Intelligent mosquito catching device

Abstract

The invention discloses an intelligent mosquito catching device, and relates to the technical field of mosquito killing. The device comprises a mosquito luring chamber, an insect passing channel, a releasing channel, a mosquito catching chamber, a channel switching mechanism, a camera and a controller, the controller is used for recognizing whether insects exist at an inlet of the insect passing channel based on a target recognition algorithm after receiving an inlet monitoring image, if yes, an insect image is intercepted, and if not, the insect image is captured; and whether the insects are mosquitoes is identified based on the convolutional neural network, if yes, the channel switching mechanism is controlled to open the outlet of the insect passing channel and close the inlet of the releasing channel, and otherwise, the channel switching mechanism is controlled to close the outlet of the insect passing channel and enable the inlet of the releasing channel to be communicated with the insect passing channel. In this way, mosquitoes can be prevented from escaping and entering the mosquito trapping chamber from the mosquito trapping chamber, other insects can be prevented from entering the mosquito trapping chamber and can escape, then the purpose of trapping and killing the mosquitoes can be achieved, and biodiversity can be protected.

Description

Technical Field

[0001] The invention belongs to the technical field of mosquito killing, and in particular relates to an intelligent mosquito catching device. Background Art

[0002] Mosquitoes are the most familiar pests. In addition to biting and sucking blood, causing pain and itching, and disturbing sleep, mosquitoes are also the vectors of many infectious diseases (such as malaria, lymphatic filariasis, epidemic encephalitis B, dengue fever, and dengue hemorrhagic fever). Therefore, effectively killing mosquitoes has a significant effect on curbing infectious diseases.

[0003] At present, the commonly used mosquito catching method is to use the mosquito's tendency to light and certain chemicals to trap and kill mosquitoes. For example, the mosquito's tendency to light is used to lure mosquitoes with ultraviolet light, and then they are killed with electricity. However, this method will also kill beneficial insects, which will have an adverse impact on biodiversity. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent mosquito catching device to solve the problem that existing mosquito catching solutions may indiscriminately kill mosquitoes and beneficial insects, thus causing adverse effects on biodiversity.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, an intelligent mosquito catching device is provided, comprising a mosquito attracting chamber, an insect passage, a release passage, a mosquito catching chamber, a passage switching mechanism, a camera, and a controller, wherein the inlet of the insect passage is connected to the mosquito attracting chamber, the outlet of the insect passage is connected to the mosquito catching chamber, the inlet of the release passage is connected to the insect passage through the passage switching mechanism, and the outlet of the release passage is connected to the outside of the room;

[0007] The channel switching mechanism is communicatively connected to the controller, and is used to, under the control of the controller, close the exit of the insect passage and connect the entrance of the release passage to the insect passage, or open the exit of the insect passage and close the entrance of the release passage;

[0008] The camera is communicatively connected to the controller and is used to collect images of the entrance of the insect passage in real time and transmit the acquired entrance monitoring images to the controller in real time;

[0009] The controller is used to perform intelligent mosquito catching according to the following steps:

[0010] When the imported monitoring image is received, the imported monitoring image is imported in real time into a pre-trained insect recognition model based on a target recognition algorithm, and an insect recognition result is output;

[0011] If the insect identification result includes an insect marking frame, an insect image is captured in real time from the imported monitoring image according to the insect marking frame;

[0012] Importing the insect image into a mosquito recognition model pre-trained based on a first convolutional neural network in real time, and outputting a mosquito recognition result;

[0013] If the mosquito identification result indicates that it is a mosquito, the channel switching mechanism is controlled to open the exit of the insect passage and close the entrance of the release passage; otherwise, the channel switching mechanism is controlled to close the exit of the insect passage and connect the entrance of the release passage to the insect passage.

[0014] Based on the above invention content, a new mosquito catching scheme for differentially catching and killing mosquitoes and beneficial insects based on a target recognition algorithm and a convolutional neural network is provided, that is, it includes a mosquito catching chamber, an insect passage, a release passage, a mosquito catching chamber, a channel switching mechanism, a camera and a controller, wherein the controller is used to first identify whether there is an insect at the entrance of the insect passage based on a target recognition algorithm after receiving an imported monitoring image, if so, intercept the insect image, and identify whether the insect is a mosquito based on a convolutional neural network, if it is a mosquito, control the channel switching mechanism to open the exit of the insect passage and close the entrance of the release passage, otherwise control the channel switching mechanism to close the exit of the insect passage and connect the entrance of the release passage to the insect passage, so that not only can the mosquito be prevented from escaping and entering the mosquito catching chamber from the mosquito catching chamber, but also other insects can be prevented from entering the mosquito catching chamber and can escape, thereby achieving the purpose of capturing and killing only mosquitoes, which is beneficial to protecting biodiversity and convenient for practical application and promotion.

[0015] In a possible design, the mosquito-attracting chamber is surrounded by a cap body, a bird-proof net and a transverse partition arranged from top to bottom, and has a built-in mosquito-attracting member, wherein the upper surface of the transverse partition is provided with an entrance to the insect passage.

[0016] In a possible design, the mosquito attracting member adopts an exhaust port for mosquito attracting gas and is arranged on the tube wall of the support tube, wherein the support tube is located inside the mosquito attracting chamber, one end of the support tube is fixedly connected to the cap body, and the other end of the support tube is vertically fixedly connected to the upper surface of the transverse partition so as to support the cap body.

