A method and apparatus for monitoring emergence of fruit fly insects

Abstract

The application discloses a method and device for monitoring the emergence of real fly insects, and the method comprises the following steps: acquiring the light intensity in the adult insect warehouse in response to the infrared signal transmitted by the infrared sensor; issuing a first instruction, which is used for controlling the camera to shoot a first image of the plane where the emergence warehouse and the adult insect warehouse are connected, and recording the first issuing time of the first instruction, wherein the light compensation lamp is turned on when the light intensity is less than the preset light intensity threshold; performing image recognition on the first image to obtain the first number of real fly insects on the plane where the emergence warehouse and the adult insect warehouse are connected and the pre-flap characteristics; determining the emergence cycle of the real fly insects and calculating the difference between the first number and the second number; determining the current emergence condition of the real fly insects according to the difference and the emergence cycle, wherein the current emergence condition comprises the emergence rate and the emergence rate. The application can effectively improve the accuracy of monitoring the emergence stage of real fly insects.

Description

Technical Field

[0001] This application relates to the field of insect trapping and monitoring technology, and in particular to a method and device for monitoring the emergence of fruit flies. Background Technology

[0002] The oriental fruit fly, commonly known as the citrus maggot, is a global plant quarantine pest and one of the most serious pests affecting citrus fruits. Because the eggs, larvae, and pupae of the oriental fruit fly are relatively hidden and difficult to detect, effective control is challenging, causing economic losses to fruit growers. Currently, effective control is mainly achieved by spraying pesticides during the oriental fruit fly's emergence stage. Therefore, monitoring the oriental fruit fly's emergence stage is crucial.

[0003] To determine whether the oriental fruit fly is in the molting stage, the primary method currently used is to set up oriental fruit fly traps in the monitoring area. The presence of molting flies in the traps is then used to determine the overall molting stage of the oriental fruit fly population within the monitored area. Existing traps typically contain a light source or food to attract and capture the flies. Using these traps, monitoring personnel can periodically check the number and condition of the captured flies to determine if they are in the molting stage. The captured adults can then be further analyzed for their physiological state, such as whether their wings are fully extended and whether their body color is mature, thus determining the molting stage and calculating the molting progress. This monitoring method is crucial for developing effective oriental fruit fly control measures and determining the timing of pesticide spraying.

[0004] However, in production, monitoring of fruit flies is often done manually by burying pupae, which not only consumes a lot of manpower, but also often leads to missing the critical period of emergence due to untimely observation, affecting the accuracy of monitoring and forecasting, and thus affecting the formulation of control measures and the best time for control. Summary of the Invention

[0005] This application provides a method and apparatus for monitoring the emergence of fruit flies, which can effectively improve the accuracy of monitoring the emergence of fruit flies.

[0006] To achieve the above objectives, the embodiments of this application adopt the following technical solutions:

[0007] Firstly, a method for monitoring the emergence of fruit flies is provided, applied to a fruit fly monitoring device. The monitoring device includes a housing, an emergence chamber, an adult chamber, and a hollow frame. The emergence chamber and the adult chamber are located within the housing. At least one infrared sensor is installed at the outlet of the emergence chamber and at the entrance of the adult chamber. The emergence chamber is entirely situated in a soil environment, while the adult chamber is entirely situated on the ground. The adult chamber has a structure that is narrower at the top and wider at the bottom, and it contains a camera and supplementary lighting for monitoring fruit flies. The top of the adult chamber is seamlessly connected to the hollow frame, and at least one counting sensor is installed within the hollow frame. The method includes:

[0008] The light intensity inside the adult insect chamber is obtained in response to the infrared signal transmitted by the infrared sensor.

[0009] A first command is issued, which controls the camera to capture a first image of the plane where the molting chamber and the adult chamber are connected, and records the first time the first command is issued. The timing starts from the first time the command is issued, wherein the supplementary light is turned on when the light intensity is less than a preset light intensity threshold.

[0010] Image recognition is performed on the first image to obtain the first number and pre-wing-spreading characteristics of fruit flies in the plane where the molting chamber and the adult chamber are connected.

[0011] Determine the emergence cycle of fruit flies and calculate the difference between the first quantity and the second quantity;

[0012] The current emergence status of fruit flies is determined based on the difference and the emergence cycle, wherein the current emergence status includes the emergence rate and the emergence speed.

[0013] In one possible implementation of the first aspect, determining the emergence cycle of fruit flies includes:

[0014] After the first emission moment, the camera is controlled to capture an image once at preset time intervals, and image recognition is performed on each captured image to obtain the number of fruit flies.

[0015] The image corresponding to the number of fruit flies is less than a preset threshold is taken as the final image, and the time of the final image is obtained.

[0016] The difference between the shooting time and the first emission time is taken as the emergence period of the fruit fly.

[0017] In one possible implementation of the first aspect, the adult chamber is further equipped with a photoelectric sensor, which, before determining the emergence cycle of the fruit fly, includes:

[0018] In response to receiving photoelectric data transmitted by the photoelectric sensor, the full wing spread characteristics of the fruit fly insects in the adult insect chamber are determined based on the photoelectric data;

[0019] A second command is issued, which controls the camera to capture a second image of the plane where the connection between the molting chamber and the adult insect chamber is located;

[0020] Image recognition is performed on the second image to obtain a second number of fruit flies on the plane where the connection between the emergence chamber and the adult chamber is located;

[0021] Acquire current environmental data, including temperature data, humidity data, and sunshine duration;

[0022] The maturity of fruit flies in the adult chamber is determined by the current environmental data and the fully spread wing characteristics.

[0023] The first quantity, the second quantity, and the current environmental data are input into a pre-trained prediction model to predict the future peak emergence time period for fruit flies.

[0024] In one possible implementation of the first aspect, the adult insect chamber is equipped with a tunable light source, and the photoelectric data includes the light intensity within the adult insect chamber and light wave reflection data from the wing surface of the fruit fly insect under the tunable light source. Determining the full wing spread characteristics of the fruit fly insect within the adult insect chamber based on the photoelectric data includes:

[0025] When the light intensity in the adult insect chamber is greater than a preset first light intensity threshold, the wavelength range of the tunable light source is determined as the target wavelength range, and the wavelength of the tunable light source is adjusted to the target wavelength range.

[0026] The spectral feature data of the wings of the giant fruit fly at different growth stages are obtained by using a preset spectral feature database. The light wave reflection data is compared with the spectral feature data to obtain the growth stage corresponding to the light wave reflection data and the number of fruit flies corresponding to each growth stage.

[0027] Based on the growth stage corresponding to the light wave reflection data, the fully spread wing characteristics of the fruit fly insects in the adult insect chamber are obtained by querying the preset feature database.

[0028] In one possible implementation of the first aspect, the step of comparing the light wave reflection data with the spectral feature data to obtain the growth stage corresponding to the light wave reflection data, and obtaining the number of fruit flies corresponding to each growth stage, includes:

[0029] Spectral analysis is performed on the light wave reflection data to obtain the first spectral bandwidth, first peak position, and first peak intensity of the light wave reflection data;

[0030] Obtain the second spectral bandwidth, second peak position, and second peak intensity of the spectral feature data;

[0031] According to a preset feature matching algorithm, the first matching degree between the first spectral bandwidth and the second spectral bandwidth, the second matching degree between the first peak position and the second peak position, and the third matching degree between the first peak intensity and the second peak intensity are calculated.

[0032] The sum of the first matching degree, the second matching degree, and the third matching degree is taken as the target matching degree between the light wave reflection data and the spectral feature data;

[0033] If the target matching degree is greater than a preset matching degree threshold, it is determined that the light wave reflection data matches the spectral feature data, and the growth stage corresponding to the spectral feature data is taken as the growth stage corresponding to the light wave reflection data that matches it.

[0034] The number of fruit flies corresponding to each growth stage was statistically determined.

[0035] In one possible implementation of the first aspect, the step of performing image recognition on the first image to obtain the first number and pre-wing-spreading characteristics of fruit flies on the plane where the molting chamber and the adult chamber are located includes:

[0036] The first image is preprocessed to obtain multiple image features;

[0037] The first number and pre-wing-spreading features of fruit flies are determined from the plane where the molting chamber and the adult chamber are located, based on a plurality of image features.

[0038] In one possible implementation of the first aspect, after performing image recognition on the first image to obtain the first number and pre-wing-spreading characteristics of fruit flies on the plane where the molting chamber and the adult chamber are connected, the method includes:

[0039] The key features are obtained by screening the pre-wing-spreading features;

[0040] Based on preset mapping rules, the key features are mapped to obtain corresponding key feature interpretations;

[0041] Based on the interpretation of the key features, determine the scope of keywords in the preset feature database;

[0042] Within the keyword range, identify and output the keywords corresponding to the key feature definitions.

