Monitoring, identifying and killing system for rats

Abstract

The invention discloses a mouse monitoring, identifying and killing system, and relates to the technical field of mouse killing, in the system, a wide-angle camera collects a wide-angle image; the controller is used for judging whether a dynamic target exists in the target area or not based on the wide-angle image; if so, controlling the optical zoom camera to align to the target; the optical zoom camera performs tracking shooting on a target at a preset magnification to obtain an optical zoom shooting image; the controller is further used for judging whether the optical zoom shot image meets a first judgment condition or not; if not, determining the target as a mouse; determining a confidence coefficient based on the optical zoom shot image by using a YOLO algorithm; when the confidence coefficient is greater than a preset confidence coefficient, edge detection is performed to obtain body parameters of the target so as to determine a category; when the mouse is in the preset category range, determining a tracking compensation angle of the killing device; and the mouse is killed by the killing device. According to the application, the requirements of safety and convenience can be met while the rodent killing efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rodent killing, in particular to a rodent monitoring, identifying and killing system. BACKGROUND

[0002] In traditional rodent control methods, mainly use traps, poison baits, biological natural enemies control and artificial auxiliary killing means. However, these traditional methods have obvious defects in practical application, and it is difficult to meet the efficient, environmentally friendly, sustainable and precise rodent control needs.

[0003] 1. Low efficiency and limited range of passive rodent control methods: Traditional traps and poison baits are passive operations, which can only act on the mice entering the specific area, and the effective range of a single operation is small. It needs to be frequently placed and replaced with baits or cleaned, and the cost of human maintenance is high. Poison baits also have serious environmental pollution risk, and are easy to cause secondary poisoning through the food chain and injure non-target animals; due to the limitation of mechanical trigger mechanism, the success rate of traps is greatly affected by the size and behavior of rodents, and the overall efficiency is low.

[0004] 2. Biological natural enemies can only achieve ecological balance and cannot control rodents as needed: Natural predators such as cats and snakes will quickly reduce their rodent-catching behavior after meeting their energy needs, and cannot operate on a large scale and continuously according to human needs; at the same time, introducing foreign natural enemies may disrupt the original ecological balance and cause new biological invasion problems.

[0005] 3. Artificial active killing method has poor sustainability and low cost-effectiveness: The current active killing methods mainly rely on artificial use of air guns, bows and arrows, and slingshots, which require high skills, physical strength and time investment from operators, and have small coverage in a single operation and cannot be executed continuously for a long time, with a comprehensive cost-effectiveness much lower than passive methods.

[0006] 4. Visual recognition type intelligent devices have bottlenecks in recognition accuracy and real-time performance: In recent years, the rise of rodent monitoring systems based on cameras can achieve target detection through image algorithms, but due to factors such as complex environmental lighting, obstruction and similar object interference, the recognition accuracy has not reached 100%. In addition, visual algorithms need to upload collected images to the edge or cloud for operation, and the delay in the link of reasoning, decision-making and issuing instructions causes the execution mechanism to miss the best killing opportunity. More importantly, pure visual solutions cannot distinguish between live mice and dead individuals, which can easily lead to false positives and repeated operations.

[0007] 5. Safety restrictions on continuous and high-power killing methods: Due to public safety requirements, traditional firearms, continuous firing devices or high-power weapons are prohibited for rodent control; if the killing power is forcibly increased, it is easy to damage surrounding facilities and injure other organisms, causing additional economic losses and legal risks.

[0008] Therefore, there is an urgent need for a rodent monitoring and identification killing system that can achieve high recognition rate, low delay, sustainability, accurate distinction of living targets, and continuous active killing with controllable and low collateral damage under the premise of safety and compliance. SUMMARY

[0009] The purpose of the present application is to provide a rodent monitoring and identification killing system to improve the efficiency of killing rodents while meeting the requirements of safety and convenience.

