Animal identification and driving method for production site of civilian explosive industry and inspection robot

Abstract

The present application relates to the field of safety production technology of civilian explosive industry, in particular to a method for animal identification and driving in the production site of civilian explosive industry and a patrol robot, image information of an animal to be identified is obtained, the image information includes animal characteristics to be identified and animal position coordinates to be identified; the animal species to be identified and corresponding animal position information in the production site are obtained according to the image information; a driving strategy is determined according to the animal species to be identified, and a driving route is determined according to the animal position information; the animal to be identified is driven out of the production site according to the driving strategy and the driving route. The patrol robot provided by the present application can identify and locate the animals in the patrol area, formulate a driving strategy and a driving route, and dynamically drive the foreign animals away. The safety hidden trouble problem caused by foreign animals in the production area of civilian explosive industry is effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety production in the civilian explosive industry, and in particular to an animal identification and driving method for a production site in the civilian explosive industry and a patrol robot, which is particularly suitable for safety protection in a dangerous goods production plant area such as explosives, detonators, and pyrotechnics. BACKGROUND

[0002] The production site in the civilian explosive industry, such as an explosives factory and a detonator production line, has extremely high safety requirements. Foreign animals such as birds, rodents, cats, and dogs entering the plant area may cause the following safety hazards: mechanical damage: animals gnawing electrical wires and damaging equipment, leading to short circuits or equipment failures; explosion risk: animals carrying flammable materials such as dry grass and garbage into the dangerous goods storage area may cause fires or explosions; pollution risk: animal excrement may contaminate raw materials or finished products, affecting product quality; interference with production: animals entering the production line may trigger false alarms or emergency shutdowns, affecting production efficiency.

[0003] Currently, the civilian explosive industry mainly uses the following methods to prevent animal intrusion: such as bird nets, fences, and ultrasonic bird repellers, but the coverage is limited and animals cannot be accurately identified and located; it relies on security personnel for regular patrols, but the efficiency is low, the cost is high, and it is difficult to guarantee at night or in bad weather; ordinary cameras can only record pictures and cannot automatically identify animals and take driving measures. The technical defect is that the existing methods are mostly passive protection and cannot actively identify and accurately drive animals.

[0004] How to provide a robot that can autonomously patrol, realize intelligent identification, accurate positioning, and dynamic driving is a technical problem to be solved by those skilled in the art. SUMMARY

[0005] Therefore, it is necessary to provide an animal identification and driving method for a production site in the civilian explosive industry and a patrol robot in view of the above technical problems.

[0006] The present application provides an animal identification and driving method for a production site in the civilian explosive industry, which comprises:

[0007] Obtaining image information of an animal to be identified, the image information comprising animal characteristics and animal position coordinates to be identified;

[0008] Obtaining the animal species to be identified and corresponding animal position information in the production site according to the image information;

[0009] Determining a driving strategy according to the animal species to be identified and determining a driving route according to the animal position information;

[0010] Driving the animal to be identified out of the production site according to the driving strategy and the driving route.

[0011] In one of the embodiments, image information of the animal to be identified is acquired, the image information including animal features to be identified and animal position coordinates to be identified, including:

[0012] The production site is monitored in real time by the image acquisition device rotating continuously by 360° horizontally and +90° to -90° vertically.

[0013] The animal to be identified collected by the image acquisition device is positioned by the positioning module.

[0014] In one of the embodiments, the animal species to be identified is obtained according to the image information, including:

[0015] An initial frame image of the animal to be identified is acquired, the initial frame image including key features of the animal to be identified and surrounding environment reference points, the key features being body shape, hair color and texture, and posture features, and the surrounding environment reference points being ground texture and vegetation distribution;

[0016] The region of the animal to be identified in the initial frame image is positioned by the image recognition module, and the region of interest centered on the animal to be identified is taken as a matching template region, and the region where the animal is located in the animal image in the species database is positioned as a matching candidate region by a deep learning algorithm;

[0017] The region features of the template region and the candidate region are extracted by the deep learning algorithm to obtain gray scale features and texture features;

[0018] The gray scale features and the texture features are subjected to similarity measurement and matching calculation to obtain the animal species to be identified.

