Intelligent continuous monitoring terminal for rats

Abstract

The utility model discloses an intelligent continuous monitoring terminal for mice, which belongs to the technical field of rodent damage monitoring and comprises an instrument host, a mouse bin is detachably communicated with the instrument host, mouse bait and a camera are arranged on the instrument host and / or the mouse bin, and a temperature and humidity sensor, a communication module, a single chip microcomputer and a power supply device are arranged on the instrument host. A display screen used for displaying the number of trapped mice is fixedly arranged on the instrument host. According to the structure, a mouse enters the mouse bin through the instrument host by utilizing the luring effect of the mouse bait, so that the mouse is continuously caught, the activity of the mouse can be kept, and the camera, the humidity sensor, the communication module, the power supply device and the display screen are controlled by utilizing the single chip microcomputer to realize continuous monitoring of the mouse; therefore, the user can check the monitoring result of the mouse locally and / or remotely at any time.

Description

Technical Field

[0001] The utility model relates to the technical field of rodent damage monitoring, in particular to an intelligent continuous monitoring terminal for rodents. Background Technique

[0002] An invention application with Chinese patent number CN201811330558.0 discloses a rodent density monitoring method, monitoring system, and an invention application with Chinese patent number CN202011615751.6 discloses a rodent situation monitoring method, system, and intelligent terminal. The above systems propose using a high-definition camera to capture the traces of mice as a monitoring means. Another example is a rodent control method disclosed in a Chinese invention patent with patent number CN202111318164.5, which monitors the rodent density by injecting a marker (and then driving away) the mice. However, the above rodent damage monitoring methods cannot capture mice. Another example is an invention application with Chinese patent number CN201810231158.8, which discloses a mouse-catching device and a rodent monitoring system. Although it can catch mice, it cannot catch mice continuously, and lacks a computer automatic recognition and automatic warning system. Another example is an invention application with Chinese patent number CN202310358880.9, which discloses a rodent intelligent monitoring device and monitoring method, which uses an intelligent mouse trap to catch mice. However, the mouse trap can only catch one mouse at a time, and the caught mouse may die quickly, and the parasites and microorganisms carried on its body will be destroyed, making it impossible to further study it. In addition, if the mouse body is not disposed of in time, it may stink and cause secondary environmental pollution. Another example is a utility model patent with Chinese patent number CN202121555943.2, which discloses an intelligent mouse detection device, which can perform intelligent detection on mice, but there is no backend recognition and display module, resulting in users being unable to judge the type of mice in time. Therefore, it is necessary to further improve. Content of the Utility Model

[0003] The utility model aims to provide an intelligent continuous monitoring terminal for rodents to overcome the deficiencies in the prior art.

[0004] An intelligent continuous monitoring terminal for rodents designed according to this purpose includes an instrument host. A mouse bin is detachably connected to the instrument host. Mouse baits and cameras are provided on the instrument host and / or the mouse bin. A temperature and humidity sensor, a communication module, a single-chip microcomputer, and a power supply device are provided on the instrument host. A display screen for displaying the number of captured mice is fixedly provided on the instrument host.

[0005] A host channel is provided on the instrument host, and a mouse bin channel is provided on the mouse bin. When the mouse bin is assembled on the instrument host, the mouse bin channel communicates with the host channel.

[0006] There is a host channel opening on the outside of the host channel, and a one-way door for the host channel is arranged inside. There is a one-way door for the mouse bin channel inside the mouse bin channel. The mouse bin channel and the host channel are interconnected through the one-way door for the host channel and the one-way door for the mouse bin channel.

[0007] The camera is arranged inside the host channel, and a fill light is also arranged on it. An infrared sensor is also arranged inside the host channel.

[0008] The single-chip microcomputer is electrically connected to the temperature and humidity sensor, the communication module, the display screen, the camera, the power supply device, and the infrared sensor respectively.

[0009] The power supply device is a rechargeable battery, and the instrument host is connected with a solar panel or a power adapter corresponding to the power supply device.