[0017] In a possible design, a storage chamber is also included, wherein the storage chamber is equipped with a carbon dioxide gas cylinder and a gas mixing chamber;

[0018] The air inlet end of the mixing chamber is connected to the carbon dioxide cylinder, and the air outlet end of the mixing chamber is connected to the support tube.

[0019] In a possible design, the camera is hung upside down on the top of the cap body, and the lens is aimed at the entrance of the insect passage.

[0020] In a possible design, a ventilation window is provided on the first side of the mosquito trapping room, a mosquito filter is installed on the ventilation window, and an exhaust fan for exhausting indoor air is installed outside the ventilation window, wherein a controlled end of the exhaust fan is communicatively connected to the controller;

[0021] The controller is also used to perform start and stop control and speed regulation control on the exhaust fan.

[0022] In a possible design, a drawer and / or a maintenance door for opening the room is installed on the second side of the mosquito catching room, wherein the drawer is located at the bottom of the mosquito catching room.

[0023] In a possible design, the channel switching mechanism includes a stepper motor and a rotating flap, wherein the controlled end of the stepper motor is communicatively connected to the controller, and the rotating flap shaft is connected to the output shaft of the stepper motor;

[0024] The rotating flap is used to close the outlet of the insect passage and connect the inlet of the release passage to the insect passage when the output shaft rotates to the first position, and to open the outlet of the insect passage and close the inlet of the release passage when the output shaft rotates to the second position.

[0025] In one possible design, the controller is also used to upload the insect image and the collection time and / or collection location of the insect image to the cloud server in real time when the mosquito identification result indicates a mosquito, so that the cloud server imports the insect image into a mosquito classification model pre-trained based on the second convolutional neural network, outputs the mosquito classification result, and performs statistical analysis based on the mosquito classification result and the collection time and / or the collection location to obtain a mosquito statistical analysis report.

[0026] In a possible design, the camera is a high-speed camera, and the controller is a Raspberry Pi microcomputer.

[0027] Beneficial effects of the above scheme:

[0028] (1) The present invention creatively provides a new mosquito catching scheme for differentially catching and killing mosquitoes and beneficial insects based on a target recognition algorithm and a convolutional neural network, that is, comprising a mosquito catching chamber, an insect passage, a release passage, a mosquito catching chamber, a passage switching mechanism, a camera and a controller, wherein the controller is used to first identify whether there is an insect at the entrance of the insect passage based on a target recognition algorithm after receiving an inlet monitoring image, if so, intercept an insect image, and identify whether the insect is a mosquito based on a convolutional neural network, if it is a mosquito, control the passage switching mechanism to open the exit of the insect passage and close the entrance of the release passage, otherwise control the passage switching mechanism to close the exit of the insect passage and connect the entrance of the release passage to the insect passage, so that not only can the mosquito be prevented from escaping and entering the mosquito catching chamber from the mosquito catching chamber, but also other insects can be prevented from entering the mosquito catching chamber and can escape, thereby achieving the purpose of capturing and killing only mosquitoes, which is beneficial to protecting biodiversity;

[0029] (2) The mosquito images and the image acquisition time and / or image acquisition location can also be uploaded to the cloud server to further obtain mosquito classification results based on the convolutional neural network, and statistical analysis can be performed in combination with the image acquisition time and / or image acquisition location to obtain a mosquito statistical analysis report, which is conducive to making appropriate mosquito control strategies and facilitating practical application and promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 A schematic diagram of the three-dimensional structure of the intelligent mosquito catching device provided in an embodiment of the present application.

[0032] Figure 2 A schematic diagram of the three-dimensional structure of a channel switching mechanism in an intelligent mosquito catching device provided in an embodiment of the present application, wherein: Figure 2 (a) shows a schematic diagram of the three-dimensional structure of the channel switching mechanism when the exit of the insect passage is closed. Figure 2 (b) shows a schematic diagram of the three-dimensional structure of the channel switching mechanism when opening the exit of the insect passage.

[0033] Figure 3 This is a front view structural schematic diagram of the intelligent mosquito catching device provided in an embodiment of the present application.

[0034] Figure 4 This is a schematic diagram of the right side structure of the intelligent mosquito catching device provided in an embodiment of the present application.

[0035] Figure 5 A schematic diagram of the top structure of the intelligent mosquito catching device provided in an embodiment of the present application.

[0036] Figure 6 A schematic diagram of the three-dimensional structure of a carbon dioxide gas enricher in an intelligent mosquito catching device provided in an embodiment of the present application.

[0037] Figure 7 This is a schematic diagram of the front view structure of a carbon dioxide gas enricher in an intelligent mosquito catching device provided in an embodiment of the present application.

[0038] Figure 8 This is a left-side structural schematic diagram of a carbon dioxide gas enricher in an intelligent mosquito catching device provided in an embodiment of the present application.

[0039] Fig. 9 A schematic diagram of the top view of the structure of a carbon dioxide gas enricher in an intelligent mosquito catching device provided in an embodiment of the present application.