[0019] In one of the embodiments, the corresponding animal position information in the production site includes:

[0020] Based on the lens parameters of the image acquisition device, the pixel coordinates are converted into the horizontal azimuth angle α and the vertical elevation angle β of the target relative to the camera;

[0021] According to the actual size of the animal to be identified, the pixel size of the animal to be identified in the image is measured by OpenCV, and the linear distance of the animal to be identified is obtained by combining the focal length f of the camera and using the similar triangle principle;

[0022] Taking the absolute coordinates Pcam of the camera of the image acquisition device as the origin, the corresponding animal position information in the production site is obtained according to the linear distance, the horizontal azimuth angle α and the vertical elevation angle β.

[0023] In one of the embodiments, the driving strategy is determined according to the animal species to be identified, including:

[0024] The animal species to be identified is a bird, the sound of its natural enemy is played, and the deterrent effect is obtained by combining the sound intensity and frequency change.

[0025] When the identified animal species is sensitive to light, a high-intensity, flickering light device is turned on to interfere with its vision and force it to leave;

[0026] When the identified animal species is small and flexible, a wind power generation device generates directional wind to change its direction of movement.

[0027] In one embodiment, the driving route is determined according to the animal position information, comprising:

[0028] Determining the animal position information;

[0029] Calculating the distance between the nearest point on the reference path to the animal position information and the animal position information, and taking the distance as the position deviation;

[0030] Calculating the control amount of the position deviation corresponding to the position deviation;

[0031] Determining the weight of each independent turning control amount according to the control amount;

[0032] According to the independent turning control amount and the corresponding weight, the target turning control amount is calculated;

[0033] According to the control amount of the position deviation and the target turning control amount, the driving route is obtained.

[0034] In one embodiment, the identified animal is driven out of the production site according to the driving strategy and the driving route, comprising:

[0035] The motion control module controls the four-legged bionic structure designed inspection robot to move towards the target position according to the planned path and positioning information.

[0036] A kind of inspection robot, comprising:

[0037] The acquisition module is used to obtain the image information of the animal to be identified, and the image information includes the characteristics of the animal to be identified and the position coordinates of the animal to be identified;

[0038] The image information analysis module is used to obtain the animal species to be identified and the corresponding animal position information in the production site according to the image information;

[0039] The driving strategy and driving route generation module is used to determine the driving strategy according to the animal species to be identified, and determine the driving route according to the animal position information;

[0040] The driving module is used to drive the animal to be identified out of the production site according to the driving strategy and the driving route.

[0041] A computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following when executing the computer program:

[0042] Obtain image information of an animal to be identified, the image information comprising animal characteristics and position coordinates of the animal to be identified;

[0043] Obtain the animal species to be identified and corresponding animal position information in the production site according to the image information;

[0044] Determine a driving strategy according to the animal species to be identified and determine a driving route according to the animal position information;

[0045] Drive the animal to be identified out of the production site according to the driving strategy and the driving route.

[0046] A computer storage medium stores a computer program, and the computer program is executed by a processor to implement the following:

[0047] Obtain image information of an animal to be identified, the image information comprising animal characteristics and position coordinates of the animal to be identified;

[0048] Obtain the animal species to be identified and corresponding animal position information in the production site according to the image information;

[0049] Determine a driving strategy according to the animal species to be identified and determine a driving route according to the animal position information;

[0050] Drive the animal to be identified out of the production site according to the driving strategy and the driving route.