[0043] In one possible implementation of the first aspect, the method further includes:

[0044] If the key features cannot be mapped to obtain corresponding key feature interpretations, image recognition is performed on the first image to obtain multiple color features;

[0045] The cross-shaped black horizontal stripe feature and the golden yellow body color feature were selected from multiple color features, and the cross-shaped black horizontal stripe feature and the golden yellow body color feature were used as the second target key features.

[0046] Obtain the associated interpretations of the key feature interpretations, and expand the keyword range based on the associated interpretations to obtain a second keyword range;

[0047] Within the second keyword range, the keywords corresponding to the second target key features are determined and output.

[0048] In one possible implementation of the first aspect, before determining the keyword range in a preset feature database based on the interpretation of the key features, the following is included:

[0049] The characteristics of each growth stage of the giant fruit fly are stored in the preset feature database, and the search range of the preset feature database is set.

[0050] The step of determining and outputting the keywords corresponding to the key feature definitions within the keyword range includes:

[0051] Within the search range, the keyword is located, and the growth stage feature corresponding to the keyword is determined, wherein the growth stage feature corresponding to the keyword is used to characterize the current growth stage of the giant water fly in the monitoring area.

[0052] Secondly, this application provides a fruit fly monitoring device, which includes a housing, an emergence chamber, an adult chamber, and a hollow frame. The emergence chamber and the adult chamber are located within the housing. At least one infrared sensor is installed at the outlet of the emergence chamber and at the entrance of the adult chamber. The emergence chamber is entirely situated in the soil environment, while the adult chamber is entirely situated on the ground. The adult chamber has a structure that is narrower at the top and wider at the bottom, and it contains a camera and a supplementary light for monitoring fruit flies. The top of the adult chamber is seamlessly connected to the hollow frame, and at least one counting sensor is installed within the hollow frame. The fruit fly monitoring device further includes:

[0053] The memory is configured to store instructions; and

[0054] The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the above-described method for monitoring the emergence of fruit flies.

[0055] Through the above technical solution, an infrared sensor is installed at the connection between the emergence chamber and the adult chamber. When the infrared sensor detects movement of fruit flies, it can automatically issue a command to control the camera to capture images and record the capture time. This effectively reduces the delay of human operation, improves the timeliness of monitoring, and can detect the movement of fruit flies in real time, reducing the time interval of periodic human checks and ensuring that the critical moment of fruit fly emergence is not missed. The camera is installed inside the adult chamber to capture images of the plane at the connection between the emergence chamber and the adult chamber. By recognizing the captured images, the number of fruit flies and their pre-wing-spreading characteristics can be obtained, effectively improving the accuracy of data collection, reducing subjective errors in human judgment, and thus improving the accuracy of monitoring. The supplementary light automatically adjusts according to the light intensity to ensure clear images even in low-light conditions. When the infrared sensor detects movement of fruit flies, it automatically obtains the light intensity inside the adult chamber and issues the first command to control the camera to capture images of the connection between the emergence chamber and the adult chamber and record the capture time. Subsequently, the captured images are recognized to obtain the number of fruit flies and their pre-wing-spreading characteristics. By combining historical and current data, the emergence cycle of fruit flies is determined, and the difference between the current and historical populations is calculated to assess the emergence status. Based on the difference and the emergence cycle, the emergence rate and emergence speed are calculated, providing data support for control measures. Automated monitoring, intelligent control, and data analysis significantly improve the accuracy of monitoring the fruit fly emergence stage, reduce the time interval of manual monitoring, ensure that critical emergence opportunities are not missed, and reduce the risk of missed detections. The application of automatic adjustment of supplemental lighting and image recognition technology improves the quality of image acquisition and the accuracy of recognition, providing a scientific basis for developing effective control measures and determining the timing of pesticide spraying.

[0056] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0057] Figure 1 A flowchart illustrating a method for monitoring the emergence of fruit flies, as provided in an embodiment of this application;

[0058] Figure 2 This is a schematic diagram of the overall structure of a fruit fly monitoring device provided in an embodiment of this application;

[0059] Figure 3 A front view of a fruit fly monitoring device provided in an embodiment of this application;

[0060] Figure 4 A rear view of a fruit fly monitoring device provided in an embodiment of this application;

[0061] Figure 5 This is a partial structural schematic diagram of a fruit fly monitoring device provided in an embodiment of this application.

[0062] Explanation of reference numerals in the attached figures

[0063] 10 Fruit fly insect monitoring device box 20 Adult insect storage 30 Feathering Chamber 40 Counting sensor 50 Camera 60 Fill light Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0065] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0066] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0067] Figure 1 The illustration schematically shows a flowchart of a method for monitoring the emergence of fruit flies according to an embodiment of this application. Figure 1As shown in the figure, this application provides a method for monitoring the emergence of fruit flies, which is applied to a fruit fly monitoring device. The fruit fly monitoring device includes a box, an emergence chamber, an adult chamber, and a hollow frame. The emergence chamber and the adult chamber are located inside the box. At least one infrared sensor is provided at the outlet of the emergence chamber and the entrance of the adult chamber. The emergence chamber is located in the soil environment, and the adult chamber is located on the ground. The adult chamber has a structure that is narrow at the top and wide at the bottom. The adult chamber is equipped with a camera and a supplementary light for monitoring fruit flies. The top of the adult chamber is seamlessly connected to the hollow frame. At least one counting sensor is provided inside the hollow frame. The method may include the following steps.

[0068] S110: In response to the infrared signal transmitted by the infrared sensor, the light intensity inside the adult insect chamber is obtained;

[0069] S120. Issue a first command. The first command is used to control the camera to capture a first image of the plane where the molting chamber and the adult chamber are connected, and record the first time the first command is issued. Start timing from the first time the command is issued. When the light intensity is less than the preset light intensity threshold, control the supplementary light to turn on.

[0070] S130. Perform image recognition on the first image to obtain the first number and pre-wing-spreading characteristics of fruit fly insects on the plane where the molting chamber and adult chamber are connected.

[0071] S140. Determine the emergence cycle of fruit flies and calculate the difference between the first and second numbers;

[0072] S150. Determine the current emergence status of fruit flies based on the difference and emergence cycle. The current emergence status includes the emergence rate and emergence speed.

[0073] In this embodiment, fruit flies can be trapped into the emergence chamber, or the fruit flies can be monitored by placing citrus fruits containing fly eggs into the emergence chamber.

[0074] In this embodiment, the fruit fly monitoring device can be applied to fruit flies. The following description uses the large fruit fly as an example. Figure 2 and Figure 3The fruit fly monitoring device housing 10 includes a molting chamber 30, an adult chamber 20, and a camera 50. The molting chamber 30, adult chamber 20, and camera 50 are all located within the monitoring area of ​​the fruit fly monitoring device housing 10, which is used to monitor the growth activities of fruit flies. The adult chamber 20 is located above the molting chamber 30. At least one infrared sensor is installed at both the outlet of the molting chamber and the entrance of the adult chamber. Due to the biological characteristics of fruit flies, the adult chamber 20 is designed with a narrow top and wide bottom structure to allow fruit flies to fly from the bottom to the top of the adult chamber 20, facilitating better monitoring. The adult chamber 20 contains the camera 50, a photoelectric sensor, and a counting sensor 40. To better simulate the natural environment, transparent observation panels are installed around the sides and top of the fruit fly monitoring device housing 10. Figure 4 As shown, the enclosure is surrounded by transparent observation panels, and the material of these panels is not limited in this application.

[0075] In practice, the emergence chamber 30 is buried in the soil to simulate the natural habitat of the oriental fruit fly, ensuring that the oriental fruit fly's environment within the monitoring area is consistent with the external environment. The oriental fruit fly monitoring device is constructed with [materials such as...] around its perimeter and above. Figure 4 The transparent monitoring panel shown is used to ensure suitable lighting and temperature within the monitoring area, facilitating observation of the area. An infrared sensor is used to determine the presence of a large fruit fly's infrared signal. If a large fruit fly's infrared signal is received within the monitoring area, the camera 50 above the device captures an image, which is automatically transmitted back and stored. Based on the biological characteristics of the large fruit fly, it will gather upwards along the cone-shaped structure of the adult chamber, gradually converging at the top of the cone. The counting sensor 40 above the cone counts the large fruit flies and stores the data using a storage device.

[0076] Specifically, Figure 5 This illustration shows a partial structural diagram of a fruit fly monitoring device provided in an embodiment of this application, as shown below. Figure 5 As shown, camera 50 is fixed inside the adult insect chamber to capture the first image of the plane where the molting chamber and the adult insect chamber are connected. After image recognition and analysis of the first image, the number of fruit flies after emerging from the molting chamber and their characteristics before wing spread can be obtained. By adjusting the light intensity in the monitoring area, the data of the reflected light waves from the wings of fruit flies in the monitoring area can be obtained through photoelectric sensors to determine the growth stage of the fruit flies.