[0010] To achieve the above-mentioned purpose, the present application provides the following solutions:

[0011] The present application provides a rodent monitoring and identification killing system, comprising: a wide-angle camera, an optical zoom camera, a controller and a killing device; the wide-angle camera, the optical zoom camera and the killing device are connected with the controller;

[0012] The wide-angle camera is used to collect wide-angle images at consecutive time points within a range of a preset radius of a target area;

[0013] The controller is used to:

[0014] Based on the wide-angle images at consecutive time points, it is determined whether there is a dynamic target in the target area;

[0015] When there is a dynamic target, the optical zoom camera is controlled to be aligned with the target;

[0016] The optical zoom camera is used to track and shoot the target at a preset magnification to obtain an optical zoom shooting image;

[0017] The YOLO algorithm is deployed in the controller; the controller is also used to:

[0018] It is determined whether the optical zoom shooting image meets a first judgment condition; the first judgment condition is that the target in the optical zoom shooting image occupies more than a set boundary of the screen pixel proportion;

[0019] If the first judgment condition is not met, it is determined that the target is a mouse;

[0020] Using the YOLO algorithm, the confidence of the mouse is determined based on the optical zoom shooting image;

[0021] When the confidence of the mouse is greater than a preset confidence and is a living body, edge detection is performed on the mouse in the optical zoom shooting image to obtain the body parameters of the target; the body parameters include the body length pixel and the tail length pixel;

[0022] Based on the body parameters, the category of the mouse is determined;

[0023] When the mouse is within the preset category range, determine the average movement speed and fixed delay time of the mouse based on the optical zoom image, determine the tracking compensation angle of the killing device, and trigger the emission signal of the killing device;

[0024] The killing device is used to kill the mouse after rotating according to the tracking compensation angle.

[0025] In an embodiment, the mouse monitoring, identifying and killing system further comprises a thermal imaging camera; the thermal imaging camera is connected to the controller;

[0026] The thermal imaging camera is used to collect a thermal distribution map of the target area after determining the category of the mouse.

[0027] The controller is further used to determine whether the mouse is a live body based on the thermal distribution map of the target area.

[0028] In an embodiment, the mouse monitoring, identifying and killing system further comprises a tripod;

[0029] The wide-angle camera, the optical zoom camera, the controller, the killing device and the thermal imaging camera are all arranged on the tripod.

[0030] In an embodiment, the mouse monitoring, identifying and killing system further comprises a scale; the scale is arranged 10 meters in front of the tripod.

[0031] In an embodiment, the preset radius is 10 meters.

[0032] In an embodiment, the killing device is an electromagnetic emitter installed with an iron projectile.

[0033] In an embodiment, based on the body parameters, the category of the mouse is determined, comprising:

[0034] When the body length pixel is greater than 300px and the ratio of the tail length pixel to the body length pixel is less than 0.8, the category of the mouse is determined to be R. norvegicus;

[0035] When the body length pixel is less than 300px and the ratio of the tail length pixel to the body length pixel is greater than 1, the category of the mouse is determined to be M. musculus;

[0036] When the ratio of the tail length pixel to the body length pixel is greater than 1.2, the category of the mouse is determined to be M. rattus.

[0037] In an embodiment, the mouse monitoring, identifying and killing system further comprises a memory; the memory is connected to the controller;

[0038] The memory is configured to store the number of killed mice and a killing log.

[0039] In an embodiment, the mouse monitoring and killing system further comprises a pan-tilt head.

[0040] The wide-angle camera, the optical zoom camera, the controller, the killing device and the thermal imaging camera are all arranged on the tripod through the pan-tilt head; the pan-tilt head is connected with the controller.

[0041] The controller is configured to control the rotation of the wide-angle camera, the optical zoom camera, the killing device and the thermal imaging camera by controlling the rotation of the pan-tilt head.