[0040] In the above drawings: 11-cap body; 12-bird-proof net; 13-lateral partition; 14-support tube; 15-support column; 2-insect passage; 21-inlet; 3-release passage; 32-exit; 4-mosquito catching room; 42-mosquito filter; 43-exhaust fan; 44-drawer; 45-maintenance door; 510-output shaft; 52-rotating flap; 6-camera; 7-controller; 8-storage room; 81-carbon dioxide cylinder; 82-mixing Air cavity; 83-air hole; 100-upper bottle body; 200-lower bottle body; 300-connecting pipe; 310-air outlet; 400-bottle body upside down flipping mechanism; 410-bracket; 411-bottom plate; 412-inverted "V"-shaped support rod; 420-rotating shaft; 430-bearing; 431-first bearing; 432-second bearing; 440-motor; 500-air supply pump; 600-high temperature filter; 700-control module. DETAILED DESCRIPTION

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structures of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these embodiments without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0042] It should be understood that although the terms first and second, etc. may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another object. For example, a first object can be referred to as a second object, and similarly, a second object can be referred to as a first object without departing from the scope of the exemplary embodiments of the present invention.

[0043] It should be understood that the term "and / or" that may appear in this document is merely a description of the association relationship between associated objects, indicating that there may be three relationships. For example, A and / or B can indicate three situations: A exists alone, B exists alone, or A and B exist at the same time. For another example, A, B and / or C can indicate the existence of any one of A, B and C or any combination of them. The term " / and" that may appear in this document describes another type of association object relationship, indicating that there may be two relationships. For example, A / and B can indicate two situations: A exists alone or A and B exist at the same time. In addition, the character " / " that may appear in this document generally indicates that the previous and next associated objects are in an "or" relationship.

[0044] Embodiment 1

[0045] like Figures 1 to 5 As shown, the intelligent mosquito catching device provided in this embodiment includes but is not limited to a mosquito luring chamber, an insect passage 2, a release passage 3, a mosquito catching chamber 4, a passage switching mechanism, a camera 6, and a controller 7, wherein the inlet 21 of the insect passage 2 is connected to the mosquito luring chamber, the outlet of the insect passage 2 is connected to the mosquito catching chamber 4, the inlet of the release passage 3 is connected to the insect passage 2 through the passage switching mechanism, and the outlet 32 ​​of the release passage 3 is connected to the outside of the room. Figure 1 As shown, the mosquito-attracting chamber is used to utilize the mosquito's tendency to light and certain chemicals to lure mosquitoes from the outdoors into the indoors; specifically, the mosquito-attracting chamber may be, but is not limited to, surrounded by a cap body 11, a bird-proof net 12, and a transverse partition 13 arranged from top to bottom, and has a built-in mosquito-attracting component, wherein the upper surface of the transverse partition 13 is provided with an entrance 21 of the insect passage 2, the bird-proof net 12 is used to prevent animals such as birds that are relatively large compared to mosquitoes from entering the mosquito-attracting chamber (i.e., only animals of the same size as mosquitoes are allowed to enter the mosquito-attracting chamber), and the mosquito-attracting component is used to lure mosquitoes there. The insect passage 2 is used to connect the mosquito-attracting chamber with the mosquito-catching chamber 4 when the outlet of the insect passage 2 is opened, so that mosquitoes can enter the mosquito-catching chamber 4 from the mosquito-attracting chamber; the release channel 3 is used to connect the mosquito-attracting chamber with the outdoors when the outlet of the insect passage 2 is closed, so that insects other than mosquitoes can escape from the mosquito-attracting chamber, as shown in FIG. Figure 2As shown, the insect passage 2 and the release passage 3 adopt an L-shaped passage structure in which the passage switching mechanism is arranged at the middle corner. The mosquito catching chamber 4 is used to imprison mosquitoes entering the chamber until they die.

[0046] The channel switching mechanism is communicatively connected to the controller 7, and is used to close the exit of the insect passage 2 and connect the entrance of the release passage 3 to the insect passage 2, or to open the exit of the insect passage 2 and close the entrance of the release passage 3 under the control of the controller 7. Figure 2 As shown, specifically, the channel switching mechanism includes but is not limited to a stepper motor and a rotating flap 52, wherein the controlled end of the stepper motor is communicatively connected to the controller 7, and the rotating flap 52 is axially connected to the output shaft 510 of the stepper motor; the rotating flap 52 is used to close the outlet of the insect passage 2 and connect the inlet of the release passage 3 to the insect passage 2 when the output shaft 510 rotates to the first position, and to open the outlet of the insect passage 2 and close the inlet of the release passage 3 when the output shaft 510 rotates to the second position. The stepper motor is used to rotate step by step under the control of the controller 7; the first position is as shown in FIG. Figure 2 (a), the second position is as follows Figure 2 As shown in (b), when the stepper motor is in the second position, if the stepper motor is controlled to rotate 180 degrees clockwise in a single step, it can be rotated to the first position, thereby achieving the purpose of closing the exit of the insect passage 2 and connecting the entrance of the release passage 3 to the insect passage 2, and when the stepper motor is in the first position, if the stepper motor is controlled to rotate 180 degrees counterclockwise in a single step, it can be rotated to the second position, thereby achieving the purpose of opening the exit of the insect passage 2 and closing the entrance of the release passage 3.

[0047] The camera 6 is communicatively connected to the controller 7, and is used to collect images of the entrance 21 of the insect passage 2 in real time, and transmit the acquired entrance monitoring images to the controller 7 in real time. Figures 3-4 As shown, specifically, the camera 6 is hung upside down on the top of the cap body 11, and the lens is aimed at the entrance 21 of the insect passage 2. In addition, in order to obtain the entrance monitoring image at a high frequency, the camera 6 is preferably a high-speed camera.

[0048] The controller 7 is specifically but not limited to a Raspberry Pi microcomputer (which is an existing microcomputer based on Linux system and has the size of a credit card) and is arranged at the top of the cap body 11, and is used for but not limited to performing intelligent mosquito catching according to the following steps S11 to S14.