[0077] When the infrared sensors in the emergence chamber or adult chamber detect the activity of fruit flies, they immediately trigger an infrared signal, which is transmitted to a controller. The controller can be a processor, server terminal, or other intelligent terminal. Simultaneously, the controller acquires the current light intensity within the adult chamber. This non-contact monitoring using infrared technology effectively avoids interference with the natural behavior of the fruit flies. In practice, if the light intensity is below a preset threshold, such as below 200 lux, the supplementary lighting 60 is automatically activated to ensure that the camera 50 can clearly capture the image, providing a high-quality data foundation for subsequent image analysis.

[0078] Next, the first command is issued, controlling camera 50 to capture the first image of the plane connecting the emergence chamber and the adult chamber, and recording the moment the command is first issued. The triggering of the first command is based on feedback from the infrared sensor, meaning that the image acquisition process will only begin when fruit fly activity is detected, thus effectively saving storage resources and processing time. In actual operation, camera 50 is configured in high-definition mode, recording video at 30 frames per second to ensure that every detail of the fruit fly emergence is captured. Simultaneously, if the lighting is insufficient, supplementary lighting 60 automatically illuminates to assist in filming.

[0079] Image analysis of the first acquired image can be performed using image recognition algorithms to automatically identify and count the number of fruit flies at the connection between the emergence chamber and the adult chamber. Simultaneously, it can identify specific morphological characteristics of the fruit fly before wing expansion, such as the presence of the pupal shell and incomplete wing development. This can be achieved through a pre-trained machine learning model, which can learn from a large number of samples and accurately distinguish different life stages and behavioral states of the fruit fly. In practice, the image recognition algorithm traverses every pixel in the image, searching for pixel clusters that match the characteristics of the fruit fly, thereby completing the counting and feature extraction.

[0080] The emergence cycle can be obtained through historical experience. By comparing the changes in the number of fruit flies in images taken at different time points, the difference between the first number (the number at the initial monitoring) and the second number in subsequent monitoring can be calculated to track the growth or decline of the fruit fly population within a specific time period, providing a quantitative basis for assessing emergence efficiency.

[0081] Finally, based on the difference in fruit fly population size and the known emergence cycle, the emergence rate and emergence speed were calculated to reflect the reproductive capacity and emergence speed of the fruit fly population. The emergence rate refers to the proportion of individuals successfully emerging as adults out of the total number of pupae, while the emergence speed describes the number of adults emerging per unit time. By analyzing the emergence rate and emergence speed, the impact of environmental factors and food sources on fruit fly emergence can be further analyzed, providing precise guidance for agricultural pest control. In summary, this fruit fly emergence monitoring method, through intelligent and automated means, achieves comprehensive and high-precision monitoring of the fruit fly emergence process, not only improving monitoring efficiency but also providing strong technical support for fruit fly ecological research and pest management practices.

[0082] Through the above technical solution, an infrared sensor is installed at the connection between the emergence chamber and the adult chamber. When the infrared sensor detects movement of fruit flies, it can automatically issue a command to control the camera to capture images and record the capture time. This effectively reduces the delay of human operation, improves the timeliness of monitoring, and can detect the movement of fruit flies in real time, reducing the time interval of periodic human checks and ensuring that the critical moment of fruit fly emergence is not missed. The camera is installed inside the adult chamber to capture images of the plane at the connection between the emergence chamber and the adult chamber. By recognizing the captured images, the number of fruit flies and their pre-wing-spreading characteristics can be obtained, effectively improving the accuracy of data collection, reducing subjective errors in human judgment, and thus improving the accuracy of monitoring. The supplementary light automatically adjusts according to the light intensity to ensure clear image capture even in low-light conditions. The supplementary light can be a tunable light source. When the infrared sensor detects movement of fruit flies, it automatically obtains the light intensity inside the adult chamber and issues the first command to control the camera to capture images of the connection between the emergence chamber and the adult chamber and record the capture time. Subsequently, the captured images are recognized to obtain the number of fruit flies and their pre-wing-spreading characteristics. By combining historical and current data, the emergence cycle of fruit flies is determined, and the difference between the current and historical populations is calculated to assess the emergence status. Based on the difference and the emergence cycle, the emergence rate and emergence speed are calculated, providing data support for control measures. Automated monitoring, intelligent control, and data analysis significantly improve the accuracy of monitoring the fruit fly emergence stage, reduce the time interval of manual monitoring, ensure that critical emergence opportunities are not missed, and reduce the risk of missed detections. The application of automatic adjustment of supplemental lighting and image recognition technology improves the quality of image acquisition and the accuracy of recognition, providing a scientific basis for developing effective control measures and determining the timing of pesticide spraying.

[0083] In one embodiment of this invention, determining the emergence cycle of fruit flies includes the following steps:

[0084] S210. After the first emission moment, the camera is controlled to capture an image once at preset time intervals, and image recognition is performed on each captured image to obtain the number of fruit flies.

[0085] S220. Take the image corresponding to the number of fruit flies less than a preset number threshold as the final image, and obtain the shooting time of the final image;

[0086] S230. The difference between the shooting time and the first emission time is taken as the emergence cycle of fruit flies.

[0087] Upon the issuance of the first command, the camera begins periodically capturing images. During this process, the camera can be automatically triggered to capture images of fruit flies at the connection between the emergence chamber and the adult chamber at preset time intervals, such as every hour or every two hours. After each capture, an image recognition algorithm is activated to accurately count the number of fruit flies in the image. In practice, the image recognition algorithm analyzes every pixel in the image, accurately identifying and counting fruit flies from a complex background by comparing their morphological characteristics, such as body size and color. Finally, a series of time-series data reflecting changes in the number of fruit flies is obtained, enabling continuous and dynamic monitoring of the fruit fly emergence process.

[0088] By analyzing time-series data, when the number of fruit flies first drops below a preset threshold, the image corresponding to that moment is marked as the final image, and the time of its capture is recorded. For example, it can be set to 10% of the initial number to capture the point when the emergence process essentially ends. This embodiment can identify the end of the emergence cycle by a significant decrease in the number of flies, thereby avoiding misjudgments caused by individual differences or accidental factors.

[0089] The emergence cycle of fruit flies can be obtained by calculating the time difference between the time the final image was captured and the time the first command was issued. For example, if the first command was issued at 9:00 AM on June 1st, and the final image was captured at 10:00 AM on June 8th, then the emergence cycle is 7 days plus 1 hour.

[0090] This embodiment, through periodic photography, quantity monitoring, and time difference calculation, can accurately and efficiently determine the emergence cycle of fruit flies. This not only greatly reduces the burden of manual monitoring but also provides a direct quantitative indicator of fruit fly emergence time, contributing to a deeper understanding of the biological characteristics of fruit flies and providing a scientific basis for developing targeted control measures, thus improving the accuracy and reliability of the data.

[0091] In one embodiment of this invention, the adult insect chamber is further equipped with a photoelectric sensor, which includes the following steps before determining the emergence cycle of fruit flies:

[0092] S310. In response to receiving photoelectric data transmitted by the photoelectric sensor, determine the full wing spread characteristics of fruit fly insects in the adult insect chamber based on the photoelectric data.

[0093] S320, Issue a second command, which is used to control the camera to capture a second image of the plane where the connection between the molting chamber and the adult chamber is located;

[0094] S330. Perform image recognition on the second image to obtain the second number of fruit fly insects on the plane where the molting chamber and the adult chamber are connected.

[0095] S340. Obtain current environmental data, including temperature data, humidity data, and sunshine duration;

[0096] S350. Determine the maturity of fruit flies in the adult chamber based on current environmental data and full wing spread characteristics.

[0097] S360. Input the first quantity, the second quantity, and the current environmental data into the pre-trained prediction model to predict the future peak emergence time of fruit flies.

[0098] In this embodiment, the photoelectric sensor can accurately sense the intensity of light and the movement of objects. When a fruit fly completes its molting and spreads its wings, the reflected light from its body shape and wings triggers the sensor, generating specific photoelectric data. By analyzing this photoelectric data and comparing it with a preset model of fully spread wings, such as the angle and area changes of wing spread, the sensor can accurately identify whether the fruit fly has reached a fully spread wing state.

[0099] A second command is issued to drive the camera to capture real-time images of the connection between the emergence chamber and the adult chamber, ensuring that the timing of image capture closely matches the emergence state of the fruit flies. The footage captured by the camera includes the crucial moment when the fruit flies move from the emergence chamber into the adult chamber, providing intuitive visual evidence for subsequent population statistics and prediction of peak emergence times.

[0100] Image recognition is performed on the captured second image to obtain the number of fruit flies in the second image, i.e., the second quantity. Then, current environmental data is acquired, including temperature, humidity, and duration of sunshine.