[0042] In an embodiment, the method for determining the average movement speed of the mouse, the fixed delay time, the tracking compensation angle of the killing device and triggering the emission signal of the killing device based on the optical zoom image comprises the following steps:

[0043] The calculation formula of the tracking compensation angle of the killing device is:

[0044] sin(a / 2)=(V×t) / (2R);

[0045] Wherein, a is the tracking compensation angle of the killing device; V is the average movement speed of the mouse; t is the fixed delay time; R is the distance from the killing device to the mouse.

[0046] According to the specific embodiments provided in the present application, the following technical effects are disclosed:

[0047] The application discloses a mouse monitoring, identifying and killing system. A wide-angle camera collects wide-angle images of continuous time points in a preset radius range of a target area; a controller judges whether there is a dynamic target in the target area based on the wide-angle images of the continuous time points; when there is a dynamic target, an optical zoom camera is controlled to be aligned with the target; the optical zoom camera tracks and photographs the target at a preset magnification to obtain an optical zoom photograph; it is judged whether the optical zoom photograph meets a first judgment condition; if the first judgment condition is not met, the target is determined to be a mouse; a YOLO algorithm is used to determine the confidence of the mouse based on the optical zoom photograph; when the confidence of the mouse is greater than a preset confidence and the mouse is a live body, edge detection is performed on the mouse in the optical zoom photograph to obtain body parameters of the target; the category of the mouse is determined based on the body parameters; when the mouse is within a preset category range, the average motion speed and the fixed delay time of the mouse are determined based on the optical zoom photograph, the tracking compensation angle of a killing device is determined, and the emission signal of the killing device is triggered; and the killing device is used to kill the mouse after rotating according to the tracking compensation angle. The application adopts multiple cameras and a controller, combines the killing device, and automatically realizes mouse identification and killing, thereby improving the efficiency of killing mice, and meeting the requirements of safety and convenience. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the related art, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0049] Fig. 1 The first view schematic diagram of the structure of the mouse monitoring, identifying and killing system provided by an embodiment of the present application is shown in FIG. 1.

[0050] Fig. 2 The second view schematic diagram of the structure of the mouse monitoring, identifying and killing system is shown in FIG. 2.

[0051] Fig. 3 The third view schematic diagram of the structure of the mouse monitoring, identifying and killing system is shown in FIG. 3. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0053] The purpose of the present application is to provide a mouse monitoring, identifying and killing system, aiming to improve the efficiency of killing mice while meeting the requirements of safety and convenience.

[0054] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0055] In an exemplary embodiment, as shown in Figs. 1-3 A mouse monitoring, identifying and killing system is provided, comprising a wide-angle camera, an optical zoom camera, a controller (not shown in the figure) and a killing device; the wide-angle camera, the optical zoom camera and the killing device are all connected with the controller.

[0056] The wide-angle camera is used to collect wide-angle images at continuous time points within a preset radius of the target area.

[0057] The controller is used to:

[0058] Based on the wide-angle images at continuous time points, it is determined whether there is a dynamic target in the target area.

[0059] When there is a dynamic target, the optical zoom camera is controlled to be aligned with the target.

[0060] The optical zoom camera is used to track and shoot the target at a preset magnification to obtain an optical zoom shooting image.

[0061] The YOLO algorithm is deployed in the controller; the controller is also used to:

[0062] It is determined whether the optical zoom shooting image meets a first determination condition; the first determination condition is that the target in the optical zoom shooting image occupies a proportion of pixels in the picture greater than a set boundary.

[0063] If the first determination condition is not met, it is determined that the target is a mouse.

[0064] Using the YOLO algorithm, the confidence of the mouse is determined based on the optical zoom shooting image.

[0065] When the confidence of the mouse is greater than a preset confidence and is a live body, edge detection is performed on the mouse in the optical zoom shooting image to obtain body parameters of the target; the body parameters include body length pixels and tail length pixels.

[0066] Based on the body parameters, the category of the mouse is determined.