[0049] S11. When the imported monitoring image is received, the imported monitoring image is imported in real time into a pre-trained insect recognition model based on a target recognition algorithm, and an insect recognition result is obtained by output.

[0050] In the step S11, the target detection algorithm is an existing artificial intelligence recognition algorithm for identifying objects in a picture and marking the location of the objects, and specifically can be but not limited to Faster R-CNN (Faster Regions with Convolutional Neural Networks features, a target detection algorithm proposed by He Kaiming et al. in 2015, which won multiple first places in the 2015 ILSVRV and COCO competitions) target detection algorithm, SSD (Single Shot MultiBox Detector, a target detection algorithm proposed by Wei Liu at ECCV 2016, which is one of the currently popular main detection frameworks) target detection algorithm or YOLO (Youonly look Once, the latest version has been developed to V5, and it is also widely used in the industry. Its basic principle is: first divide the input image into 7×7 grids, predict 2 borders for each grid, then remove the target window with relatively low probability according to the threshold, and finally use the border merging method to remove the redundant window to obtain the detection result) target detection algorithm, etc., so the insect image recognition real-time processing can be performed based on the target detection algorithm to obtain the insect recognition result. Specifically, the target recognition algorithm can be exemplified by but not limited to the YOLO v4 target detection algorithm, wherein the specific model structure of the YOLO v4 target detection algorithm consists of three parts, namely the backbone network backbone, the neck network neck and the head network head. The backbone network Backbone can use the CSPDarknet53 (CSP stands for Cross Stage Partial) network for feature extraction. The neck network is composed of SPP (Spatial Pyramid Pooling block) blocks and PANet (Path Aggregation Network) networks. The former is used to increase the receptive field and separate the most important features, and the latter is used to ensure that semantic features are received from high-level layers and fine-grained features are received from low-level layers of the lateral backbone network at the same time. The head network head is based on anchor frame detection and detects three feature maps of different sizes of 13×13, 26×26 and 52×52, which are used to detect targets from large to small (here, feature maps with large sizes contain richer information, so feature maps with 52×52 sizes are used to detect small targets, and vice versa). The aforementioned insect recognition model can be trained by conventional sample training methods, so that after inputting the test image, the recognition results of whether there are insect images and their confidence prediction values ​​and other information can be output.In addition, the insect recognition model can be pre-trained on a server with rich computing resources and then deployed on the controller 7 with limited computing resources.

[0051] In step S11, considering that the import monitoring image is acquired at a high frequency, it is preferred to select a frame of the import monitoring image with the highest clarity (the specific clarity algorithm is an existing algorithm) from a plurality of consecutive frames of the import monitoring image and import it into the insect recognition model, so that it is not necessary to perform insect recognition for each frame of the acquired image, thereby achieving the purpose of saving computing resources. In addition, after the import monitoring image is cut by the region of interest (i.e., the region where the import 21 is located, since the import 21 and the camera 6 are fixed, the region of interest can be pre-delimited), the obtained region of interest image can be imported into the insect recognition model for insect recognition.

[0052] S12. If the insect identification result includes an insect marking frame, an insect image is captured in real time from the imported monitoring image according to the insect marking frame.

[0053] In step S12, if the insect identification result includes the insect marking frame, it indicates that there is an insect at the entrance 21, and vice versa, so it is necessary to further identify whether the insect is a mosquito based on the insect image. In addition, if the insect identification result includes multiple insect marking frames, the corresponding insect image can be cut for each insect marking frame (generally speaking, when the size of the entrance 21 is appropriately designed, the insect identification result will include at most one insect marking frame).

[0054] S13. Importing the insect image into a mosquito recognition model pre-trained based on a first convolutional neural network in real time, and outputting a mosquito recognition result.

[0055] In the step S13, the convolutional neural network (CNN) is a type of feedforward neural network (CNN) that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning and is often used to classify and identify input images. Therefore, the mosquito recognition model can be trained based on a certain amount of positive sample images (i.e., sample images of mosquitoes) and negative sample images (i.e., sample images of other insects) through a conventional calibration verification modeling method (its specific process includes a calibration process and a verification process of the model, that is, first by comparing the model simulation results with the measured data, and then adjusting the model parameters according to the comparison results, so that the simulation results are consistent with the actual process). In addition, the mosquito recognition model can also be pre-trained on a server with rich computing resources, and then deployed on the controller 7 with limited computing resources.

[0056] S14. If the mosquito identification result indicates that it is a mosquito, the channel switching mechanism is controlled to open the exit of the insect passage 2 and close the entrance of the release passage 3; otherwise, the channel switching mechanism is controlled to close the exit of the insect passage 2 and connect the entrance of the release passage 3 to the insect passage 2.

[0057] In step S14, since the channel switching mechanism is controlled to open the exit of the insect passage 2 and close the entrance of the release passage 3 when the mosquito identification result indicates a mosquito, the mosquito can be prevented from escaping and allowed to enter the mosquito catching chamber 4 from the mosquito attracting chamber, and since the exit of the insect passage 2 is closed and the entrance of the release passage 3 is connected to the insect passage 2 when the mosquito identification result indicates other insects other than mosquitoes, the other insects can be prevented from entering the mosquito catching chamber 4 and allowed to escape, thereby achieving the purpose of capturing and killing only mosquitoes, which is beneficial to protecting biodiversity and convenient for practical application and promotion. In addition, in order to ensure that the channel switching mechanism can be controlled in time to close the exit of the insect passage 2 and to connect the entrance of the release passage 3 to the insect passage 2, the length of the insect passage 2 needs to be positively correlated with the time required for the execution of the aforementioned steps S11 to S14, that is, the longer the time required for the execution of the aforementioned steps S11 to S14, the longer the length of the insect passage 2 needs to be.