[0101] Next, by combining the full wing spread characteristics of fruit flies with current environmental data, we can further analyze their maturity. By constructing a mathematical model that comprehensively considers the physiological state and external environment of fruit flies, we can infer the time required for fruit flies to reach a reproductive state, i.e., maturity. This helps to more accurately predict the peak emergence period and population dynamics of fruit flies.

[0102] Specifically, the maturity of fruit flies in the adult chamber can be determined by constructing a nonlinear regression model, based on current environmental data and full wing spread characteristics.

[0103] First, assume that the maturity (M) of fruit flies is a function of factors such as age (A), temperature (T), and humidity (H):

[0104] M = f(A, T, H, ...);

[0105] Here, f() is a nonlinear function. Using current environmental data (temperature, humidity, food supply, light conditions, etc.) and full wing spread characteristics (age, wing extent, etc.), the parameters in the nonlinear regression model can be estimated using least squares, maximum likelihood estimation, or other optimization algorithms. Cross-validation is used to evaluate the model's predictive performance, and the model structure or parameters are adjusted based on the validation results to improve prediction accuracy. Finally, the model is applied to predict the maturity of fruit flies in the new environment. For example, given a specific set of age, temperature, humidity, and other conditions, the model can output a predicted maturity value. Assume a simplified nonlinear regression model where the maturity (M) of the fruit fly is only related to age (A) and temperature (T), in the following form:

[0106] ;

[0107] Here, T0 is a baseline temperature, representing the temperature at which the maturity growth of fruit flies begins to accelerate. The maturity of fruit flies is observed and data is recorded under different temperatures and instars. By fitting the observed data, the value of T0, as well as other parameters in the model, can be estimated. To predict the maturity of a 5-day-old fruit fly at 25°C, simply substitute A=5 and T=25 into the nonlinear regression model for calculation. In this embodiment, maturity is presented in the form of time or probability, that is, the time required for a fruit fly to reach maturity under specific conditions, or the probability that the fruit fly has matured under the current conditions. For example, suppose that a fruit fly begins to exhibit active courtship behavior after reaching a temperature of 35 degrees Celsius, a light exposure duration of 7 hours, a weight of 0.1 grams, a body length exceeding 1 cm, and fully extended wings. Based on the above data, a nonlinear regression mathematical model can be constructed, using temperature, light exposure duration, weight, body length, and wing extension as input variables, and the output variable being the maturity probability of the fruit fly. When a newly captured fruit fly is measured and input into the model, the model outputs a probability value between 0 and 1, indicating the probability that the fruit fly has matured.

[0108] Finally, the first number (initial emergence count), the second number (number of fruit flies at the junction), and the current environmental data are input into a pre-trained prediction model. Based on historical data and machine learning algorithms, the prediction model can analyze the complex relationships between the data and predict the future peak emergence period of fruit flies. By predicting the future peak emergence period of fruit flies, agricultural producers can prepare for prevention and control in advance, reducing the potential threat posed by fruit flies to crops.

[0109] This embodiment utilizes precise monitoring by photoelectric sensors, image capture by cameras, and predictive models to effectively grasp the emergence cycle and population dynamics of fruit flies, providing strong scientific support for agricultural production.

[0110] In one embodiment of this invention, the adult insect chamber is equipped with a tunable light source. The photoelectric data includes the light intensity inside the adult insect chamber and the light wave reflection data of the wing surface of the fruit fly insect under the tunable light source. Determining the full wing spread characteristics of the fruit fly insect in the adult insect chamber based on the photoelectric data includes the following steps:

[0111] S410. When the light intensity in the adult insect chamber is greater than the preset first light intensity threshold, determine the wavelength range of the tunable light source as the target wavelength range, and adjust the wavelength of the tunable light source to the target wavelength range.

[0112] S420. Obtain spectral feature data of the wings of the giant fruit fly at different growth stages through a preset spectral feature database, and compare the light wave reflection data with the spectral feature data to obtain the growth stage corresponding to the light wave reflection data, and obtain the number of fruit flies corresponding to each growth stage.

[0113] S430. Based on the growth stage corresponding to the light wave reflection data, query the preset feature database to obtain the fully spread wing characteristics of fruit flies in the adult insect chamber.

[0114] Inside the adult insect chamber, a tunable light source is used to accurately capture the wing characteristics of fruit flies. This source adjusts the wavelength to provide specific light wavelengths according to different needs. When the light intensity inside the chamber exceeds a preset first light intensity threshold, it indicates sufficient ambient light, which may interfere with the accuracy of the light reflection data. In this case, the system automatically adjusts the wavelength range of the tunable light source to a pre-determined target wavelength range. This target wavelength range is selected based on research into the reflective characteristics of fruit fly wings, maximizing the highlighting of characteristic differences on the wing surface. During implementation, a light intensity sensor monitors the light intensity inside the chamber in real time. If the threshold is exceeded, the light source adjustment mechanism is triggered, ensuring that subsequent light reflection data acquisition is conducted at the optimal wavelength. This effectively reduces interference from ambient light and improves the accuracy and stability of data acquisition.

[0115] Next, a pre-set spectral feature database is used, which stores spectral feature data of the wing reflectance of fruit flies at different growth stages. This data was obtained through long-term observation and experimentation. The currently collected light wave reflectance data is compared with the data in the database to determine the best matching spectral features, and based on this, the growth stage of the fruit fly is determined. Simultaneously, the number of fruit flies at each growth stage is counted to provide data support for subsequent maturity assessment. In practice, the comparison results can be displayed through a user interface, allowing users to intuitively see the number and proportion of fruit flies at different growth stages, facilitating user decision-making.

[0116] After determining the growth stages of the fruit fly, the full wing spread characteristics of the fruit flies in the adult chamber are queried from another preset feature database based on the growth stages corresponding to the light wave reflection data. The preset feature database includes detailed characteristic descriptions of fruit flies at each stage from larva to adult. Based on the query results and the current light wave reflection data, a comprehensive judgment is made as to whether the fruit flies have reached the full wing spread state.

[0117] In this embodiment, the light wave reflection data is acquired through a photoelectric sensor, which is used to detect the reflection of light waves of different wavelengths in order to obtain the light wave reflection data of the wing surface of fruit flies.

[0118] Specifically, the light reflection data includes the first spectral bandwidth, the first peak position, and the first peak intensity. Light reflection data from the wing surface of fruit flies is collected under different light intensities because fruit fly emergence is influenced by environmental factors, primarily temperature, which varies with light intensity. Therefore, light intensity is used as the independent variable for monitoring fruit fly emergence to analyze the impact of different light intensities on the emergence stage of fruit flies.

[0119] In one embodiment, light reflection data from the wing surface of fruit flies can be collected under different light intensities. That is, multiple light intensity monitoring data points are acquired, and multiple fruit fly light reflection data points are collected under these multiple light intensity monitoring data points. For example, light intensity monitoring data at 12:00 noon and 1:00 PM are selected as a first preset time period. If the average light intensity of these two light intensity data points is greater than a first light intensity threshold, the light wavelength is defined as the target wavelength range. Light intensity monitoring data at 5:00 PM and 6:00 PM are selected as a second preset time period. If the average light intensity of these two light intensity data points is greater than a second light intensity threshold, the light wavelength is defined as the second wavelength range. Light reflection data from the wing surface of fruit flies is collected within both the target wavelength range and the second wavelength range. The changes in the light reflection data from the wing surface of fruit flies are compared between the two time periods to determine whether the fruit flies are in full wing-spread condition.

[0120] Spectral characteristics are one of the properties of matter; every substance has its unique spectrum. Different substances have different spectra. Light wave reflectance data is obtained through light wave reflectance. The curve obtained by plotting wavelength on the x-axis and reflectance on the y-axis is called the reflectance spectral characteristic curve of the object. When a light source shines on the surface of an object, the object selectively reflects electromagnetic waves of different wavelengths. Reflectance refers to the ratio of the luminous flux reflected by the object in a certain wavelength band to the luminous flux incident on the object; it is an essential property of the object's surface. Spectral reflectance is the object's inherent characteristic of color. However, when the light intensity is too strong, directly using the current light source may result in excessively intense reflected light, causing noise or saturation in the captured image, affecting the accuracy of determining the full wing spread feature. Therefore, it is necessary to adjust the wavelength of the tunable light source to a suitable range to obtain more accurate reflectance data.

[0121] Tunable light sources are essential instruments for wavelength adjustment in optical communication testing, allowing for precise and rapid adjustment of the appropriate wavelength to meet testing requirements. To acquire light reflection data from the wing surfaces of fruit flies under a tunable light source, the wavelength and frequency of the light are first adjusted to ensure the monitored light wave is within the target wavelength range. This reduces interference from excessively strong light on the reflection data, guaranteeing the clarity and accuracy of the images captured by the camera and the accuracy of data acquisition.