[0051] The animal identification and driving method for a production site in the civilian explosive industry comprises the following steps: obtaining image information of an animal to be identified, the image information comprising animal characteristics and position coordinates of the animal to be identified; obtaining the animal species to be identified and corresponding animal position information in the production site according to the image information; determining a driving strategy according to the animal species to be identified and determining a driving route according to the animal position information; and driving the animal to be identified out of the production site according to the driving strategy and the driving route. The inspection robot provided by the application can identify, locate, determine a driving strategy and a driving route, and dynamically drive away the alien animals, thereby effectively solving the problem of the safety hazards caused by the alien animals in the production area of the civilian explosive industry. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0053] Figure 1 A flow chart of a method for identifying and driving animals away from a production site in the civilian explosive industry in one embodiment.

[0054] Figure 2 An internal structure diagram of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to facilitate the understanding of the present application, the present application will be described in more detail below with reference to the relevant drawings. The drawings show embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0057] It can be understood that the terms "first", "second", etc. used in the present application can be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0058] In one embodiment, as shown in Figure 1 , a method for identifying and driving animals away from a production site in the civilian explosive industry is provided. Taking a patrol robot in Figure 1 as an example, the method includes the following steps: obtaining image information of an animal to be identified, the image information including animal characteristics to be identified and animal position coordinates to be identified; obtaining animal species to be identified and corresponding animal position information in the production site according to the image information; determining a driving strategy according to the animal species to be identified and determining a driving route according to the animal position information; and driving the animal to be identified out of the production site according to the driving strategy and the driving route.

[0059] Specifically, the inspection robot adopts a network high-definition camera with a holder, such as a S series 400 million 32 times network infrared T holder iDS-2DY5432IXR(T5), which can realize 360° continuous rotation in the horizontal direction and +90~ -90° rotation in the vertical direction, and realize panoramic scanning. It has an open network video interface RJ45, can be linked with an image recognition module to transmit data, has an external MicroSD card slot, supports external storage, and solves the short-term storage capacity demand of video images. The working temperature supports -40℃~70℃; the humidity is less than 95%, which meets the requirements of outdoor working environment in the factory area. The camera has a built-in 3D positioning function, which obtains its own positioning information through a GPS / Beidou dual-mode positioning system. The positioning module is connected with the camera through the RJ45 network port and works cooperatively to position the animals to be identified photographed by the camera. In the positioning module, the computer vision library OpenCV is used for image preprocessing, recognition, positioning and measurement, etc. Image processing algorithms are used to obtain the two-dimensional coordinate distance of the animal to be identified relative to the camera. Combined with the positioning information of the camera itself, the similar triangle principle formula is used to estimate the distance between the animal to be identified and the camera, and the positioning information of the animal to be identified is calculated.

[0060] The above-mentioned animal identification and driving method for the production site of the civilian explosive industry obtains image information of the animal to be identified, the image information including animal characteristics and animal position coordinates; obtains the animal species and corresponding animal position information in the production site according to the image information; determines the driving strategy according to the animal species, and determines the driving route according to the animal position information; drives the animal to be identified out of the production site according to the driving strategy and the driving route. The inspection robot provided by the present application can identify, position, formulate driving strategies and driving routes for animals in the patrol area, and dynamically drive away the alien animals. The safety hidden trouble problem caused by alien animals in the production area of the civilian explosive industry is effectively solved.

[0061] In one embodiment, the image information of the animal to be identified is obtained, and the image information includes animal characteristics and animal position coordinates, including: real-time monitoring of the production site by a 360° continuous rotation in the horizontal direction and +90~ -90° rotation in the vertical direction image acquisition device; positioning the animal to be identified collected by the image acquisition device through a positioning module.

[0062] Specifically, the positioning module is connected with the camera through the RJ45 network port and cooperates to position the to-be-identified animal photographed by the camera. In the positioning module, the image processing algorithms such as image preprocessing, recognition, positioning and measurement are performed by using the computer vision library OpenCV to obtain the two-dimensional coordinate distance of the to-be-identified animal relative to the camera. In combination with the positioning information of the camera itself, the distance between the object and the camera is estimated by using the similar triangle principle calculation formula to calculate the positioning information of the animal target.