[0058] Preferably, the mosquito attracting member adopts an exhaust port for attracting mosquito gas and is arranged on the wall of the support tube 14, wherein the support tube 14 is located inside the mosquito attracting chamber, one end of the support tube 14 is fixedly connected to the cap body 11, and the other end of the support tube 14 is vertically fixedly connected to the upper surface of the transverse partition 13 to support the cap body 11. Figures 4-5 As shown, in addition to the support tube 14, there are three support columns 15 for supporting the cap body 11 to ensure the stability of the support. The mosquito-attracting gas can be, but is not limited to, a conventional mixture of carbon dioxide gas, air, and mosquito-attracting bait odor. Specifically, it also includes a storage chamber 8, wherein the storage chamber 8 is built with a carbon dioxide gas cylinder 81 and a gas mixing chamber 82; the gas inlet end of the gas mixing chamber 82 is connected to the carbon dioxide gas cylinder 81, and the gas outlet end of the gas mixing chamber 82 is connected to the support tube 14. In this way, the mosquito-attracting gas can be continuously supplied to attract mosquitoes.

[0059] Preferably, a ventilation window is provided on the first side of the mosquito catching room 4, and a mosquito filter 42 is installed on the ventilation window, and an exhaust fan 43 for exhausting indoor air is also installed outside the ventilation window, wherein the controlled end of the exhaust fan 43 is communicatively connected to the controller 7; the controller 7 is also used to start and stop the exhaust fan 43 and / or adjust the speed (for example, when the mosquito identification result indicates a mosquito, the speed of the exhaust fan 43 is increased, and when the mosquito identification result indicates a non-mosquito, the speed of the exhaust fan 43 is slowed down). Figures 3-4 As shown, specifically, the ventilation window is provided on the rear side of the mosquito catching chamber 4. The mosquito filter 42 is used to prevent indoor mosquitoes from escaping. The exhaust fan 43 is used to generate sufficient indoor negative pressure after being started, so as to suck the mosquitoes passing through the inlet 21 from the mosquito attracting chamber into the mosquito catching chamber 4, and prevent indoor mosquitoes from escaping from the mosquito catching chamber to the mosquito attracting chamber through the insect passage 3. Specifically, the ventilation window is provided on the vertical partition between the mosquito catching chamber 4 and the storage chamber 8, and the exhaust fan 43 is located in the storage chamber 8, and a plurality of air holes 83 are provided on the side wall of the storage chamber 8, so as to ensure smooth exhaust airflow while making the entire intelligent mosquito catching device neat and beautiful. The exhaust fan 43 is always running after being turned on. In order to further ensure smooth exhaust airflow, the outlet of the insect passage 2 can be set to be normally open, and the inlet of the release passage 3 can be set to be normally closed, so that when the mosquito identification result indicates that it is not a mosquito, the channel switching mechanism is controlled to close the outlet of the insect passage 2 and the inlet of the release passage 3 is connected to the insect passage 2. In addition, in order to ensure that the channel switching mechanism can be controlled in time to close the outlet of the insect passage 2 and the inlet of the release passage 3 is connected to the insect passage 2, the length of the insect passage 2 also needs to be positively correlated with the rotation speed of the exhaust fan 43, that is, the faster the rotation speed of the exhaust fan 43, the longer the length of the insect passage 2 needs to be increased accordingly.

[0060] Preferably, a drawer 44 and / or a maintenance door 45 for opening the interior is installed on the second side of the mosquito catching chamber 4, wherein the drawer 44 is located at the bottom of the mosquito catching chamber 4. Figures 4-5 As shown, the drawer 44 and the maintenance door 45 are installed on the main side of the mosquito catching chamber 4, wherein the drawer 44 is used to contain the killed indoor mosquitoes, so it can be pulled out at any time for cleaning; the maintenance door 45 is used to be opened at any time to repair the mosquito filter 42 or other indoor components.

[0061] Preferably, the controller 7 is also used to upload the insect image and the collection time and / or collection location of the insect image to the cloud server in real time when the mosquito identification result indicates a mosquito, so that the cloud server imports the insect image into the mosquito classification model pre-trained based on the second convolutional neural network, outputs the mosquito classification result, and performs statistical analysis based on the mosquito classification result and the collection time and / or the collection location to obtain a mosquito statistical analysis report. The classification capability of the mosquito classification model may include but is not limited to the classification capability of mosquito species, the classification capability of mosquito growth stage and / or the classification capability of mosquito sex, etc. Therefore, the mosquito statistical analysis report specifically includes but is not limited to the statistical analysis results of mosquito species based on time latitude and / or geographical latitude, the statistical analysis results of mosquito growth stage and the statistical analysis results of mosquito sex, etc., which can help to make appropriate mosquito prevention and control strategies. Since the classification ability of the mosquito classification model is stronger than that of the mosquito recognition model, the second convolutional neural network will be more complex than the first convolutional neural network (for example, the second convolutional neural network adopts the existing RCNN network or ResNet network) and requires more computing resources, and thus needs to be deployed on the cloud server to complete the mosquito classification task. In addition, the mosquito classification model can also be trained based on a certain amount of sample images and sample labels through conventional calibration verification modeling methods.