[0122] Secondly, light wave reflection data from the surface of the wings of fruit flies is collected within the target wavelength range. Specifically, the first spectral bandwidth, first peak position, and first peak intensity of the light wave reflection data are collected to obtain the reflectivity of the fruit fly wing surface. The formula for calculating the reflectivity of the fruit fly wing is as follows:

[0123] ;

[0124] The incident light intensity is The intensity of the reflected light is .

[0125] Record the intensity of incident light on the wings of fruit flies. The intensity of reflected light from the wings of fruit flies The reflectance of the wings of fruit flies within the target wavelength range was calculated. .

[0126] By plotting wavelength as the x-axis and reflectivity... The curve with y as the vertical axis yields the first reflectance spectral characteristic curve of the wing surface of fruit flies, providing the first spectral bandwidth, first peak position, and first peak intensity. Spectral characteristic data of fruit flies in the eclosion stage are acquired, and the light wave reflectance data is compared with the spectral characteristic data of fruit flies in the eclosion stage to obtain the matching degree between the light wave reflectance data and the spectral characteristic data. The matching degree determines whether the fruit fly is in the eclosion stage, and if the fruit fly is in the eclosion stage, the growth stage characteristics of the fruit fly are identified as eclosion stage characteristics.

[0127] This embodiment obtains spectral feature data of the wings of fruit flies at different growth stages by using a preset spectral feature data database, and compares the light wave reflection data with the spectral feature data to obtain the growth stage corresponding to the light wave reflection data. This can accurately identify the fully spread wings of fruit flies in the adult chamber, providing reliable data support for fruit fly maturity assessment and emergence cycle prediction. It not only improves the accuracy and stability of data collection, but also makes it easier for agricultural producers to monitor the growth and development of fruit flies.

[0128] In one embodiment of this invention, the light wave reflection data and spectral feature data are compared to obtain the growth stage corresponding to the light wave reflection data, and the number of fruit flies corresponding to each growth stage, including the following steps:

[0129] S510. Perform spectral analysis on the light wave reflection data to obtain the first spectral bandwidth, first peak position, and first peak intensity of the light wave reflection data;

[0130] S520: Obtain the second spectral bandwidth, second peak position, and second peak intensity of the spectral feature data;

[0131] S530. Calculate the first matching degree between the first spectral bandwidth and the second spectral bandwidth, the second matching degree between the first peak position and the second peak position, and the third matching degree between the first peak intensity and the second peak intensity according to the preset feature matching algorithm.

[0132] S540. The sum of the first matching degree, the second matching degree and the third matching degree is taken as the target matching degree between the light wave reflection data and the spectral feature data.

[0133] S550. When the target matching degree is greater than the preset matching degree threshold, determine that the light wave reflection data matches the spectral feature data, and take the growth stage corresponding to the spectral feature data as the growth stage corresponding to the light wave reflection data that matches it.

[0134] S560. Statistically obtain the number of fruit flies corresponding to each growth stage.

[0135] In this embodiment, spectral analysis technology is first used to extract the first spectral bandwidth, first peak position, and first peak intensity of the light wave reflection data. The spectral bandwidth refers to the frequency range occupied by the light wave reflection data within a certain range, reflecting the color richness of the light reflected from the surface of the fruit fly's wings. The peak position refers to the wavelength corresponding to the point of maximum intensity in the reflected light wave, which is directly related to a specific structure or pigment on the surface of the fruit fly's wings. The peak intensity is the reflection intensity of the light wave at this wavelength, reflecting the light reflection capability of the fruit fly's wing surface. Using these three parameters, a preliminary characteristic profile of the light wave reflected from the fruit fly's wings can be constructed, laying the foundation for subsequent feature matching. In practice, the collected light wave reflection data can be automatically processed using preset spectral analysis software to extract the required spectral feature parameters.

[0136] Next, spectral feature data of the reflection from the wings of fruit flies at different growth stages needs to be obtained from a pre-defined spectral feature database. This spectral feature data includes the second spectral bandwidth, the second peak position, and the second peak intensity. The spectral feature data in the database was obtained through long-term observation and experimental measurement of a large number of fruit fly samples, representing the typical characteristics of light waves reflected from the wings of fruit flies at different growth stages. In practice, a database query can be used to quickly retrieve the corresponding spectral feature data based on the fruit fly species and growth stage, preparing for subsequent feature comparison.

[0137] To determine whether light wave reflection data corresponds to spectral feature data, feature matching can be used. A pre-defined feature matching algorithm can calculate the first degree of matching between the first and second spectral bandwidths, the second degree of matching between the first and second peak positions, and the third degree of matching between the first and second peak intensities. These matching degrees are calculated by comparing the similarity between the two sets of spectral feature parameters, typically using similarity metrics such as Euclidean distance or cosine similarity. A higher matching degree indicates a greater similarity between the two sets of data, meaning a higher probability that the light wave reflection data corresponds to the spectral feature data.

[0138] After obtaining the three matching degrees, the sum of the first matching degree, the second matching degree, and the third matching degree is used as the target matching degree to fully consider the importance of each matching degree in feature matching and ensure that the final matching result is more accurate and reliable.

[0139] When the target matching degree exceeds a preset matching degree threshold, it can be determined that the light wave reflection data matches the spectral feature data. At this point, the growth stage corresponding to the spectral feature data is taken as the growth stage corresponding to the matching light wave reflection data. This achieves a mapping from light wave reflection data to fruit fly growth stages, providing crucial information for subsequent fruit fly population statistics and maturity assessment. Subsequently, after determining the growth stage corresponding to each light wave reflection data point, the number of fruit flies corresponding to each growth stage can be statistically calculated. In practice, the statistical results can be presented to users in the form of charts or reports, allowing them to intuitively understand the growth status and population distribution of the fruit fly population.

[0140] In another embodiment, the reflectance of light wave reflection data from the wing surface of fruit flies within the target wavelength range can be calculated using the reflectance formula. By plotting wavelength as the x-axis and reflectivity... Using the curve with ordinate as the vertical axis, the second reflectance spectral characteristic curve of the wing surface of fruit flies is obtained, yielding the second spectral bandwidth, second peak position, and second peak intensity. The matching degree between the obtained first spectral bandwidth, first peak position, and first peak intensity and the second spectral bandwidth, second peak position, and second peak intensity is calculated using the following formula:

[0141] ;

[0142] Where A represents the first spectral bandwidth, first peak position, and first peak intensity in the light wave reflection data; and B represents the second spectral bandwidth, second peak position, and second peak intensity.

[0143] The first matching degree is calculated by using the matching degree formula between the first spectral bandwidth and the second spectral bandwidth; the second matching degree is calculated by using the matching degree formula between the first peak position and the second peak position; the third matching degree is calculated by using the matching degree formula between the first peak intensity and the second peak intensity; the sum of the first matching degree, the second matching degree and the third matching degree is compared with the matching degree of the spectral feature data. If the matching degree is greater than the first preset value, it is determined that the giant fly is in the fully spread wing stage.

[0144] Specifically, the matching degree calculation formula includes:

[0145] ;

[0146] In the formula, M is any one of the first matching degree, the second matching degree, and the third matching degree. Based on M being the first matching degree, B1 is the first spectral bandwidth and B2 is the second spectral bandwidth. Based on M being the second matching degree, B1 is the first peak position and B2 is the second peak position. Based on M being the third matching degree, B1 is the first peak intensity and B2 is the second peak intensity.

[0147] This embodiment can accurately identify the growth stages of fruit flies within the adult fly enclosure and count the number of fruit flies corresponding to each growth stage. This not only improves the accuracy and efficiency of fruit fly growth stage identification but also provides important decision-making information for agricultural producers. By monitoring the growth and population distribution of fruit flies in real time, agricultural producers can formulate more scientific control measures, reduce the damage of fruit flies to crops, and improve agricultural production efficiency. Furthermore, this technical solution can also be applied to the identification of growth stages and population statistics of other insects or animals.

[0148] In one embodiment of this invention, image recognition is performed on the first image to obtain the first number and pre-wing-spreading characteristics of fruit flies on the plane where the molting chamber and adult chamber are located, including:

[0149] S610. Preprocess the first image to obtain multiple image features;

[0150] S620. Determine the first number and pre-wing-spreading features of fruit flies on the plane where the molting chamber and adult chamber are located from multiple image features.