[0067] When the mouse is within a preset category range, the average motion speed and the fixed delay time of the mouse are determined based on the optical zoom shooting image, the tracking compensation angle of the killing device is determined, and the trigger signal of the killing device is triggered.

[0068] The killing device is used for killing the mouse after rotating according to the tracking compensation angle.

[0069] Specifically, in the use process of the present application, wide-angle camera is used for patrol identification in a large range of 10 meters radius, and wide-angle shooting images in the large range of 10 meters radius are obtained, and the wide-angle shooting images in the large range of 10 meters radius are sent to the controller in real time.

[0070] The controller determines whether there is a dynamic target in the target area based on the wide-angle shooting image of the target area at the continuous moment; if there is a dynamic target, the controller controls the picture center of the optical zoom camera to align the target, and at the same time, the optical zoom camera is zoomed in to a preset magnification for shooting, and an optical zoom shooting image is obtained.

[0071] The controller determines whether the optical zoom shooting image meets the first judgment condition; the first judgment condition is that the proportion of the dynamic target in the optical zoom shooting image is greater than the set boundary of 640*480.

[0072] If the first judgment condition is met, the controller determines that the target is not a mouse.

[0073] If the first judgment condition is not met, the controller determines that the target is a mouse, and a mouse YOLO algorithm (including but not limited to YOLOv5s algorithm) is enabled to determine the confidence of the mouse; when the confidence of the mouse is greater than the preset confidence and is a living body, edge detection is performed on the mouse in the optical zoom shooting image to obtain the body length pixel and tail length pixel of the mouse when the mouse occupies the widest pixel in the picture, so as to determine the category of the mouse.

[0074] It is judged whether the category of the mouse is a preset killing mouse species, if yes, the average motion speed and the fixed delay time of the mouse are determined based on the optical zoom shooting image, the tracking compensation angle of the killing device is determined, and the trigger signal of the killing device is triggered. After the wide-angle camera confirms that there is no other large moving target approaching and no background object that may cause accidental injury, the trigger signal of the electromagnetic emitter is triggered to kill the target.

[0075] Among them, the wide-angle camera performs wide area monitoring when it is in standby, and marks a 10-meter fence line on the screen of the monitoring software. After the mouse individual crosses the fence line, the optical zoom camera for aiming is triggered to zoom in for aiming.

[0076] The optical zoom camera identifies the mouse image in the picture after preset zooming. Since the fixed zooming ratio is adopted, the upper limit of the mouse picture projection size in the fence range is fixed. After collecting the edge profile data of the mouse projection picture, the mouse data model embedded in the system is matched. If the size exceeds the mouse model data, it is considered as a non-mouse organism and is not processed. The picture is automatically switched back to the wide-angle camera for patrol standby, and the non-target is excluded. In order to quickly analyze, the resolution is generally reduced, and the image with 640*480 resolution is used as a standard. The horizontal 640 pixel grid and the vertical 480 pixel grid form an array. The scale of 20 cm at 10 meters is used. After zooming the picture by a fixed zooming ratio, the scale length occupies 600 horizontal pixels, and each pixel represents an actual length of about 0.33 mm. After shooting at 10 meters, the body length occupies 550 pixels, and the body length is 181.5 cm.

[0077] As an optional implementation, the mouse monitoring, identifying and killing system further comprises a thermal imaging camera; the thermal imaging camera is connected with the controller.

[0078] The thermal imaging camera is used to collect a thermal distribution map of the target area after determining the category of the mouse.

[0079] The controller is further used to determine whether the mouse is a live body based on the thermal distribution map of the target area.

[0080] Specifically, for the killed mouse, the target filtering and exclusion are performed by the thermal imaging camera during the patrol process. The repeated killing of the killed target is prevented.

[0081] As an optional implementation, the mouse monitoring, identifying and killing system further comprises a tripod.