[0062] In summary, the intelligent mosquito catching device provided by this embodiment has the following technical effects:

[0063] (1) This embodiment provides a new mosquito catching scheme for differentially catching and killing mosquitoes and beneficial insects based on a target recognition algorithm and a convolutional neural network, that is, it includes a mosquito catching chamber, an insect passage, a release passage, a mosquito catching chamber, a passage switching mechanism, a camera and a controller. The controller is used to first identify whether there is an insect at the entrance of the insect passage based on a target recognition algorithm after receiving an inlet monitoring image. If so, it intercepts an insect image and identifies whether the insect is a mosquito based on a convolutional neural network. If it is a mosquito, it controls the passage switching mechanism to open the exit of the insect passage and close the entrance of the release passage. Otherwise, it controls the passage switching mechanism to close the exit of the insect passage and connect the entrance of the release passage to the insect passage. In this way, not only can mosquitoes be prevented from escaping and entering the mosquito catching chamber from the mosquito catching chamber, but other insects can also be prevented from entering the mosquito catching chamber and allowing them to escape. In this way, the purpose of capturing and killing only mosquitoes can be achieved, which is beneficial to protecting biodiversity.

[0064] (2) The mosquito images and the image acquisition time and / or image acquisition location can also be uploaded to the cloud server to further obtain mosquito classification results based on the convolutional neural network, and statistical analysis can be performed in combination with the image acquisition time and / or image acquisition location to obtain a mosquito statistical analysis report, which is conducive to making appropriate mosquito control strategies and facilitating practical application and promotion.

[0065] Embodiment 2

[0066] Based on the technical solution of the first embodiment, this embodiment further provides a new intelligent mosquito catching solution in which the carbon dioxide gas cylinder 81 is replaced by a carbon dioxide gas enricher, that is, Figures 6 to 9 As shown, the carbon dioxide gas enricher for replacing the carbon dioxide gas cylinder 81 includes but is not limited to an upper bottle body 100, a lower bottle body 200, a connecting pipe 300, a bottle body upside-down turning mechanism 400, an air supply pump 500 and a high-temperature filter 600, wherein the bottle mouth of the upper bottle body 100 and the bottle mouth of the lower bottle body 200 are arranged opposite to each other up and down and are connected to each other through the connecting pipe 300, and the interior of the lower bottle body 200 contains a carbon dioxide adsorbent solution; the bottle body upside-down turning mechanism 400 is used to upside down the upper bottle body 100, the lower bottle body 200 and the connecting pipe The bottle body assembly structure is composed of 300; the output port of the air supply pump 500 is connected to the bottom of the inner cavity of the lower bottle body 200, so as to introduce the air outside the bottle into the carbon dioxide adsorbent solution, so that the carbon dioxide gas in the air reacts with the carbon dioxide adsorbent in the carbon dioxide adsorbent solution to form water-insoluble carbonate crystals; the high-temperature filter 600 is arranged inside the connecting pipe 300, so as to intercept the carbonate crystals when the bottle body assembly structure is turned upside down, and decompose the carbonate crystals at high temperature to obtain carbon dioxide gas and the carbon dioxide adsorbent that can fall down by itself.

[0067] like Figures 6 to 9 As shown, in the specific structure of the carbon dioxide gas enricher, the upper bottle body 100 and the lower bottle body 200 are a pair of bottles that can be interchanged by turning upside down, so as to contain the carbon dioxide adsorbent solution in time; Figures 6-7 As shown, specifically, the upper bottle body 100 and / or the lower bottle body 200 can be, but not limited to, a conical flask. The connecting tube 300 is used to connect the upper bottle body 100 and the lower bottle body 200, so that the carbon dioxide adsorbent solution can flow from one bottle body to another bottle body during the upside-down flipping process. In order to be able to export the air from which the carbon dioxide gas has been adsorbed and the carbon dioxide gas obtained by high-temperature decomposition, specifically, an air outlet 310 is opened on the side wall of the connecting tube 300; and in order to further be able to export the air from which the carbon dioxide gas has been adsorbed and the carbon dioxide gas obtained by high-temperature decomposition to different flow directions at different time periods, the air outlet 310 can be connected to the air inlet port of a three-way valve, so that when the air outside the bottle is passed into the carbon dioxide adsorbent solution, the air inlet port of the three-way valve and the first air outlet port of the three-way valve are selected to be connected, so as to realize the air from which the carbon dioxide gas has been adsorbed. The purpose of the air of carbon dioxide gas is to be led out to the first flow direction, and when the carbonic acid crystals are decomposed at high temperature, the air inlet port of the three-way valve and the second air outlet port of the three-way valve are connected to achieve the purpose of leading the carbon dioxide gas obtained by high temperature decomposition to the second flow direction (for example, the flow direction leading to the mosquito killer, that is, the second air outlet port of the three-way valve is connected to the air inlet end of the mixing chamber 82); at the same time, in order to prevent the carbon dioxide adsorbent solution from leaking out through the air outlet 310 during the upside-down turning process, a gas separation membrane is preferably arranged in the air outlet 310. The air supply pump 500 can be implemented by, but not limited to, an existing air pump. The high temperature in the high temperature filter 600 is a temperature that is relatively high compared to normal temperature and needs to reach the decomposition temperature point of the carbonic acid crystals (for example, above 35 degrees Celsius), which generally needs to be heated after intercepting the carbonic acid crystals, that is, the high temperature filter 600 can be, but not limited to, woven by electric heating wire / and non-electric heating wire (specifically, woven by electric heating wire, or woven by electric heating wire and non-electric heating wire), so that the high temperature can be obtained by heating the electric heating wire. In addition, the carbon dioxide adsorbent may specifically be, but is not limited to, guanidine or meta-xylylenediamine.