[0010] A shielding plate is arranged on the top of the instrument host, and the periphery of the shielding plate extends towards the outside of the instrument host. The solar panel is fixedly arranged on the shielding plate or externally arranged outside the instrument host.

[0011] The instrument host and the mouse bin are detachably assembled and matched with each other through buckles, fasteners, or magnetic attraction.

[0012] Ventilation holes are arranged on the side of the mouse bin.

[0013] A handle is arranged on the top of the mouse bin.

[0014] The utility model utilizes the attracting effect of the mouse bait to enable the mouse to enter the mouse bin through the instrument host, thereby realizing the continuous capture of the mouse and keeping the activity of the mouse. The single-chip microcomputer is used to control the camera, humidity sensor, communication module, power supply device, and display screen, so as to take pictures of the mouse when it passes through the instrument host, enable the humidity sensor to sense the temperature and humidity information in the environment where the instrument host is located, enable the power supply device to provide working power for the instrument host, the display screen to display the number of mice captured in the mouse bin, and the communication module to enable the instrument host to remotely connect to the cloud database, display and warning module, so as to transmit the high-definition pictures and environmental information to the cloud database. The cloud database then analyzes and stores the high-definition pictures and environmental information, and the display and warning module displays and warns the analysis results and stored information, so as to facilitate the user to view the monitoring results of rodents locally and / or remotely at any time. Description of the Drawings

[0015] Figure 1 It is a front structural schematic diagram of an embodiment of the utility model.

[0016] Figure 2 It is a side structural schematic diagram of an embodiment of the utility model.

[0017] Figure 3 It is a cooperation schematic diagram of the intelligent terminal, cloud database, and display and warning module.

[0018] Figure 4 This is a working diagram of the improved YOLOV5 algorithm.

[0019] Figure 5 , Figure 6 This is a working diagram of the NMS non-maximum suppression algorithm.

[0020] Figure 7 This is a schematic diagram of the control interface of this display and warning module. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned purposes, features and advantages of the utility model more obvious and easy to understand, the specific implementation methods of the utility model are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the utility model. However, the utility model can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the utility model, so the utility model is not limited by the specific embodiments disclosed below.

[0022] The utility model is further described below in conjunction with the accompanying drawings and embodiments.

[0023] See also Figures 1 - 3 The intelligent continuous monitoring terminal for rodents includes an instrument host 12, which is detachably connected to a mouse bin 13, the instrument host 12 and / or the mouse bin 13 are provided with mouse bait 18 and a camera 6, the instrument host 12 is provided with a temperature and humidity sensor 2, a communication module 3, a single-chip microcomputer 4 and a power supply device 7, and the instrument host 12 is fixedly provided with a display screen 5 for displaying the number of mice captured.

[0024] This embodiment utilizes the luring effect of mouse bait 18 to make mice enter the mouse bin 13 through the instrument host 12, thereby realizing continuous capture of mice and maintaining the activity of mice, and uses the single-chip computer 4 to control the camera 6, humidity sensor 2, communication module 3, power supply device 7, and display screen 5, so that mice are photographed when passing through the instrument host 12, so that the humidity sensor 2 can sense the temperature and humidity information of the environment in which the instrument host 12 is located, the power supply device 7 can provide working electricity for the instrument host 12, and the display screen 5 displays the number of mice captured in the mouse bin 13. The communication module 3 can enable the instrument host 12 to remotely connect to the cloud database, display and early warning module to transmit high-definition pictures and environmental information to the cloud database, and the cloud database then analyzes and stores the high-definition pictures and environmental information, and the display and early warning module displays and warns the analysis structure and stored information, so that users can view the monitoring results of rodents locally and / or remotely at any time.

[0025] The main instrument body 12 is provided with a main body channel 20, and the mouse chamber 13 is provided with a mouse chamber channel 17. When the mouse chamber 13 is assembled on the main instrument body 12, the mouse chamber channel 17 communicates with the main body channel 20.