[0151] In this embodiment, the acquired first image is first preprocessed to improve image quality and enhance the features of fruit flies, facilitating subsequent identification and analysis. Preprocessing steps include image grayscale conversion, filtering, and contrast enhancement. Grayscale conversion transforms a color image into a grayscale image to simplify subsequent processing; filtering removes noise, making the insect features clearer; and contrast enhancement highlights the difference between the insect and the background, improving identification accuracy. In practice, image processing software, such as OpenCV, can be used for automated preprocessing of the first image. For example, a specific grayscale conversion algorithm can be used to convert the color image to grayscale; then, a suitable filter, such as a Gaussian filter, can be selected to smooth the image; finally, the brightness and contrast of the image can be adjusted to make the insect features more prominent. After preprocessing, the fruit flies in the image will be easier to identify and count.

[0152] After image preprocessing, the next step is to determine the initial quantity and pre-wing-spreading features of fruit flies in the plane connecting the emergence chamber and the adult chamber from multiple image features. First, image segmentation techniques are used to separate the fruit flies from the background. This can be achieved through thresholding, edge detection, and region growing. Then, feature extraction techniques are used to extract key features of the fruit flies, such as shape, size, and texture. In practice, machine learning algorithms, such as Support Vector Machines (SVM) and Convolutional Neural Networks (CNN), can be used to classify and identify the extracted features. By training the model, it can learn the characteristics of fruit flies at different growth stages and accurately identify and count the insects in the image. Furthermore, by comparing image data from different time points, the growth trend and quantity changes of fruit flies can be analyzed, providing important information for subsequent control measures.

[0153] Specifically, the image recognition process in this embodiment is divided into information acquisition, preprocessing, feature extraction and selection, classifier design, and classification decision. In this embodiment, information acquisition mainly involves monitoring infrared signals using an infrared sensor. If an infrared signal is detected, a camera is used to capture and acquire the information. The steps for processing the captured images are not detailed here.

[0154] In another embodiment, convolutional neural network (CNN) algorithms can be used to extract important features from images to identify and classify the shape, texture, and pattern of fruit flies during their growth stages, thereby achieving accurate classification. A CNN is a type of feedforward neural network that includes convolutional computations and is used to process high-dimensional input data such as images. CNN algorithms efficiently extract important features from images through convolutional and pooling layers, then process the compressed image information through fully connected layers and output the results. This method reduces model parameters, lowers computational cost, and effectively handles the translation invariance of images, meaning that different locations of objects in an image have the same meaning. This characteristic of CNN algorithms makes them very suitable for image recognition tasks. Converting images into feature vectors typically requires image processing and machine learning algorithms, especially CNNs in deep learning.

[0155] Specifically, the process of converting an image into a feature vector includes the following steps:

[0156] S1. Convolutional and pooling layers from a convolutional neural network algorithm are used to extract features from the image. These layers capture key information such as the edges and colors of fruit flies, while also reducing the dimensionality of the data, making subsequent processing more efficient.

[0157] S2. Through a fully connected layer, the extracted features are mapped to a higher-dimensional space. Representations in a higher-dimensional space are more conducive to classification or recognition tasks.

[0158] S2. The output of a convolutional neural network algorithm can be viewed as a feature vector. This feature vector contains key information about the image and can be used for image classification. It allows us to obtain the features of the target image through the feature vector and then select the features representing the outward-spreading wings from among these features.

[0159] In summary, by utilizing the convolutional neural network algorithm, with its unique architecture and learning capabilities, the image recognition and processing becomes more efficient and accurate by converting the image into a feature vector to obtain the target image features and then identifying the pre-wing features within the target image features.

[0160] This embodiment utilizes image recognition to accurately identify and count fruit flies on the plane connecting the emergence chamber and the adult chamber, while simultaneously acquiring their pre-wing-spreading characteristics. This not only improves the accuracy and efficiency of fruit fly identification and counting but also provides important decision-making support for agricultural producers.

[0161] In one embodiment of this invention, after performing image recognition on the first image to obtain the first number and pre-wing-spreading characteristics of fruit flies on the plane where the molting chamber and adult chamber are connected, the following steps are included:

[0162] S710. Screen the features before wing spread to obtain key features;

[0163] S720. Based on preset mapping rules, key features are mapped to obtain corresponding key feature interpretations;

[0164] S730. Determine the scope of keywords in the preset feature database based on the definition of key features;

[0165] S740. Within the keyword range, identify and output the keywords that correspond to the key feature definitions.

[0166] After obtaining the pre-wing-spreading features of fruit flies, these features are screened to select the most representative and discriminative features from a large pool, facilitating subsequent analysis and identification. Principal component analysis can be used to screen key features, identifying those that contribute most to classification and identification. In practice, feature selection algorithms can be used to automate the screening of pre-wing-spreading features. Based on indicators such as variance, correlation, and mutual information, features are ranked and selected, ultimately retaining the most representative key features. These key features allow for a more accurate description of the pre-wing-spreading state of fruit flies.

[0167] After obtaining the key features, these features are mapped according to pre-defined mapping rules to obtain corresponding key feature interpretations. The mapping rules associate key features with specific biological meanings or classification labels. In practice, a feature-interpretation mapping table can be constructed, including various key features and their corresponding interpretations. For example, a key feature corresponding to the wing length of fruit flies would be interpreted as "wing length is an important characteristic for distinguishing different species of fruit flies." Through these mapping rules, abstract key features can be transformed into biologically meaningful interpretations, providing more intuitive and easily understood information for subsequent analysis and identification.

[0168] This embodiment uses a mapping method to interpret current key features and automatically map them to corresponding keywords. If a keyword cannot be mapped, other features of fruit flies are selected for secondary mapping to obtain the keyword. These other features can be the body color and morphological characteristics of fruit flies. For example, the large fruit fly has a golden-yellow body color after emergence, a black longitudinal stripe in the center of the dorsal surface of the mesothorax running from the base to the abdomen, and a wide black transverse stripe near the anterior margin of the third abdominal segment. The longitudinal and transverse stripes intersect to form a cross shape, which are the body characteristics of adult fruit flies. The corresponding keywords are then retrieved from a preset database using an algorithm. Within the search range, the keywords are located, and the growth stage characteristics corresponding to the keywords are determined, thus obtaining the current growth stage of fruit flies in the monitoring area.

[0169] After obtaining the key feature definitions, the keyword range is determined within a pre-defined feature database. This database contains a large amount of characteristic information about fruit flies, allowing for rapid retrieval of relevant keywords based on the feature definitions. In practice, database queries can be used to perform fuzzy or exact matching based on the key feature definitions to find a range of keywords related to those definitions. For example, if the key feature definition is "wing length," the keyword range might include terms related to wing length such as "long wings," "short wings," and "wingspan." Determining the keyword range further narrows the search space, improving the accuracy and efficiency of subsequent keyword selection.

[0170] Once the keyword range is determined, the keywords corresponding to the key feature definitions can be identified and output. In practice, text mining and natural language processing techniques can be used to automatically filter and rank the vocabulary within the keyword range. For example, the degree of matching between keywords and key feature definitions can be assessed by calculating similarity scores; simultaneously, the frequency and weight of keywords in relevant literature or databases can be considered to determine their popularity and relevance. Finally, the keywords with the highest scores or those that best meet the requirements can be selected as output results, providing important information for subsequent classification, identification, or control of fruit flies.

[0171] In this embodiment, the mapping rules are first determined. The growth stages of fruit flies include four stages: egg, larva, pupa, and adult. The characteristics of each stage are recorded in the mapping rules. The unique characteristics of the four growth stages of fruit flies include: Egg stage: The egg is oblong, slightly pointed at one end, slightly curved, milky white in the middle, and relatively transparent at both ends. Larval stage: Milky white, conical, pointed at the front end, and stout at the rear end. The mouth hook is black and often retracted into the prothorax. Pupa stage: Oval, golden yellow, turning yellowish-brown before emergence. The nipple-like protrusions of the anterior spiracles are still clearly visible during the larval stage. Adult stage: The adult resembles a fly in shape, but is smaller, and its body color resembles that of a bee, with alternating black and yellow markings. The adult body length is 10-13 mm, the wingspan is about 21 mm, and the entire body is pale yellowish-brown. There is a black stripe in the center of the dorsal surface of the mesothorax, running from the base to the abdomen. Near the anterior margin of the third abdominal segment, there is a relatively wide black horizontal stripe, which intersects to form a cross shape.

[0172] After determining the mapping rules, principal component analysis (PCA) can be used to identify key features. First, each pre-spreading feature is averaged (decentered) to ensure the dataset's mean is zero, facilitating subsequent calculations. The second step is to calculate the covariance matrix: Next, the covariance between all pre-spreading features is calculated, forming the covariance matrix. The covariance matrix reflects the correlation between different pre-spreading features. Then, the covariance matrix is ​​decomposed into pre-spreading eigenvalues ​​to obtain key eigenvalues ​​and key eigenvectors. Key eigenvalues ​​represent the importance of each principal component, while key eigenvectors define the direction of the principal components. Based on the magnitude of the key eigenvalues, the eigenvectors corresponding to the k largest eigenvalues ​​are selected. These eigenvectors constitute a new feature space, i.e., the principal components. Finally, the original data is projected onto the k principal components to obtain the dimensionality-reduced key features.