[0068] The working principle of the aforementioned carbon dioxide gas enricher may include, but is not limited to, the following: (1) introducing air outside the bottle into the carbon dioxide adsorbent solution through the air supply pump 500, so that the carbon dioxide gas in the air reacts with the carbon dioxide adsorbent in the carbon dioxide adsorbent solution to form water-insoluble carbonate crystals; (2) after a period of time, turning off the air supply pump 500, and flipping the bottle assembly structure upside down through the bottle body upside down mechanism 400, so that the upper bottle body 100 and the lower bottle body 200 are interchanged, and the carbon dioxide adsorbent solution flows into the new lower bottle body after the identity change (i.e., the upper bottle body 100 before the identity change) after passing through the high-temperature filter 600. At this time, due to the The high temperature filter 600 intercepts the carbonate crystals; (3) the carbonate crystals are decomposed by the high temperature filter 600 to obtain carbon dioxide gas and carbon dioxide adsorbent that can fall down by itself (i.e., the carbon dioxide adsorbent solution that falls down and dissolves to obtain a carbon dioxide adsorbent solution with a higher concentration). Although there is air in the bottle at this time, the carbon dioxide gas obtained by decomposition is adsorbed from a large amount of air outside the bottle. Therefore, after the carbon dioxide gas obtained by decomposition is mixed with the air in the bottle, the purpose of carbon dioxide gas enrichment can still be achieved (note that the purpose of carbon dioxide gas purification is not achieved here); (4) the air outside the bottle is introduced into the carbon dioxide adsorbent solution through the air supply pump 500 again to start the next cycle. Therefore, the aforementioned carbon dioxide gas enricher provided by this embodiment can greatly simplify the air separation system and reduce the system volume while achieving the purpose of carbon dioxide gas enrichment, so that it can be applied to small equipment such as mosquito killing devices that use the tendency of mosquitoes to carbon dioxide gas to lure and kill mosquitoes, which is convenient for practical application and promotion.

[0069] Preferably, the bottle body upside-down flipping mechanism 400 includes but is not limited to a bracket 410, a rotating shaft 420, a bearing 430 and a motor 440, wherein the rotating shaft 420 is fixed and horizontally passes through the middle of the connecting pipe 300; the rotating shaft 420 is horizontally installed on the top of the bracket 410 through the bearing 430, and the bottle body assembly structure is suspended; the output shaft of the motor 440 is connected to the rotating shaft 420. Figures 6 to 9 As shown, the purpose of easily flipping the bottle body assembly structure can be achieved through the specific design of the above-mentioned bottle body upside down flipping mechanism. In order to further improve the stability during the flipping process, it is further preferred that the bracket 410 includes but is not limited to a bottom plate 411 and an inverted "V"-shaped support rod 412, wherein the two ends of the inverted "V"-shaped support rod 412 are respectively fixedly connected to the upper surface of the bottom plate 411, so that the middle inflection point of the inverted "V"-shaped support rod 412 forms the top of the bracket 410; the number of the inverted "V"-shaped support rod 412 can be as follows: Figure 6 There are two as shown, but there may be one or more. Also in order to further improve the stability during the flipping process, it is further preferred that the bearing 430 includes but is not limited to a first bearing 431 and a second bearing 432; one end of the rotating shaft 420 is installed on the top side of the bracket 410 through the first bearing 431, and the other end of the rotating shaft 420 is installed on the other side of the top of the bracket 410 through the second bearing 432, so that the bottle body assembly structure is suspended between the top side of the bracket 410 and the other side of the top of the bracket 410. The motor 440 can be as follows Figure 6 The motor 440 is installed on the top of the bracket 410 as shown, and can also be installed on the bottom plate 411. In addition, the motor 440 can be implemented by a stepper motor but is not limited to it.

[0070] Preferably, the number of the air supply pumps 500 is two, and their output ports are connected to the bottom of the inner cavities of the two bottles that are the lower bottles in the bottle assembly structure in a one-to-one correspondence. Figures 6 to 9 As shown, based on the characteristic that the upper bottle body 100 and the lower bottle body 200 are a pair of bottles that can be interchanged by turning upside down, carbon dioxide gas can be continuously enriched to improve the enrichment efficiency.

[0071] Preferably, a control module 700 is also included, wherein the output end of the control module 700 is respectively connected to the controlled end of the bottle body upside down flip mechanism 400 (specifically the controlled end of the motor 440) and the controlled end of the air supply pump 500. In this way, the flipping action of the bottle body upside down flip mechanism 400 and the start and stop action of the air supply pump 500 can be controlled by the control module 700 (specifically realized by the internal conventional control circuit), thereby facilitating the realization of the intelligent working purpose. In addition, the output end of the control module 700 can also be connected to the controlled end of the power supply module of the electric heating wire and the controlled end of the three-way valve, so that the heating start and stop action of the electric heating wire and the channel switching action of the three-way valve are also controlled by the control module 700, further facilitating the realization of the intelligent working purpose.