[0026] An opening 9 of the main body channel 20 is arranged on the outside, and a one-way door 8 of the main body channel is arranged inside. A one-way door 21 of the mouse chamber channel is arranged inside the mouse chamber channel 17. The mouse chamber channel 17 and the main body channel 20 communicate with each other through the one-way door 8 of the main body channel and the one-way door 21 of the mouse chamber channel.

[0027] In this embodiment, a one-way door 8 of the main body channel and a one-way door 21 of the mouse chamber channel are respectively arranged inside the main body channel 20 and the mouse chamber channel 17. Therefore, when a mouse enters the main body channel 20 through the opening 9 of the main body channel and passes through the one-way door 8 of the main body channel and the one-way door 21 of the mouse chamber channel respectively, it cannot escape from the main body channel 20 and the mouse chamber channel 17, improving the capture stability of the mouse.

[0028] The camera 6 is arranged inside the main body channel 20 and a fill light is also arranged thereon. When the camera 6 works, it uses the fill light to illuminate the main body channel 20 to improve the clarity of the camera 6's photographing.

[0029] An infrared sensor 10 is also arranged inside the main body channel 20. When a mouse enters the main body channel 20, it will be sensed by the infrared sensor 10, causing the main instrument body 12 to work. This not only improves the working stability of the intelligent terminal but also avoids energy consumption waste when no mouse enters the intelligent terminal.

[0030] The single-chip microcomputer 4 is electrically connected to the temperature and humidity sensor 2, the communication module 3, the display screen 5, the camera 6, the power supply device 7, and the infrared sensor 10 respectively.

[0031] The power supply device 7 is a rechargeable battery. The main instrument body 12 is connected with a solar panel 19 or a power adapter corresponding to the power supply device 7. The power supply device 7 can be charged through the solar panel 19 or by plugging in the power adapter.

[0032] A shielding plate 22 is arranged on the top of the main instrument body 12. The periphery of the shielding plate 22 extends towards the outside of the main instrument body 12. Thus, the size of the shielding plate 22 is larger than that of the main instrument body 12 to shade and protect the main instrument body 12 from rain. In addition, the cross-section of the shielding plate 22 is fan-shaped to enable rainwater to fall naturally.

[0033] The solar panel 19 is fixedly arranged on the shielding plate 22 or externally arranged outside the main instrument body 12.

[0034] In this embodiment, the solar panel 19 is independently arranged outside the instrument main body 12 and electrically connected to the power supply device 7, so that the power supply device 7 can be automatically charged to ensure the long-term stable operation of the instrument main body 12.

[0035] The instrument main body 12 and the mouse bin 13 are detachably and assembled with each other through a buckle 14, a fastener, or magnetic attraction.

[0036] In this embodiment, a buckle 14 is provided between the instrument main body 12 and the mouse bin 13, and the two are quickly disassembled and assembled through the buckle 14 to facilitate the replacement of the mouse bin 13. In addition, the size inside the mouse bin 13 determines the number of live mice that can be accommodated, generally 3-10.

[0037] Vent holes 15 are provided on the side of the mouse bin 13. In order to ensure the survival of the mice inside the mouse bin 13, vent holes 15 are provided on the side of the mouse bin 13 to facilitate the air circulation inside and outside the mouse bin 13.

[0038] A handle 11 is provided on the top of the mouse bin 13. When the number of mice reaches a certain amount or as needed, the user can easily remove the mouse bin 13 containing live mice through the handle 11 and replace it with a new mouse bin 13.

[0039] In this embodiment, the cloud database has a storage module and an identification module. The storage module stores pictures of different types of mice and forms an original data set. The identification module automatically reads the original data set using the improved YOLOV5 algorithm and classifies and trains the mouse features using the large data set. Before training, the identification module at least performs data enhancement processing on the data using image segmentation and / or random rotation methods. When the identification module receives a new image from the intelligent terminal, it also performs data enhancement processing on the new image using the image segmentation method, calls the pre-trained result for AI identification, merges the prediction results of each region and the prediction result of the whole image after identification, then uses the NMS non-maximum suppression algorithm to filter the results, anchors the target object and performs classification statistics, and then transmits the relevant information of the identified picture and the classification quantity back to the storage module for storage.

