of data, third type of data, and fourth type of data. That is, classify and mark the data types of the data segments according to the combined categories of the first reader and the second reader: first type of data, second type of data, third type of data, and fourth type of data. The specific rules are as follows: Determine whether all the reader numbers of each data segment are the first reader. If so, mark it as the first type of data; if not, continue to determine whether all the reader numbers of each data segment are the second reader. If so, mark it as the second type of data. If not, continue to determine whether the last reader number of each data segment is the first reader. If so, mark it as the third type of data. If not, mark it as the fourth type of data;

[0011] Step 5000: Read the data segments of each of the pollinating bees in sequence according to the scanning timestamps. Find the data segments that can be marked as leaving the hive based on the data types of the data segments and the data types or data tags of the next data segment. Then read the next data segment, and find the data segments that can be marked as returning to the hive based on the tag types or data types of this data segment until all the data segments are read.

[0012] In the above method for analyzing the foraging behavior of pollinating bees based on RFID data, step 2000 includes the following steps:

[0013] Step 2100: Set a working coefficient and an environmental coefficient respectively according to the types of pollinating bees and the pollination environment;

[0014] Step 2200: After normalizing the working coefficient and the environmental coefficient, set the foraging behavior threshold in combination with the standard foraging behavior threshold.

[0015] In the above method for analyzing the foraging behavior of pollinating bees based on RFID data, the rule for determining whether the data segment can be marked as leaving the nest in step 5000 is as follows:

[0016] If the read data segment is the first type of data, determine whether the first reader number of the next data segment is the first reader or whether it is marked as potentially leaving the nest. If so, skip this data segment and read the next data segment. If not, mark it as leaving the nest;

[0017] If the read data segment is the second type of data, determine whether the next data segment is marked as potentially leaving the nest. If so, skip this data segment and read the next data segment. If not, mark it as leaving the nest;

[0018] If the read data segment is the third type of data, determine whether the first reader number of the next data segment is the first reader or whether it is marked as potentially leaving the nest. If so, skip this data segment and read the next data segment. If not, mark it as leaving the nest;

[0019] If the read data segment is the fourth type of data, mark it as leaving the nest.

[0020] In the above method for analyzing the foraging behavior of pollinating bees based on RFID data, the rule for determining whether the data segment can be marked as returning to the nest in step 5000 is as follows:

[0021] If the data segment marked as leaving the nest is marked as the first type of data, read the next data segment and determine whether this data segment is marked as potentially returning to the nest. If not, read the next data segment and re - judge. If so, mark it as returning to the nest;

[0022] If the data segment marked as leaving the nest is marked as the second type of data, read the next data segment and determine whether this data segment is not marked as the second type of data or is marked as potentially returning to the nest. If not, read the next data segment and re - judge. If so, mark it as returning to the nest;

[0023] If the data segment marked as leaving the nest is marked as the third type of data, read the next data segment and determine whether this data segment is not marked as the second type of data. If not, read the next data segment and re - judge. If so, mark it as returning to the nest;

[0024] If the data segment marked as the said out-of-hive is marked as the fourth type of data, read the next data segment and mark it as the in-hive.

[0025] In the above method for analyzing the foraging behavior of pollinating bees based on RFID data, the step 5000 includes the following steps:

[0026] Step 5100: Read the data segments of each pollinating bee in sequence according to the scanning timestamp. Determine whether a data segment can be marked as the out-of-hive based on the data type of the data segment and the data type or data label of the next data segment. If it cannot be marked as the out-of-hive, skip this data segment and read the next data segment until a data segment that can be marked as the out-of-hive is found.

[0027] In the above method for analyzing the foraging behavior of pollinating bees based on RFID data, the step 5000 further includes the following steps:

[0028] Step 5200: Read the next data segment of the data segment marked as the out-of-hive. Determine whether it can be marked as the in-hive based on the label type or data type of this data segment. If it cannot be marked as the in-hive, skip this data segment and read the next data segment until a data segment that can be marked as the in-hive is found.