[0173] During the retrieval process, a block-based search approach can be adopted. The search set, including the body shape and appearance characteristics of fruit flies at all growth stages, is entered into a pre-defined database. Specifically, the lookup table containing information on all growth stages of fruit flies is first divided into several blocks. Elements within each block do not need to be ordered, but the blocks themselves must be ordered. For each block, an element is selected and placed into an index table, which is sorted. The search process may include: first, searching the index table for the block containing the keyword; then, performing a sequential search within that block to determine the keyword corresponding to the current key feature; finally, determining the position of the keyword and retrieving it from the pre-defined database to ascertain the current growth state of the fruit fly.

[0174] This embodiment can screen key features from the pre-wing-spread characteristics of fruit flies and map them to biologically meaningful interpretations. Then, it determines the keyword range from a pre-set feature database and finally identifies the keywords corresponding to the key feature interpretations. This not only improves the accuracy and efficiency of fruit fly feature analysis and identification but also provides agricultural producers with more intuitive and easily understood information. By monitoring and analyzing the key features of fruit flies in real time, agricultural producers can formulate more scientific control measures to reduce the damage of fruit flies to crops.

[0175] In one embodiment of this example, the following steps are also included:

[0176] S810. In the case where the key features cannot be mapped to obtain the corresponding key feature interpretation, perform image recognition on the first image to obtain multiple color features.

[0177] S820. Select the cross-shaped black horizontal stripe feature and the golden yellow body color feature from multiple color features, and use the cross-shaped black horizontal stripe feature and the golden yellow body color feature as the second target key features.

[0178] S830. Obtain the related interpretations of the key feature definitions, and expand the keyword range based on the related interpretations to obtain the second keyword range;

[0179] S840. Within the second keyword range, identify and output the keywords corresponding to the key features of the second objective.

[0180] In image recognition, when key features cannot be mapped to corresponding key feature interpretations, it indicates that the current feature library or mapping rules may not cover all feature cases. To compensate for this deficiency, further image recognition strategies can be adopted. Specifically, more in-depth image recognition can be performed on the first image to obtain multiple color features. Color features are a commonly used feature in image recognition; they describe the color information of objects or regions in an image and are of great significance for distinguishing different objects or recognizing specific patterns. In implementation, color space conversion algorithms can be used to convert the image from the RGB color space to the HSV or Lab color space to better separate color information; then, multiple color features in the image can be extracted using methods such as color histograms and color moments.

[0181] After acquiring multiple color features, representative features are selected as the second target key features. Based on the biological characteristics of fruit flies, the cross-shaped black stripes and golden-yellow body color are two prominent and common color features. The cross-shaped black stripes typically appear on the back or wings of fruit flies, forming cross-shaped or intersecting black stripes; while the golden-yellow body color is a common feature of fruit flies, exhibiting a bright golden hue. In implementation, feature selection algorithms can be used to automatically filter and compare the extracted color features to find those that best match the cross-shaped black stripes and golden-yellow body color. These selected features will serve as the second target key features for subsequent identification and keyword determination.

[0182] After identifying the key features of the second target, the keyword range needs to be further expanded to find corresponding keywords. This can be achieved by obtaining related definitions of the key features. Related definitions refer to other definitions that are related to or similar to the key feature definitions, and they may contain more information or broader descriptions. In practice, semantic similarity calculation techniques can be used to find related definitions similar to the key feature definitions; then, based on these related definitions, the keyword range is further searched and expanded in a pre-defined feature database to obtain the second keyword range. The second keyword range will contain more keywords related to the key features of the second target, improving the accuracy and diversity of subsequent keyword determination.

[0183] After obtaining the second keyword range, keywords corresponding to the second target key features within this range are identified and output. In practice, text mining and natural language processing techniques can be used to automatically filter and rank the vocabulary within the second keyword range. For example, the degree of matching between keywords and the second target key features can be assessed by calculating similarity scores; simultaneously, the frequency and weight of keywords in relevant literature or databases can be considered to determine their popularity and relevance. Finally, the keywords with the highest scores or those best meeting the requirements can be selected as output results, providing important information for subsequent classification, identification, or control of fruit flies.

[0184] This embodiment can identify second target key features by further extracting color features and filtering them when the corresponding key feature definitions cannot be obtained through mapping. Then, by acquiring associated definitions and expanding the keyword range, the keywords corresponding to the second target key features are finally determined. This not only improves the flexibility and accuracy of fruit fly feature analysis and identification but also provides agricultural producers with richer and more diverse information. By monitoring and analyzing the color features of fruit flies in real time, agricultural producers can formulate more scientific control measures and reduce the damage of fruit flies to crops.

[0185] In one embodiment of this example, before determining the keyword range in a preset feature database based on the definition of key features, the following steps are also included:

[0186] S910. Store the characteristics of each growth stage of the giant fruit fly in a preset feature database, and set the search range of the preset feature database.

[0187] S920. Within the keyword range, identify and output the keywords corresponding to the key feature definitions, including:

[0188] S930. Within the search scope, locate keywords and determine the growth stage characteristics corresponding to the keywords. The growth stage characteristics corresponding to the keywords are used to characterize the current growth stage of the giant water fly in the monitoring area.

[0189] To effectively identify the growth stages of the giant water fly, it is first necessary to store the characteristics of each growth stage in a pre-defined feature database. The growth stages of the giant water fly typically include egg, larva, pupa, and adult, each with its unique morphological and biological characteristics. In practice, this can be achieved by collecting a large amount of image, video, or textual descriptions of each growth stage, and then using image processing and feature extraction methods to extract representative features such as morphology, color, and texture, which are then stored in the pre-defined feature database, forming a giant water fly growth stage feature library. Simultaneously, to improve retrieval efficiency, it is also necessary to set the search scope of the pre-defined feature database, that is, to limit the search to specific growth stages or specific feature types according to actual needs, thereby reducing unnecessary calculations and data comparisons.

[0190] After defining the keyword range, the keywords are located within that range, and the corresponding growth stage characteristics are determined. In practice, fuzzy or exact matching can be performed on the database based on the keywords to find all feature records related to them. Then, by comparing these feature records with the characteristics of the giant water fly's current growth stage, the best-matching growth stage characteristics are determined. For example, if the keyword is "fully spread wings," then the database will be searched for all features related to fully spread wings, such as body size, color, and behavior, and compared with the currently monitored giant water fly characteristics to determine whether the fly is currently in the fully spread wings stage.

[0191] Furthermore, S930 can not only identify the growth stage characteristics corresponding to keywords, but also use these characteristics to characterize the current growth stage of the giant mealybug in the monitoring area. This means that through this step, agricultural producers or researchers can understand the growth status of the giant mealybug in real time, providing an important basis for formulating effective control measures or research plans. For example, if it is determined that the giant mealybug is currently in the adult stage, targeted pesticide application measures can be taken to control its growth and reproduction.

[0192] This embodiment stores characteristic information of each growth stage of the giant water fly in a pre-set feature database and sets the search scope to improve search efficiency. Then, it locates keywords within the keyword range and determines the growth stage characteristics corresponding to the keywords, thereby characterizing the current growth stage of the giant water fly in the monitoring area in real time. This not only improves the accuracy and efficiency of identifying the growth stage of the giant water fly but also provides agricultural producers and researchers with more intuitive and easy-to-understand information. By monitoring the growth status of the giant water fly in real time, control measures and research plans can be formulated more scientifically, reducing the damage of the giant water fly to crops and promoting the sustainable development of agricultural production.

[0193] like Figure 2 and Figure 3As shown in the illustration, this application also provides a fruit fly monitoring device, which includes a monitoring device housing 10, an emergence chamber 30, an adult chamber 20, and a hollow frame. The emergence chamber 30 and the adult chamber 20 are located inside the monitoring device housing 10. At least one infrared sensor is provided at the outlet of the emergence chamber 30 and the entrance of the adult chamber 20. The emergence chamber 30 is entirely located in the soil environment, while the adult chamber 20 is entirely located on the ground. The adult chamber 20 has a structure that is narrower at the top and wider at the bottom, and it contains a camera 50 and a supplementary light 60 for monitoring fruit flies. The top of the adult chamber 20 is seamlessly connected to the hollow frame, and at least one counting sensor 40 is provided inside the hollow frame. The fruit fly monitoring device also includes:

[0194] The memory is configured to store instructions; and

[0195] The processor is configured to retrieve instructions from memory and, when executing the instructions, to implement the aforementioned method for monitoring the emergence of fruit flies.