[0040] A display and warning module, which includes a statistical number display module and a warning module. The statistical number display module is communicatively connected to the cloud database, automatically reads the information of the cloud database, automatically calculates the capture rate of mice, the types of mice, and the density of mice in the environment where the intelligent terminal is located according to the information of the cloud database, and automatically generates a warning message for display. The warning module processes the warning message and gives a warning when the warning message exceeds the expected value.

[0041] The monitoring terminal in this embodiment continuously captures mice through bait, can keep the mice alive, takes pictures of the mice, and then transmits the high-definition pictures to the cloud database. In addition, the captured live mice can be collected for the user to analyze and study. The recognition module of the cloud database uses the improved YOLOV5 algorithm, regularly calls the API interface from the storage module, judges the picture size, decides whether to intervene in training and recognition for the image segmentation method, automatically counts after identifying the types of mice, and writes the relevant data into the storage module. The display and warning module analyzes according to the information in the cloud database and displays and warns in real time, and the user can view the warning information locally and / or remotely at any time.

[0042] The improved YOLOV5 algorithm is as follows: when the image is greater than or equal to [1280, 1280], or the training / detection target area is less than or equal to 1% of the entire picture, or there are multiple training / prediction targets, use the image segmentation method for training and prediction, use the NMS non-maximum suppression algorithm to filter the results, the size of the side length parameter slice_size of the cut sub-image is [400, 400], and the overlap ratio between sub-images is set by the parameter overlap_ratio = 0.4.

[0043] In this embodiment, the improved YOLOV5 algorithm includes a backbone main network, a neck, and a detect detection head. Its default input value is [640, 640], and the input size of the picture during training, parallel computing, etc. are related to the target size and picture size.

[0044] The NMS non-maximum suppression algorithm is as follows: calculate the intersection over union IOU between slices, IOU = the overlapping part of two pictures / (the sum of two pictures - the overlapping part). Among them, the NMS non-maximum suppression algorithm relies on the classifier to obtain multiple candidate boxes and the probability values of the candidate boxes belonging to the category, and sorts according to the category classification probability obtained by the classifier.

[0045] Suppose there is a set B of candidate BOXES and its corresponding set S of SCORES, and the following steps are used for processing:

[0046] 1) Find the box M with the highest score;

[0047] 2) Delete the BOX corresponding to M from B;

[0048] 3) Add the deleted BOX to the set D;

[0049] 4) Delete other boxes in B whose overlapping area with the BOX corresponding to M is greater than the threshold Nt;

[0050] 5) Repeat the above steps 1 to 4;

[0051] If it exceeds the set threshold, it is considered that the objects inside the two boxes belong to the same category, for example, they both belong to the category of mice. We only need to keep the box diagram with the highest probability of one category.

[0052] The above is the NMS non-maximum suppression algorithm.

[0053] The formula representation of the NMS non-maximum suppression algorithm is:

[0054]

[0055] Among them, S is the score of the box, M is the box with the highest current score, Bi is the box to be processed, and the larger the IOU between Bi and M, the more severely the score of Bi drops. Nt is the threshold, and Nt is generally set to 0.

[0056] The warning information at least includes: the relationship information between the mouse activity frequency in the current environment where the monitoring terminal is located and the factors of the current season, temperature, humidity, and wind speed, the mouse species distribution information at different latitudes and longitudes in the historical environment where the monitoring terminal is located, and the dominant mouse population information.

[0057] The display forms of the display and warning module at least include: statistical table display, statistical graph display, change curve display, and visual display.

[0058] In this embodiment, the control interface of the display and warning module is as Figure 7 shown, with monitoring options at the top, a warning information display area in the middle, and data management options at the bottom.