[0063] In one of the embodiments, the species of the to-be-identified animal is obtained according to image information, including: obtaining an initial frame image of the to-be-identified animal, the initial frame image including key features of the to-be-identified animal and surrounding environment reference points, the key features being body shape contour, fur texture and posture feature, and the surrounding environment reference points being ground texture and vegetation distribution; positioning the region of the to-be-identified animal in the initial frame image by the image recognition module, taking the region of interest centered on the to-be-identified animal as a matching template region, and positioning the region where the animal is located in the animal image in the species database as a matching candidate region by a deep learning algorithm; extracting region features of the template region and the candidate region by the deep learning algorithm to obtain gray features and texture features; and performing similarity measurement and matching calculation on the gray features and the texture features to obtain the species of the to-be-identified animal.

[0064] Specifically, the image recognition module of the inspection robot positions the region of the to-be-identified animal in the initial frame image by a deep learning algorithm, determines the region of interest (ROI, such as a rectangular region containing the whole body of the animal and the surrounding environment) centered on the animal as a matching “template region”, and positions the region where the animal is located in the animal image in the species database as a matching “candidate region”. The image recognition module extracts region features of the template region and the candidate region by a deep learning algorithm as a matching basis. The gray features: calculate the average value, variance and gradient histogram (such as Sobel gradient) of the gray scale in the region to represent the brightness change and edge distribution of the region. The texture features: extract the texture information of the region by LBP (Local Binary Pattern) or GLCM (Gray Level Co-occurrence Matrix), which is suitable for distinguishing animal fur texture and environment vegetation texture.

[0065] Similarity measurement and matching calculation:

[0066] Sum of Squared Difference (SSD): the sum of squares of the difference between the corresponding pixel gray values of the two regions is calculated, and the smaller the value is, the more similar the regions are.

[0067] Formula: SSD = ∑(x, y) e ROI (I1(x, y) - I2(x, y))2

[0068] Where I1 is the gray value of the template region, and I2 is the gray value of the candidate region.

[0069] Normalized Cross Correlation (NCC): By normalizing the gray scale distribution of two regions, the influence of light change is eliminated, and the similarity range is [-1, 1], the closer to 1, the higher the matching degree.

[0070] Formula:

[0071] Wherein μ1, μ2 are the average gray values of the two regions. The image recognition module accurately obtains the species of the animal to be identified through the above measurement method.

[0072] In one of the embodiments, the corresponding animal position information in the production site includes: based on the lens parameters of the image acquisition device, the pixel coordinates are converted into the horizontal azimuth angle a and the vertical elevation angle β of the target relative to the camera; according to the actual size of the animal to be identified, the pixel size of the animal to be identified in the image is measured through OpenCV, and the linear distance of the animal to be identified is obtained by combining the focal length f of the camera and using the similar triangle principle; taking the absolute coordinates Pcam of the camera of the image acquisition device as the origin, the linear distance, the horizontal azimuth angle a and the vertical elevation angle β are obtained to obtain the corresponding animal position information in the production site.

[0073] Specifically, the positioning module of the inspection robot is a GPS / Beidou dual-mode positioning, and the real-time absolute coordinates of the camera are set as Pcam=(Xcam, Ycam, Zcam, wherein Zcam is the camera height, which can be preset through the parameters of the inspection robot itself.

[0074] The relative information of the animal to be identified relative to the camera is calculated through OpenCV

[0075] Target detection and pixel coordinate extraction: the target detection algorithm (such as YOLO, SSD) of OpenCV is used to identify the contour of the animal in the image, and the pixel coordinates (set as (u, v), that is, the column and row pixel positions in the two-dimensional plane of the image) of the animal in the image are determined.