[0082] The wide-angle camera, the optical zoom camera, the controller, the killing device and the thermal imaging camera are all arranged on the tripod.

[0083] As an optional implementation, the mouse monitoring, identifying and killing system further comprises a scale; the scale is arranged at a position 10 meters in front of the tripod.

[0084] As an optional implementation, the preset radius is 10 meters.

[0085] As an optional implementation, the killing device is an electromagnetic launcher provided with iron bullets.

[0086] Specifically, the electromagnetic launcher is an electromagnetic gun, the standby bullets are 10-20, the bullets are 8*20 mm iron with a weight of about 8 g, the initial speed is 100 m / s, the kinetic energy = 8 g / 1000*100 2 / 2=80 joule, according to the proportion of the cross section, the kinetic energy of the emitter exit is 80 / (0.4 2 ×3.14)=159 joule / square centimeter, which has enough lethal energy for the mouse.

[0087] As an optional implementation, the category of the mouse is determined based on the body parameter, including:

[0088] When the body length pixel is greater than 300px and the ratio of the tail length pixel to the body length pixel is less than 0.8, the category of the mouse is determined as R. norvegicus.

[0089] When the body length pixel is less than 300px and the ratio of the tail length pixel to the body length pixel is greater than 1, the category of the mouse is determined as M. musculus.

[0090] When the ratio of the tail length pixel to the body length pixel is greater than 1.2, the category of the mouse is determined as M. rattus.

[0091] As an optional implementation, the rodent monitoring, identifying and killing system further includes a memory; the memory is connected with the controller.

[0092] The memory is used for storing the number of killed mice and the killing log.

[0093] As an optional implementation, the rodent monitoring, identifying and killing system further includes a gimbal.

[0094] The wide-angle camera, the optical zoom camera, the controller, the killing device and the thermal imaging camera are all arranged on the tripod through the gimbal; the gimbal is connected with the controller.

[0095] The controller is used for controlling the rotation of the gimbal, thereby controlling the rotation of the wide-angle camera, the optical zoom camera, the killing device and the thermal imaging camera.

[0096] As an optional implementation, the average motion speed and the fixed delay time of the mouse are determined based on the image captured by the optical zoom camera, the tracking compensation angle of the killing device is determined, and the trigger signal of the killing device is triggered, including:

[0097] The calculation formula of the tracking compensation angle of the killing device is:

[0098] sin(a / 2)=(V×t) / (2R).

[0099] Wherein, a is the tracking compensation angle of the killing device; V is the average motion speed of the mouse; t is the fixed delay time; R is the distance from the killing device to the mouse.

[0100] The application: replace complex feature learning with fixed resolution, simplified scale comparison, and adapt to low-cost controllers; Double insurance: optical recognition + thermal imaging verification, avoid misjudgment of dead bodies or non-living things; Safe and controllable: wide-angle alert + energy adjustment, comply with regulations and reduce collateral damage; High efficiency and continuity: electromagnetic emitter supports 10-20 continuous shots, combined with dynamic tracking to achieve fast response. These modules work together to realize the full-process automation of "recognition-tracking-killing-safety verification", balancing efficiency and safety.