[0059] Among them, the monitoring options include adult mosquito monitoring, mosquito egg monitoring, cockroach monitoring, fly monitoring, and mouse monitoring from left to right. Thus, users can select the corresponding monitoring through the monitoring options.

[0060] The warning information display area has an instrument distribution map, a density schematic diagram, a daily change curve, a ranking in the first area, and a ranking in the second area. Thus, users can view the relevant monitoring information through the warning information display area.

[0061] The data management options include exporting pictures and detailed data analysis. Thus, users can export the photographed pictures and analyze and view the more detailed monitoring data.

[0062] There is also a view warning record on the right side of the monitoring options. Thus, users can view the warning information in the historical records.

[0063] There is an existing old YOLOV5 algorithm (select the YOLOV5X version, with about 86 million parameters). Whether for training or prediction, the original image is reshaped to [640, 640] in size, and padding = 0. The improved YOLOV5 algorithm includes judging the size of the image, the proportion of the training / recognition target, or there are multiple target objects in the image to be recognized. At this time, both the training set and the recognition target need to be cut for training / recognition. After the recognition, the cut images are filtered using the NMS non-maximum suppression algorithm. Since the monitoring terminal uses a high-definition camera 6, the size of the captured images is high-definition images with a resolution of [4032, 3024] and above. In order to make the recognition module have high accuracy and stability, and also make the recognition module have a certain degree of generalization, both training and recognition use the improved YOLOV5 algorithm, and the image segmentation method is introduced to cut and process the data.

[0064] As Figure 5 shown, the original image belongs to a high-definition [4032, 3024] and multi-target image. The improved YOLOV5 algorithm needs to perform cut training / recognition. The cut parameters are: slice_size = [400, 400], overlap_ratio = 0.4. As Figure 6 shown, the sizes of the solid / double-dashed / dotted boxes are all [400, 400]. The left and right overlap ratio of the solid box and the double-dashed box is 40%, and the up and down overlap ratio of the solid box and the dotted box is 40%. The entire original image is cut into more than a hundred sub-images. In Figure 6 , for the first mouse on the left, the probability of the dotted box is greater than that of the solid and double-dashed boxes. The dotted box is retained, and the solid / double-dashed boxes are suppressed during merging (NMS non-maximum suppression algorithm). Similarly, when recognizing the second mouse from the left, the probability of the double-dashed box is greater than that of the solid and dotted boxes. The double-dashed box is retained, and the solid / dotted boxes are suppressed (NMS non-maximum suppression algorithm) and not displayed. When the window slides across the entire original image, all the mice are accurately recognized.

[0065] For mosquitoes, mosquito eggs, etc. which are inherently tiny objects, the improved YOLOV5 algorithm is more suitable for the training and recognition of small and medium-sized objects such as mice.

[0066] In this embodiment, the image segmentation method is the SAHI image segmentation method.

[0067] The above-mentioned intelligent continuous monitoring method for mice is as follows:

[0068] The monitoring terminal continuously captures mice through the mouse bait 18, can keep the mice alive in the mouse bin 13, takes pictures of the mice, and then transmits the high-definition pictures to the cloud database. After a certain number of captured mice or according to the monitoring requirements, the mouse bin 13 is replaced. All the mice in the replaced mouse bin 13 are live mice, and users can take them back to the laboratory for further research, such as the pathogenic fleas, pathogenic bacteria, etc. carried by the mice.

[0069] The recognition module uses the improved YOLOV5 algorithm, regularly calls the API interface from the cloud database, judges the picture size, decides whether to intervene in training and recognition for the image segmentation method, automatically counts after recognizing the types of mice, and writes the relevant data into the cloud database.

[0070] The display and warning module analyzes according to the information in the cloud database and then displays and warns in real time. Users can view the warning information and receive warnings locally and / or remotely at any time.