[0076] Relative azimuth angle calculation: based on the lens parameters (horizontal / vertical field of view FOVw, FOVh) of the camera, the pixel coordinates (u, v) are converted into the horizontal azimuth angle a and the vertical elevation angle β of the target relative to the camera:

[0077] Horizontal azimuth angle a=(u−W / 2)×FOVw / W (W is the image width pixel), positive value indicates the right side of the camera, negative value indicates the left side;

[0078] Vertical elevation angle β=(H / 2−v)×FOVh / H (H is the image height pixel), positive value indicates the upper side of the camera, negative value indicates the lower side.

[0079] Relative distance estimation: The actual size of the animal is obtained from the species database after the animal species has been identified by the image recognition module. From the actual size of the known target animal (such as body length L), its pixel size l in the image is measured by OpenCV, combined with the camera focal length f, and the distance D is calculated using the similarity triangle principle: D = (L x f) / l, where D is the straight-line distance from the camera to the animal.

[0080] With the camera absolute coordinates Pcam as the origin, combined with the relative distance D, the horizontal azimuth angle a and the vertical elevation angle b, the absolute coordinates of the animal in the factory coordinate system Panimal = (Xa, Ya, Za) are calculated by the three-dimensional coordinate conversion formula:

[0081] Horizontal plane (X-Y axis) coordinates: Xa = Xcam + D x cos(b) x sin(a), Ya = Ycam + D x cos(b) x cos(a)

[0082] Vertical height (Z axis) coordinates: Za = Zcam + D x sin(b). If the animal is on the ground, Za can be simplified as the ground height, which is calibrated by environmental sensors or factory terrain data.

[0083] In one of the embodiments, the driving strategy is determined according to the animal species to be identified, including: if the animal species to be identified is a bird, the sound of its natural enemy is played, and the deterrent effect is obtained by combining the sound intensity and frequency change; if the animal species to be identified is an animal sensitive to light, a high-intensity, flickering light device is turned on to interfere with its vision and force it to leave; if the animal species to be identified is a small-sized and flexible animal, a wind power generation device will generate directional wind to change its direction of movement.

[0084] In one of the embodiments, the driving route is determined according to the animal position information, including: determining the animal position information; calculating the distance between the nearest point on the reference path to the animal position information and the animal position information, and taking the distance as the position deviation; calculating the control amount of the position deviation corresponding to the position deviation; determining the weight of each independent corner control amount according to the control amount; calculating the target corner control amount according to the independent corner control amount and the corresponding weight; obtaining the driving route according to the control amount of the position deviation and the target corner control amount.

[0085] Specifically, the nearest point to the animal position information in the reference path is determined, and the distance between the nearest point to the animal position information in the reference path and the animal position information is taken as the position deviation. The position deviation control amount is calculated according to the following proportional-integral-derivative equation.

[0086] wherein, is a position deviation control quantity, is a proportional parameter, is an integral parameter, is a differential parameter, represents the Nth position deviation in the current control period, represents the (N-1)th position deviation in the current control period, and represents the ith position deviation in the current control period.

[0087] According to the current speed of the inspection robot, the proportional parameter, the integral parameter, and the differential parameter are determined by looking up a mapping relationship between the speed and the position deviation proportional-integral-differential control parameters that are previously calibrated.

[0088] In this embodiment, in order to make the inspection robot keep advancing in the path, the animal to be identified is first recognized, and then the point closest to the animal to be identified in the reference path is determined, the distance between the point closest to the animal to be identified in the reference path and the animal to be identified is taken as the position deviation, and the position deviation control quantity is calculated through the proportional-integral-differential equation.

[0089] In one of the embodiments, the animal to be identified is driven out of the production site according to the driving strategy and the driving route, which comprises: the motion control module controls the inspection robot with the four-legged bionic structure design to travel towards the target position for driving according to the planned path and the positioning information.