[0029] In the above method for analyzing the foraging behavior of pollinating bees based on RFID data, the step 5000 further includes the following steps:

[0030] Step 5300: Determine whether the data segment marked as the in-hive is marked as the potential out-of-hive. If it is marked as the potential out-of-hive, determine whether the number of data segments between the data segment marked as the out-of-hive and the corresponding data segment marked as the in-hive is less than 3, and proceed to step 5310; if it is not marked as the potential out-of-hive, proceed to step 5320;

[0031] Step 5310: If the number of data segments between the data segment marked as the out-of-hive and the corresponding data segment marked as the in-hive is less than 3, the data segment marked as the in-hive is accurate; if it is greater than or equal to 3, correct the previous data segment of the data segment marked as the in-hive to in-hive;

[0032] Step 5320: Read the next data segment of the data segment marked as the in-hive and determine whether this data segment is marked as the potential in-hive. If it is not marked as the potential in-hive, the data segment marked as the in-hive is accurate;

[0033] If it is marked as the potential homing, it is determined whether the number of data segments included between the data segment marked as the leaving nest and the corresponding data segment marked as the homing is greater than or equal to 3. If it is greater than or equal to 3, the data segment immediately preceding the data segment marked as the homing is corrected to be a homing. If it is less than 3, the data segment marked as the homing is accurate.

[0034] Through the RFID technology, the behavioral data of pollinating bees can be conveniently collected. Then, by analyzing and processing the RFID scanning data of each pollinating bee, the in-and-out nest behavioral states of each pollinating bee can be determined, thereby completing the analysis of the foraging behaviors of the pollinating bee colony. And abnormal data can be corrected, improving the utilization rate of the data and the accuracy of the analysis results. Brief Description of the Drawings

[0035] Figure 1 is a schematic flowchart of the steps of an embodiment of the present invention;

[0036] Figure 2 is one of the schematic data processing flowcharts of an embodiment of the present invention;

[0037] Figure 3 is the other schematic data processing flowchart of an embodiment of the present invention. Detailed Embodiments

[0038] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In addition, in the following descriptions, the descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts of the present invention.

[0039] The embodiments shown in the present invention will be described below with reference to the accompanying drawings.

[0040] RFID (Radio Frequency Identification) technology is a technology that uses radio waves to identify and track targets. The basic working principle is that the reader sends a radio frequency signal of a specific frequency through the transmitting antenna. When the electronic tag enters the effective working area, an induced current is generated, so that it can obtain energy and be activated. Therefore, a small RFID tag can be worn on each pollinating bee, and each RFID tag stores ID information of a unique identity. A first reader and a second reader are set at the inlet and outlet channels of the beehive. The first reader is located on the side of the inlet and outlet channel close to the beehive, and the second reader is located on the side of the inlet and outlet channel far from the beehive. When the pollinating bee leaves the nest, it is scanned by the first reader and the second reader in sequence. When the pollinating bee returns to the nest, it is scanned by the second reader and the first reader in sequence. Thus, it is possible to identify whether the pollinating bee leaves or returns to the nest by the set of reader numbers passed by each pollinating bee, and further analyze its foraging behavior. The first reader and the second reader are responsible for collecting the identity ID information, timestamp, signal strength, etc. of the RFID tags worn by each pollinating bee.

[0041] Reference Figure 1 And in combination with Figure 2 And Figure 3 , a method for analyzing the foraging behavior of pollinating bees based on RFID data shown in this embodiment specifically includes the following steps:

[0042] Step 1000: Obtain the types of pollinating bees, pollination environment, and RFID scan data of each pollinating bee, where the RFID scan data includes the ID information of the pollinating bee, scan timestamp, and set of reader numbers passed through.

[0043] Step 2000: Set a foraging behavior threshold for each pollinating bee according to the types of pollinating bees, pollination environment, and the preset standard foraging behavior threshold based on historical data. Divide the set of reader numbers into several data segments according to the scan timestamp and the foraging behavior threshold.

[0044] Step 2100: Set a working coefficient and an environmental coefficient respectively according to the types of pollinating bees and the pollination environment.

[0045] Step 2200: After normalizing the working coefficient and the environmental coefficient, set the foraging behavior threshold in combination with the standard foraging behavior threshold.

[0046] In this embodiment, the pollinating bees may include honeybees and bumblebees, where the working coefficient of honeybees is 1, and the working coefficient of bumblebees is 1.5. The pollination environment of the pollinating bees may include greenhouses and fields, where the environmental coefficient of greenhouses is 1, and the environmental coefficient of fields is 2.5. According to the standard collection behavior threshold preset based on historical data, which is 2 minutes, after normalizing the working coefficients of different pollinating bee species and the environmental coefficients of different pollination environments, the collection behavior threshold can be set according to the standard collection behavior threshold.