[0101] The technical features of the above embodiments can be combined in any way. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0102] The principles and implementation modes of the present application are described by applying specific examples herein, and the above embodiments are only used to help understand the system and core idea of the present application; meanwhile, for those skilled in the art, the specific implementation modes and application ranges will be changed according to the idea of the present application. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A rodent monitoring, identification, and killing system, characterized in that, The rodent monitoring, identification, and killing system includes: a wide-angle camera, an optical zoom camera, a controller, and a killing device; the wide-angle camera, the optical zoom camera, and the killing device are all connected to the controller; The wide-angle camera is used to capture wide-angle images of the target area at continuous times within a preset radius. The controller is used for: Based on wide-angle images at continuous time intervals, determine whether there are dynamic targets within the target area; When a dynamic target is present, control the optical zoom camera to aim at the target; The optical zoom camera is used to track and capture images of the target at a preset magnification to obtain an optical zoom image. The controller deploys the YOLO algorithm; the controller is also used for: Determine whether the optical zoom image meets the first determination condition; the first determination condition is that the proportion of pixels occupied by the target in the optical zoom image is greater than a set boundary. If the first judgment condition is not met, then the target is determined to be a mouse; Using the YOLO algorithm, the confidence level of the mouse is determined based on the optical zoom image; When the confidence level of the mouse is greater than the preset confidence level and the mouse is alive, edge detection is performed on the mouse in the optical zoom image to obtain the body parameters of the target; the body parameters include body length pixels and tail length pixels; Based on the aforementioned physical parameters, the mouse category is determined; When the mouse is within a preset category range, the average movement speed and fixed delay time of the mouse are determined based on the optical zoom image, the tracking compensation angle of the killing device is determined, and the firing signal of the killing device is triggered. The killing device is used to kill the rat after rotating according to the tracking compensation angle.

2. The rodent monitoring, identification, and killing system according to claim 1, characterized in that, The rodent monitoring, identification, and killing system further includes: a thermal imaging camera; the thermal imaging camera is connected to the controller; The thermal imaging camera is used to acquire a thermal distribution map of the target area after the type of mouse has been determined. The controller is also used to determine whether a mouse is alive based on a heat map of the target area.

3. The rodent monitoring, identification, and killing system according to claim 2, characterized in that, The rodent monitoring, identification, and killing system also includes: a tripod; The wide-angle camera, the optical zoom camera, the controller, the killing device, and the thermal imaging camera are all mounted on the tripod.

4. The rodent monitoring, identification, and killing system according to claim 3, characterized in that, The rodent monitoring, identification, and killing system also includes a scale; the scale is set 10 meters in front of the tripod.

5. The rodent monitoring, identification, and killing system according to claim 1, characterized in that, The preset radius is 10 meters.

6. The rodent monitoring, identification, and killing system according to claim 1, characterized in that, The killing device is an electromagnetic launcher equipped with an iron projectile.

7. The rodent monitoring, identification, and killing system according to claim 1, characterized in that, Based on the aforementioned physical parameters, the mouse is categorized, including: When the body length in pixels is greater than 300 pixels and the ratio of the tail length in pixels to the body length in pixels is less than 0.8, the mouse is classified as a brown rat. When the body length in pixels is less than 300 pixels and the ratio of the tail length in pixels to the body length in pixels is greater than 1, the mouse is classified as a house mouse. When the ratio of tail length pixels to body length pixels is greater than 1.2, the mouse is classified as a black rat.

8. The rodent monitoring, identification, and killing system according to claim 1, characterized in that, The rodent monitoring, identification, and killing system further includes: a memory; the memory is connected to the controller; The memory is used to store the number of mice killed and the kill log.

9. The rodent monitoring, identification, and killing system according to claim 3, characterized in that, The rodent monitoring, identification, and killing system also includes: a pan-tilt unit; The wide-angle camera, the optical zoom camera, the controller, the killing device, and the thermal imaging camera are all mounted on the tripod via the gimbal; the gimbal is connected to the controller; The controller is used to control the rotation of the wide-angle camera, optical zoom camera, killing device, and thermal imaging camera by controlling the rotation of the gimbal.

10. The rodent monitoring, identification, and killing system according to claim 1, characterized in that, Based on the optical zoom image, the average movement speed and fixed delay time of the mouse are determined, the tracking compensation angle of the killing device is determined, and the transmission signal of the killing device is triggered, including: The formula for calculating the tracking compensation angle of the killing device is: sin(a / 2)=(V×t) / (2R); Where a is the tracking compensation angle of the killing device; V is the average movement speed of the mouse; t is the fixed delay time; and R is the distance from the killing device to the mouse.

Images