[0072] In summary, in addition to the technical effects of the first embodiment, this embodiment also has the following technical effects: (1) by replacing the carbon dioxide cylinder with a carbon dioxide gas enricher, not only can the purpose of carbon dioxide gas enrichment be achieved to continuously supply mosquito-attracting gas, but the air separation system can also be greatly simplified and the system volume can be reduced, which is conducive to the miniaturized design of the entire intelligent mosquito catching device, further facilitating practical application and promotion.

[0073] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent mosquito catching device, characterized in that: The invention comprises a mosquito attracting chamber, an insect passage (2), a releasing passage (3), a mosquito catching chamber (4), a passage switching mechanism, a camera (6) and a controller (7), wherein the inlet (21) of the insect passage (2) is connected to the mosquito attracting chamber, the outlet of the insect passage (2) is connected to the mosquito catching chamber (4), the inlet of the releasing passage (3) is connected to the insect passage (2) through the passage switching mechanism, and the outlet (32) of the releasing passage (3) is connected to the outside of the room; The channel switching mechanism is communicatively connected to the controller (7) and is used to, under the control of the controller (7), close the exit of the insect passage (2) and connect the entrance of the release passage (3) to the insect passage (2), or open the exit of the insect passage (2) and close the entrance of the release passage (3); The camera (6) is communicatively connected to the controller (7) and is used to collect images of the entrance (21) of the insect passage (2) in real time, and transmit the acquired entrance monitoring images to the controller (7) in real time; The controller (7) is used to perform intelligent mosquito catching according to the following steps: When the imported monitoring image is received, the imported monitoring image is imported in real time into a pre-trained insect recognition model based on a target recognition algorithm, and an insect recognition result is output; If the insect identification result includes an insect marking frame, an insect image is captured in real time from the imported monitoring image according to the insect marking frame; Importing the insect image into a mosquito recognition model pre-trained based on a first convolutional neural network in real time, and outputting a mosquito recognition result; If the mosquito identification result indicates that it is a mosquito, the channel switching mechanism is controlled to open the exit of the insect passage (2) and close the entrance of the release passage (3); otherwise, the channel switching mechanism is controlled to close the exit of the insect passage (2) and connect the entrance of the release passage (3) to the insect passage (2).

2. The intelligent mosquito catching device according to claim 1, characterized in that: The mosquito attracting chamber is surrounded by a cap body (11), a bird-proof net (12) and a transverse partition (13) arranged from top to bottom, and has a built-in mosquito attracting member, wherein the upper surface of the transverse partition (13) is provided with an entrance (21) of the insect passage (2).

3. The intelligent mosquito catching device according to claim 2, characterized in that: The mosquito attracting member adopts an exhaust port for mosquito attracting gas and is arranged on the tube wall of a support tube (14), wherein the support tube (14) is located inside the mosquito attracting chamber, one end of the support tube (14) is fixedly connected to the cap body (11), and the other end of the support tube (14) is vertically fixedly connected to the upper surface of the transverse partition (13) so as to support the cap body (11).

4. The intelligent mosquito catching device according to claim 3, characterized in that: It also includes a storage chamber (8), wherein the storage chamber (8) has a carbon dioxide gas cylinder (81) and a gas mixing chamber (82) built therein; The gas inlet end of the gas mixing chamber (82) is connected to the carbon dioxide gas cylinder (81), and the gas outlet end of the gas mixing chamber (82) is connected to the support tube (14).

5. The intelligent mosquito catching device according to claim 2, characterized in that: The camera (6) is hung upside down on the top of the cap body (11), and its lens is aimed at the entrance (21) of the insect passage (2).

6. The intelligent mosquito catching device according to claim 1, characterized in that: A ventilation window is provided on a first side surface of the mosquito trapping chamber (4), a mosquito filter (42) is installed on the ventilation window, and an exhaust fan (43) for exhausting indoor air is installed outside the ventilation window, wherein a controlled end of the exhaust fan (43) is communicatively connected to the controller (7); The controller (7) is also used to perform start and stop control and speed regulation control on the exhaust fan (43).

7. The intelligent mosquito catching device according to claim 1, characterized in that: A drawer (44) and / or a maintenance door (45) for opening the interior of the mosquito catching room (4) are installed on the second side of the mosquito catching room (4), wherein the drawer (44) is located at the bottom of the mosquito catching room (4).

8. The intelligent mosquito catching device according to claim 1, characterized in that: The channel switching mechanism comprises a stepping motor and a rotating flap (52), wherein the controlled end of the stepping motor is communicatively connected to the controller (7), and the rotating flap (52) is axially connected to the output shaft (510) of the stepping motor; The rotating flap (52) is used to close the outlet of the insect passage (2) and connect the inlet of the release passage (3) to the insect passage (2) when the output shaft (510) rotates to a first position, and to open the outlet of the insect passage (2) and close the inlet of the release passage (3) when the output shaft (510) rotates to a second position.

9. The intelligent mosquito catching device according to claim 1, characterized in that: The controller (7) is also used to upload the insect image and the time and / or location of collection of the insect image to a cloud server in real time when the mosquito identification result indicates a mosquito, so that the cloud server imports the insect image into a mosquito classification model pre-trained based on a second convolutional neural network, outputs a mosquito classification result, and performs statistical analysis based on the mosquito classification result and the time and / or location of collection to obtain a mosquito statistical analysis report.

10. The intelligent mosquito catching device according to claim 1, characterized in that: The camera (6) is a high-speed camera, and the controller (7) is a Raspberry Pi microcomputer.

Images