[0047] Step 2300: Traverse the RFID scan data of each pollinating bee. When the time difference between the scan timestamps of the two card readers is greater than the collection behavior threshold, segment the card reader number set, and divide it into several data segments.

[0048] For example, assume that the number of the first card reader is 1, the number of the second card reader is 2, and the card reader number set passed by a certain pollinating bee is: 111112222211112222211111222222. According to the scan timestamp and the collection behavior threshold, the passed card reader number set is divided into 4 data segments, namely: 1: (1111122222) / 2: (1111) / 3: (2222211111) / 4: (222222).

[0049] Step 3000: Traverse the data segments of each pollinating bee and mark data labels, where the data labels include: potential out-of-nest and potential return-to-nest. The rule for marking data labels is: merge consecutive identical card reader numbers within each data segment, and sequentially determine whether the first two card reader numbers in the merged data segment are the first card reader and the second card reader respectively. If so, mark this data segment as potential out-of-nest, and continue to determine whether the last two card reader numbers in the merged data segment are the second card reader and the first card reader respectively. If so, mark this data segment as potential return-to-nest.

[0050] For example, the card reader number set passed by a certain pollinating bee is: 1: (1111122222) / 2: (1111) / 3: (2222211111) / 4: (111222222111). Merge the consecutive identical card reader numbers within each data segment to get: 1: (12) / 2: (1) / 3: (21) / 4: (121).

[0051] After the first data segment is merged, the first two card reader numbers are: the first card reader and the second card reader, so it is marked as potentially out of the nest. After the second data segment is merged, the card reader number is only the first card reader, and no data label can be marked. After the third data segment is merged, the first two card reader numbers are: the second card reader and the first card reader, so it is marked as potentially returning to the nest. After the fourth data segment is merged, the first two card reader numbers are: the first card reader and the second card reader, so it is marked as potentially out of the nest. At the same time, the last two card reader numbers are: the second card reader and the first card reader, marked as potentially returning to the nest. So we get: 1: (1111122222) [Potentially out of the nest] / 2: (1111) / 3: (2222211111) [Potentially returning to the nest] / 4: (111222222111) [Potentially out of the nest & Potentially returning to the nest],

[0052] Step 4000: Traverse the data segments of each pollinating bee and mark the data types, where the data types include: the first type of data, the second type of data, the third type of data, and the fourth type of data. The rule for marking the data type is: Determine whether the card reader numbers of each data segment are all the first card reader. If so, mark it as the first type of data; if not, continue to determine whether the card reader numbers of each data segment are all the second card reader. If so, mark it as the second type of data. If not, continue to determine whether the last card reader number of each data segment is the first card reader. If so, mark it as the third type of data. If not, mark it as the fourth type of data.

[0053] For example, the set of card reader numbers passed by a certain pollinating bee is: 1: (1111122222) / 2: (1111) / 3: (2222211111) / 4: (222222). The first data segment is marked as the fourth type of data, the second data segment is marked as the first type of data, the third data segment is marked as the third type of data, and the fourth data segment is marked as the second type of data.

[0054] So we get: 1: (1111122222) [Fourth type of data] / 2: (1111)) [First type of data] / 3: (2222211111)) [Third type of data] / 4: (222222)) [Second type of data].

[0055] Step 5000: Read the data segments of each pollinating bee in sequence according to the scan timestamp. Find the data segments that can be marked as out of the nest according to the data type of the data segment and the data type or data label of the next data segment. Then read the next data segment and find the data segments that can be marked as returning to the nest according to the label type or data type of this data segment until all data segments are read.

[0056] When pollinating bees pass through the entrance and exit channels of the beehive, they may crawl repeatedly in the entrance and exit channels or remain stationary under the card reader for a long time, resulting in repeated scanning in a short period. Moreover, a large number of pollinating bees may pass through the entrance and exit channels of a beehive at the same time, and some pollinating bees may not be successfully scanned by the first card reader and / or the second card reader. As a result, some abnormal data will appear. Therefore, when determining whether a data segment can be marked as leaving the nest or returning to the nest, it is necessary to assist in the judgment with the next data segment to improve the utilization rate of the data.