[0196] The fruit fly monitoring device includes a molting chamber 30, an adult chamber 20, and a camera 50. The molting chamber 30, adult chamber 20, and camera 50 are all located in the monitoring area of ​​the device housing 10, which is used to monitor the growth activities of fruit flies. The adult chamber 20 is positioned above the molting chamber 30. Due to the biological characteristics of fruit flies, the adult chamber 20 is designed with a narrow top and wide bottom structure to allow fruit flies to fly from the bottom to the top of the chamber, facilitating better monitoring. The adult chamber 20 contains the camera 50, a photoelectric sensor, and a counting sensor 40. One, two, or more infrared sensors can be used; this application does not impose any limitations on this.

[0197] In specific implementation, the fruit fly monitoring device housing 10 is also equipped with a counting sensor 40 for counting. The counting sensor 40 can be a photoelectric sensor, an ultrasonic sensor, a vibration sensor, etc. Taking an ultrasonic sensor as an example, it can count fruit flies by the change in the ultrasonic waves reflected when the insect passes through the sensor, so as to better determine the degree of emergence of fruit flies in the monitoring area, and thus determine the current growth status of fruit flies. The counting sensor 40 is located in a hollow square frame.

[0198] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0199] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0200] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0201] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0202] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0203] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0204] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0205] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0206] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method of emergence monitoring of a Drosophilidae insect, characterized by, The application is applied to a fruit fly monitoring device, which comprises a box body, an emergence bin, an adult bin and a hollow square box. The emergence bin and the adult bin are arranged in the box body. At least one infrared sensor is arranged at the outlet of the emergence bin and the inlet of the adult bin. The emergence bin is arranged in the soil environment, and the adult bin is arranged on the ground. The adult bin has a narrow top and a wide bottom. A camera and a light supplement lamp for monitoring fruit flies are arranged in the adult bin. The top end of the adult bin is seamlessly connected with the hollow square box. At least one counting sensor is arranged in the hollow square box. The method comprises the following steps: In response to the infrared signal transmitted by the infrared sensor, the light intensity in the adult bin is acquired; A first instruction is sent, which is used to control the camera to shoot a first image of the plane where the emergence bin and the adult bin are connected, and the first sending time of the first instruction is recorded. The time is counted from the first sending time. If the light intensity is less than the preset light intensity threshold, the light supplement lamp is turned on; The first image is subjected to image recognition to obtain the first number of fruit flies on the plane where the emergence bin and the adult bin are connected and the pre-flap characteristic; The emergence period of fruit flies is determined, and the difference between the first number and the second number is calculated; According to the difference and the emergence period, the current emergence situation of fruit flies is determined, which includes the emergence rate and the emergence rate; The adult bin is also provided with a photoelectric sensor. Before the emergence period of fruit flies is determined, the following steps are included: In response to the photoelectric data transmitted by the photoelectric sensor, the fully-flapped characteristic of fruit flies in the adult bin is determined according to the photoelectric data; A second instruction is sent, which is used to control the camera to shoot a second image of the plane where the emergence bin and the adult bin are connected; The second image is subjected to image recognition to obtain the second number of fruit flies on the plane where the emergence bin and the adult bin are connected; Current environmental data are acquired; The maturity of fruit flies in the adult bin is determined through the current environmental data and the fully-flapped characteristic; The first number, the second number and the current environmental data are input into a pre-trained prediction model to predict the future emergence peak period of fruit flies; The adult bin is provided with a tunable light source. The photoelectric data includes the light intensity in the adult bin and the light wave reflection data of the wing surface of the fruit flies under the tunable light source. The fully-flapped characteristic of fruit flies in the adult bin is determined according to the photoelectric data, which includes the following steps: When the light intensity in the adult bin is greater than the preset first light intensity threshold, the wavelength range of the tunable light source is determined as the target wavelength range, and the wavelength of the tunable light source is adjusted to the target wavelength range. Obtaining spectral feature data of wing reflection of the fruit fly at different growth stages through a preset spectral feature database, and performing data comparison between the light wave reflection data and the spectral feature data to obtain a growth stage corresponding to the light wave reflection data and a number of real fly insects corresponding to each growth stage; According to the growth stage corresponding to the light wave reflection data, the fully expanded wing feature of the real fly insects in the adult warehouse is obtained by querying a preset feature database.

2. A method of emergence monitoring of a Drosophilidae insect according to claim 1, characterized in that, The determination of the eclosion cycle of the real fly insects includes: After the first sending time, the camera is controlled to take an image every interval of a preset time period, and the number of real fly insects is obtained by image recognition of each taken image; When the number of real fly insects is less than a preset number threshold, the corresponding image is taken as a final image, and a shooting time of the final image is obtained; The difference between the shooting time and the first sending time is taken as the eclosion cycle of the real fly insects.

3. A method of emergence monitoring of a Drosophilidae insect according to claim 1, characterized in that, The data comparison between the light wave reflection data and the spectral feature data to obtain a growth stage corresponding to the light wave reflection data and a number of real fly insects corresponding to each growth stage includes: Performing spectral analysis on the light wave reflection data to obtain a first spectral bandwidth, a first peak position and a first peak intensity of the light wave reflection data; Obtaining a second spectral bandwidth, a second peak position and a second peak intensity of the spectral feature data; According to a preset feature matching algorithm, a first matching degree of the first spectral bandwidth and the second spectral bandwidth, a second matching degree of the first peak position and the second peak position, and a third matching degree of the first peak intensity and the second peak intensity are calculated; The sum of the first matching degree, the second matching degree and the third matching degree is taken as a target matching degree of the light wave reflection data and the spectral feature data; In the case where the target matching degree is greater than a preset matching degree threshold, it is determined that the light wave reflection data and the spectral feature data match, and the growth stage corresponding to the spectral feature data is taken as the growth stage corresponding to the light wave reflection data matched therewith; The number of real fly insects corresponding to each growth stage is obtained by statistics.

4. A method of emergence monitoring of a Drosophilidae insect according to claim 1, characterized in that, The image recognition of the first image to obtain the first number of real fly insects and the pre-flap feature of the plane where the eclosion warehouse and the adult warehouse are connected includes: Preprocessing the first image to obtain a plurality of image features; Determining the first number of real fly insects and the pre-flap feature of the plane where the eclosion warehouse and the adult warehouse are connected among the plurality of image features.

5. A method of emergence monitoring of a Drosophilidae insect according to claim 1, wherein After the image recognition of the first image to obtain the first number of real fly insects and the pre-flap feature of the plane where the eclosion warehouse and the adult warehouse are connected, it includes: Screening the pre-flap feature to obtain a key feature; Mapping the key feature based on a preset mapping rule to obtain a corresponding key feature interpretation; According to the key feature interpretation, determining a keyword range in a preset feature database; In the keyword range, the keyword corresponding to the key feature interpretation is determined and output.

6. A method of emergence monitoring of a Drosophilidae insect according to claim 5, wherein The method further comprises: In the case that the key feature cannot be mapped to obtain the corresponding key feature interpretation, image recognition is performed on the first image to obtain a plurality of color features; Cross black horizontal stripe features and golden yellow body color features are screened from the plurality of color features, and the cross black horizontal stripe features and the golden yellow body color features are taken as second target key features; The associated interpretation of the key feature interpretation is obtained, and the keyword range is expanded according to the associated interpretation to obtain a second keyword range; In the second keyword range, the keyword corresponding to the second target key feature is determined and output.

7. A method of emergence monitoring of a Drosophilidae insect according to claim 5, wherein Before the keyword range is determined in the preset feature database according to the key feature interpretation, the method comprises: Each growth stage feature of the Delia Antiqua is stored in the preset feature database, and a retrieval range of the preset feature database is set; The keyword corresponding to the key feature interpretation is determined and output in the keyword range, which comprises: In the retrieval range, the keyword is located, and the growth stage feature corresponding to the keyword is determined, wherein the growth stage feature corresponding to the keyword is used to represent the current growth stage of the Delia Antiqua in the monitoring area.

8. A Drosophila monitoring device, characterized in that The fruit fly insect monitoring device comprises a box body, an eclosion bin, an adult bin and a hollow square box, the eclosion bin and the adult bin are arranged in the box body, at least one infrared sensor is arranged at the outlet of the eclosion bin and the inlet of the adult bin, the eclosion bin is arranged in a soil environment as a whole, the adult bin is arranged on the ground as a whole, the adult bin has a narrow upper part and a wide lower part, a camera and a fill light for monitoring fruit fly insects are arranged in the adult bin, the top end of the adult bin is seamlessly connected with the hollow square box, at least one counting sensor is arranged in the hollow square box, and the fruit fly insect monitoring device further comprises: a memory configured to store instructions; and a processor configured to call the instructions from the memory and enable the fruit fly insect eclosion monitoring method according to any one of claims 1 to 7 when the instructions are executed.

Images