[0071] Specifically, the mice are attracted by the mouse bait 18, enter the main channel 20 through the opening 9 of the main channel, and then trigger the infrared sensor 10 after passing through the one-way door 8 of the main channel. The infrared sensor 10 sends a working signal to the single-chip microcomputer 4. After receiving the working signal, the single-chip microcomputer 4 controls the camera 6 to take pictures of the mice in the main channel 20, records the environmental information of the current temperature, humidity, wind speed, longitude and latitude through the temperature and humidity sensor 2, and then sends the taken high-definition pictures and environmental information to the cloud database through the communication module 3. The mice continue to be attracted by the mouse bait 18, enter the mouse bin channel 17 through the main channel 20, and then enter the inside of the mouse bin 13 through the one-way door 21 of the mouse bin channel to achieve the live capture of the mice. There is mouse food for the mice to survive inside the mouse bin 13.

[0072] Among them, the single-chip microcomputer 4 sends a transmission request to the port API of the cloud database every minute through the communication module 3. Only when the infrared sensor 10 has a signal (generally marked as 1) given to the single-chip microcomputer 4, will the single-chip microcomputer 4 send work information to the cloud database, and the cloud database will respond and open the API for the single-chip microcomputer 4 to perform data transmission. In addition to sending high-definition pictures to the cloud database, the single-chip microcomputer 4 also integrates GPS and will simultaneously transmit the environmental parameters of the monitoring terminal when capturing mice, such as information on temperature, humidity, wind speed, longitude and latitude in the environment. After the single-chip microcomputer 4 finishes transmitting the information, the signal of the infrared sensor 10 is cleared (the signal becomes 0) and waits for the next attraction and capture of mice.

[0073] Meanwhile, after the old mouse trap 13 is removed from the instrument mainframe 12, a new mouse trap 13 is replaced on the instrument mainframe 12. A "reset" button is set on the display screen 5. After the user presses the "reset" button, the monitoring terminal can start a new round of monitoring work. For the removed old mouse trap 13, the user can take it back to the laboratory for further research and analysis, such as whether there are disease-carrying fleas, disease-carrying parasites or disease-carrying bacteria on the live mice. Moreover, the old mouse trap 13 can be reused after steps such as cleaning and disinfection.

[0074] The recognition module reads the information from the cloud database and, during training and prediction, first compares the picture size with the target value. Before training, the target object has been annotated using COCO. According to the setting, if the target size is less than 1% of the original picture, it can be determined as a small target. If the picture is larger than [1280, 1280] and the target is defined as a small target, then image slicing processing is required. The size of the side length parameter slice_size of the sub-image is [400, 400], and the overlap ratio between sub-images is set through the parameter overlap_ratio = 0.4. After image slicing, the recognition uses the NMS non-maximum suppression algorithm for filtering and image stitching.

[0075] Among them, the improved YOLOV5 on the recognition module can be a powerful practical tool for supervised learning. During training and prediction, it first compares the picture size with the target value. Before training, the target object has been annotated using COCO. According to the setting defined by the International Society for Optics and Photonics SPIE, if the target size is less than 1% of the original picture, it can be determined as a small target. If the picture is larger than [1280, 1280] and the target is defined as a small target, then image slicing processing is required. The size of the side length parameter slice_size of the sub-image is [400, 400], and the overlap ratio between sub-images is set through the parameter overlap_ratio. For better training and prediction effects, this value is set to 0.5. After image slicing, the recognition uses the NMS non-maximum suppression algorithm for filtering and image stitching.

[0076] The recognition module has stored the results of big data training. The big data training has processed more than 100,000 pictures and recorded the characteristics of common house mice, brown rats, yellow-breasted rats, and striped field mice that have an impact on human health.

[0077] The comparison of the recognition rates between the existing old YOLOV5 algorithm and the improved YOLOV5 algorithm is as follows:

[0078] After comparison, the average recognition rate of the four types of mice is 86%, among which *Apodemus agrarius* reaches 92%. Compared with the existing old YOLOV5 algorithm, the recognition rate of this improved YOLOV5 algorithm has increased by 4%, indicating the high recognition rate and stability of the recognition module. If multi-target recognition pictures are recognized, the performance of the improved YOLOV5 algorithm can be improved by more than 8% compared with the original, and the generality of the recognition module has been improved.