[0057] Step 5000 specifically includes the following steps:

[0058] Step 5100: Read the data segments of each pollinating bee in sequence according to the scanning timestamp, and determine whether the data segment can be marked as leaving the nest based on the data type of the data segment and the data type or data label of the next data segment. If it cannot be marked as leaving the nest, skip this data segment and read the next data segment until a data segment that can be marked as leaving the nest is found.

[0059] Step 5200: Read the next data segment of the data segment marked as leaving the nest, and determine whether it can be marked as returning to the nest based on the label type or data type of this data segment. If it cannot be marked as returning to the nest, skip this data segment and read the next data segment until a data segment that can be marked as returning to the nest is found.

[0060] The rule for determining whether a data segment can be marked as leaving the nest in Step 5000 is:

[0061] If the read data segment is of the first type of data, determine whether the first card reader number of the next data segment is the first card reader or whether it is marked as potentially leaving the nest. If so, skip this data segment and read the next data segment. If not, mark it as leaving the nest;

[0062] If the read data segment is of the second type of data, determine whether the next data segment is marked as potentially leaving the nest. If so, skip this data segment and read the next data segment. If not, mark it as leaving the nest;

[0063] If the read data segment is of the third type of data, determine whether the first card reader number of the next data segment is the first card reader or whether it is marked as potentially leaving the nest. If so, skip this data segment and read the next data segment. If not, mark it as leaving the nest;

[0064] If the read data segment is of the fourth type of data, mark it as leaving the nest.

[0065] The rule for determining whether a data segment can be marked as returning to the nest in Step 5000 is:

[0066] If the data segment marked as out-of-nest is marked as the first type of data, read the next data segment and determine whether this data segment is marked as potential in-nest. If not, read the next data segment and re-determine. If so, mark it as in-nest;

[0067] If the data segment marked as out-of-nest is marked as the second type of data, read the next data segment and determine whether this data segment is not marked as the second type of data or is marked as potential in-nest. If not, read the next data segment and re-determine. If so, mark it as in-nest;

[0068] If the data segment marked as out-of-nest is marked as the third type of data, read the next data segment and determine whether this data segment is not marked as the second type of data. If not, read the next data segment and re-determine. If so, mark it as in-nest;

[0069] If the data segment marked as out-of-nest is marked as the fourth type of data, read the next data segment and mark it as in-nest.

[0070] To improve the accuracy of data segment determination, after the data segment is marked as in-nest, a secondary determination and correction are still required. Therefore, step 5000 further includes:

[0071] Step 5300: Determine whether the data segment marked as in-nest is marked as potential out-of-nest. If it is marked as potential out-of-nest, determine whether the number of data segments between the data segment marked as out-of-nest and the corresponding data segment marked as in-nest is less than 3, and enter step 5310; if it is not marked as potential out-of-nest, enter step 5320.

[0072] Step 5310: If the number of data segments between the data segment marked as out-of-nest and the corresponding data segment marked as in-nest is less than 3, the data segment marked as in-nest is accurate. If it is greater than or equal to 3, correct the previous data segment of the data segment marked as in-nest to in-nest.

[0073] Step 5320: Read the next data segment of the data segment marked as in-nest and determine whether this data segment is marked as potential in-nest. If it is not marked as potential in-nest, the data segment marked as in-nest is accurate;

[0074] If it is marked as potential in-nest, determine whether the number of data segments between the data segment marked as out-of-nest and the corresponding data segment marked as in-nest is greater than or equal to 3. If it is greater than or equal to 3, correct the previous data segment of the data segment marked as in-nest to in-nest. If it is less than 3, the data segment marked as in-nest is accurate.

[0075] The reader number set of a certain pollinating bee is: 111111111222211111222221111111.

[0076] Divide the data segments according to the scanning timestamp and the acquisition behavior threshold, and get: 1: (1111111112222) / 2: (11111) / 3: (222221111111),

[0077] Mark data labels and data types for each data segment, and get: 1: (1111111112222) [Potential out-of-nest & Fourth type of data] / 2: (11111) [First type of data] / 3: (222221111111) [Potential in-nest & Third type of data].

[0078] Read the first data segment. The first data segment is marked as the fourth type of data and marked as out-of-nest.

[0079] The first data segment is marked as out-of-nest. Then start looking for the corresponding in-nest behavior from the second data segment. Since the first data segment is marked as the fourth type of data, the second scanned data segment is determined as in-nest.

[0080] Next, perform a secondary judgment and correction on the in-nest. The second data segment is not marked as potential out-of-nest, so read the next data segment. The third data segment is marked as potential in-nest. There are a total of 2 data segments between the first data segment and the second data segment, which is less than 3. So the second data segment is correctly marked as in-nest.

[0081] The reader scan result of another pollinating bee is: 11111111111122211111111221111111111222221111111111111.

[0082] Divide it into 6 data segments according to the timestamp and the acquisition behavior threshold, and get: 1: (111111111) / 2: (11122211) / 3: (11111122111) / 4: (1111111) / 5: (22222111111) / 6: (1111111).

[0083] Mark data labels and data types for each data segment, and get: 1: (111111111) [First type of data] / 2: (11122211) [Potential out-of-nest & Potential in-nest & Third type of data] / 3: (11111122111) [Potential out-of-nest & Potential in-nest & Third type of data] / 4: (1111111) [First type of data] / 5: (22222111111) [Potential in-nest & Third type of data] / 6: (1111111) [First type of data].

[0084] Read the first data segment. The first data segment is of the first type. Read the next data segment. The first card reader number of the second data segment is the first card reader and it is marked as potentially out of the nest, so skip the first data segment.

[0085] Read the second data segment. The second data segment is of the third type. Read the next data segment. The first card reader number of the third data segment is the first card reader and it is marked as potentially out of the nest, so skip the second data segment.

[0086] Read the third data segment. The third data segment is of the third type. Read the next data segment. The first card reader number of the fourth data segment is the first card reader, so skip the third data segment.

[0087] Read the fourth data segment. The fourth data segment is of the first type. Read the next data segment. The first card reader number of the fifth data segment is the second card reader and it is not marked as potentially out of the nest, so mark the fourth data segment as out of the nest.

[0088] The fourth data segment is marked as out of the nest. Look for the corresponding return to the nest from the fifth data segment. Read the fifth data segment. The fourth data segment is marked as of the first type and the fifth data segment is marked as potentially returning to the nest, which meets the requirements, so mark the fifth data segment as a return to the nest behavior.

[0089] Next, perform a secondary judgment and correction on the return to the nest. The fifth data segment is not marked as potentially out of the nest, so read the next data segment. The sixth data segment is not marked as potentially returning to the nest, so mark the fifth data segment as correctly returned to the nest.

[0090] Read the sixth data segment. Since the sixth data segment is the last data segment, the loop ends.

[0091] It should be understood that the above specific embodiments of the present invention are only used for exemplary illustration or explanation of the principle of the present invention, and do not constitute a limitation to the present invention. Therefore, any modifications, equivalent replacements, improvements, etc. made without departing from the spirit and scope of the present invention shall be included within the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modification examples falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A method for analyzing pollinating bee collection behavior based on RFID data, characterized in that: The steps include: Step 1000: Acquire the pollinating bee species, pollination environment, and RFID scanning data of each pollinating bee, wherein the RFID scanning data includes the ID information of the pollinating bee, the scanning timestamp, and the number set of the card reader passed; Step 2000: setting a collection behavior threshold for each of the pollinating bees according to the pollinating bee species, the pollination environment, and a standard collection behavior threshold preset according to historical data, and dividing the card reader number set into a plurality of data segments according to the scanning timestamp and the collection behavior threshold; Step 3000: traverse the data segment of each pollinating bee and mark the data label, wherein the data label includes: potential leaving the nest and potential returning to the nest; Step 4000: traverse the data segment of each pollinating bee and mark the data type, wherein the data type includes: first-category data, second-category data, third-category data and fourth-category data, and the rule for marking the data type is: determine whether the card reader number of each data segment is all first-category data, if yes, mark it as the first-category data; if not, continue to determine whether the card reader number of each data segment is all second-category data, if yes, mark it as the second-category data, if not, continue to determine whether the last card reader number of each data segment is the first card reader, if yes, mark it as the third-category data, if not, mark it as the fourth-category data; Step 5000: Read the data segment of each pollinating bee in sequence according to the scanning timestamp, and search for a data segment that can be marked as leaving the nest according to the data type of the data segment and the data type or data label of the next data segment. Then read the next data segment, and search for the data segment that can be marked as returning to the nest according to the label type or data type of the data segment, until all the data segments are read.