[0079] When a mouse is caught and confirmed by the recognition module, the cloud database returns the recognition data to the monitoring terminal. The display screen 5 on the monitoring terminal displays the warning information in real time. The staff can also use intelligent devices to connect to the monitoring terminal remotely and directly read the warning information. The staff can also replace the mouse bin 13 according to needs. After the replacement of the mouse bin 13 is completed, the number of mice captured on the display screen 5 is cleared manually or automatically, and the monitoring terminal then starts the next round of continuous monitoring of mice.

[0080] The display and warning module is set according to GBT27770-2011, that is, a certain level needs to be reached indoors, outdoors, and in units (especially food production, circulation, and sales enterprises). For example, the capture rate of Class A indoors needs to be ≤1%, or there is no occurrence of major related infectious diseases, such as plague. The warning information it displays can provide guidance and data support for the next step of rodent control work.

[0081] The above is the preferred solution of the present invention, which shows and describes the basic principle, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A rodent intelligent continuous monitoring terminal, comprising an instrument host (12), characterized in that: The instrument host (12) is detachably connected to a mouse bin (13); the instrument host (12) and / or the mouse bin (13) are provided with mouse bait (18) and a camera (6); the instrument host (12) is provided with a temperature and humidity sensor (2), a communication module (3), a single-chip microcomputer (4) and a power supply device (7); and a display screen (5) for displaying the number of mice captured is fixedly provided on the instrument host (12).

2. The rodent intelligent continuous monitoring terminal according to claim 1, characterized in that: The instrument mainframe (12) is provided with a mainframe channel (20), and the mouse compartment (13) is provided with a mouse compartment channel (17). When the mouse compartment (13) is assembled on the instrument mainframe (12), the mouse compartment channel (17) and the mainframe channel (20) are interconnected.

3. The rodent intelligent continuous monitoring terminal according to claim 2, characterized in that: A host channel opening (9) is arranged outside the host channel (20), a host channel one-way door (8) is arranged inside the host channel, and a mouse bin channel one-way door (21) is arranged inside the mouse bin channel (17). The mouse bin channel (17) and the host channel (20) are connected to each other via the host channel one-way door (8) and the mouse bin channel one-way door (21).

4. The rodent intelligent continuous monitoring terminal according to claim 2, characterized in that: The camera (6) is arranged in the host channel (20), and a fill light is also arranged on the camera. An infrared sensor (10) is also arranged in the host channel (20).

5. The rodent intelligent continuous monitoring terminal according to claim 4, characterized in that: The single-chip computer (4) is electrically connected to the temperature and humidity sensor (2), the communication module (3), the display screen (5), the camera (6), the power supply device (7), and the infrared sensor (10).

6. The rodent intelligent continuous monitoring terminal according to claim 1, characterized in that: The power supply device (7) is a rechargeable battery, and the instrument host (12) is connected to a solar panel (19) or a power adapter corresponding to the power supply device (7).

7. The rodent intelligent continuous monitoring terminal according to claim 6, characterized in that: A shielding plate (22) is arranged on the top of the instrument mainframe (12), the outer periphery of the shielding plate (22) extends toward the outside of the instrument mainframe (12), and the solar panel (19) is fixedly arranged on the shielding plate (22) or externally arranged outside the instrument mainframe (12).

8. The rodent intelligent continuous monitoring terminal according to claim 1, characterized in that: The instrument host (12) and the mouse compartment (13) are mutually assembled and disassembled through buckles (14), fasteners, or magnetic attraction.

9. The rodent intelligent continuous monitoring terminal according to claim 1, characterized in that: The side of the mouse compartment (13) is provided with ventilation holes (15).

10. The rodent intelligent continuous monitoring terminal according to claim 1, characterized in that: A handle (11) is provided on the top of the mouse bin (13).

Images