2. The method for analyzing pollinating bee collection behavior based on RFID data according to claim 1, characterized in that: The step 2000 includes the following steps: Step 2100: setting a working coefficient and an environmental coefficient according to the pollinating bee species and the pollination environment; Step 2200: After normalizing the working coefficient and the environmental coefficient, the collection behavior threshold is set in combination with the standard collection behavior threshold.

3. The method for analyzing pollinating bee collection behavior based on RFID data according to claim 1, characterized in that: The rule for determining whether the data segment can be marked as the out-of-nest in step 5000 is: If the data segment read is the first type of data, determine whether the first card reader number of the next data segment is the first card reader, or whether it is marked as the potential out-of-nest, if yes, skip the data segment and read the next data segment, if not, mark it as the out-of-nest; If the data segment read is the second type of data, determining whether the next data segment is marked as the potential out-of-nest, if yes, skipping the data segment and reading the next data segment, if no, marking it as the potential out-of-nest; If the data segment read is the third type of data, determining whether the first card reader number of the next data segment is the first card reader, or whether it is marked as the potential nest exit; if so, skipping the data segment and reading the next data segment; if not, marking it as the nest exit; If the read data segment is the fourth type of data, it is marked as the nest exit.

4. The method for analyzing pollinating bee collection behavior based on RFID data according to claim 3, characterized in that: The rule for determining whether the data segment can be marked as the homing in step 5000 is: If the data segment marked as the out-of-nest is marked as the first type of data, then read the next data segment and determine whether the data segment is marked as the potential back-to-nest; if not, then read the next data segment and re-determine; if yes, then mark it as the back-to-nest; If the data segment marked as the out-of-nest is marked as the second type of data, then read the next data segment, and determine whether the data segment is not marked as the second type of data, or whether it is marked as the potential back-to-nest; if not, read the next data segment and re-determine; if yes, mark it as the back-to-nest; If the data segment marked as the out-of-nest is marked as the third category of data, read the next data segment and determine whether the data segment is not marked as the second category of data; if not, read the next data segment and re-determine; if yes, mark it as the back-to-nest; If the data segment marked as the out-of-nest is marked as the fourth type of data, the next data segment is read and marked as the in-nest.

5. The method for analyzing pollinating bee collection behavior based on RFID data according to claim 4, characterized in that: The step 5000 includes the following steps: Step 5100: Read the data segment of each pollinating bee in sequence according to the scanning timestamp, and determine whether the data segment can be marked as leaving the nest based on the data type of the data segment and the data type or data label of the next data segment. If it cannot be marked as leaving the nest, skip the data segment and read the next data segment until the data segment that can be marked as leaving the nest is found.

6. The method for analyzing pollinating bee collection behavior based on RFID data according to claim 5, characterized in that: The step 5000 also includes the following steps: Step 5200: Read the next data segment of the data segment marked as the out-of-nest, and determine whether it can be marked as the back-in-nest according to the tag type or the data type of the data segment. If it cannot be marked as the back-in-nest, skip the data segment and read the next data segment until the data segment that can be marked as the back-in-nest is found.

7. The method for analyzing pollinating bee collection behavior based on RFID data according to claim 6, characterized in that: The step 5000 also includes the following steps: Step 5300: Determine whether the data segment marked as the return to the nest is marked as the potential out-of-nest. If it is marked as the potential out-of-nest, determine whether the number of data segments included between the data segment marked as the out-of-nest and the corresponding data segment marked as the return to the nest is less than 3, and proceed to step 5310; If it is not marked as the potential nest exit, go to step 5320; Step 5310: If the number of data segments included between the data segment marked as the out-of-nest and the corresponding data segment marked as the back-to-nest is less than 3, the data segment marked as the back-to-nest is accurate; if it is greater than or equal to 3, the previous data segment of the data segment marked as the back-to-nest is corrected to be the back-to-nest; Step 5320: read the next data segment of the data segment marked as the return-to-nest, and determine whether the data segment is marked as the potential return-to-nest; if it is not marked as the potential return-to-nest, the data segment marked as the return-to-nest is accurate; If it is marked as the potential return to the nest, determine whether the number of data segments included between the data segment marked as the out-of-nest and the corresponding data segment marked as the return to the nest is greater than or equal to 3. If it is greater than or equal to 3, the previous data segment of the data segment marked as the return to the nest is corrected to the return to the nest. If it is less than 3, the data segment marked as the return to the nest is accurate.

Images