[0090] It should be understood that, although Figure 1 The steps in the flowchart of the method are displayed in sequence according to the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps in the method can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0091] In one embodiment, an inspection robot is provided, characterized in that the inspection robot comprises:

[0092] The acquisition module is configured to acquire image information of the animal to be identified, the image information comprising animal features and animal position coordinates of the animal to be identified.

[0093] The image information analysis module is configured to obtain animal species and corresponding animal position information in the production site according to the image information.

[0094] The driving strategy and driving route generation module is configured to determine a driving strategy according to the animal species and determine a driving route according to the animal position information.

[0095] The driving module is configured to drive the animal to be identified out of the production site according to a driving strategy and a driving route.

[0096] In an embodiment, the image information of the animal to be identified is acquired, and the image information includes features of the animal to be identified and position coordinates of the animal to be identified, and includes:

[0097] The real-time monitoring module is configured to monitor the production site in real time by rotating the image acquisition device by 360° horizontally and by +90° to -90° vertically.

[0098] The positioning module is configured to position the animal to be identified collected by the image acquisition device through the positioning module.

[0099] In an embodiment, the species of the animal to be identified is obtained according to the image information, and includes:

[0100] The initial frame acquisition module is configured to acquire an initial frame image of the animal to be identified, and the initial frame image includes key features of the animal to be identified and surrounding environment reference points, the key features are body shape, hair color and texture, and posture features, and the surrounding environment reference points are ground texture and vegetation distribution.

[0101] The region selection module is configured to position a region of the animal to be identified in the initial frame image through the image recognition module, take a region of interest centered on the animal to be identified as a template region for matching, and position a region where an animal is located in an animal image in the species database through a deep learning algorithm as a candidate region for matching.

[0102] The feature extraction module is configured to extract region features of the template region and the candidate region through a deep learning algorithm to obtain grayscale features and texture features.

[0103] The analysis module is configured to perform similarity measurement and matching calculation on the grayscale features and the texture features to obtain the species of the animal to be identified.

[0104] In an embodiment, the corresponding animal position information in the production site includes:

[0105] The pixel coordinate conversion module is configured to convert pixel coordinates into a horizontal azimuth angle a and a vertical elevation angle β of the target relative to the camera based on lens parameters of the image acquisition device.

[0106] The straight line distance analysis module is configured to measure a pixel size of the animal to be identified in an image according to an actual size of the animal to be identified through OpenCV, combine a focal length f of the camera, and obtain a straight line distance of the animal to be identified by using a similar triangle principle.

[0107] The position information analysis module is configured to take the absolute coordinates Pcam of the camera of the image acquisition device as the origin, and obtain the corresponding animal position information in the production site according to the straight-line distance, the horizontal azimuth angle alpha and the vertical elevation angle beta.

[0108] In one embodiment, the driving strategy is determined according to the animal species to be identified, including:

[0109] The bird driving module is configured to, when the animal species to be identified is a bird, play the sound of its natural enemy, and obtain a deterrent effect by combining the sound intensity and frequency change.

[0110] The light-sensitive driving module is configured to, when the animal species to be identified is an animal sensitive to light, turn on a high-intensity and flickering light device to interfere with the vision of the animal and force it to leave.

[0111] The small-sized animal driving module is configured to, when the animal species to be identified is a small-sized and flexible animal, generate directional wind by the wind power generation device to change the moving direction of the animal.

[0112] In one embodiment, the driving route is determined according to the animal position information, including:

[0113] The position determination module is configured to determine the animal position information.

[0114] The position deviation module is configured to calculate the distance between the nearest point on the reference path to the animal position information and the animal position information, and take the distance as the position deviation.

[0115] The control amount analysis module is configured to calculate the control amount of the position deviation corresponding to the position deviation.

[0116] The weight control module is configured to determine the weight of each independent turning angle control amount according to the control amount.

[0117] The target turning angle control amount module is configured to calculate the target turning angle control amount according to the independent turning angle control amount and the corresponding weight.

[0118] The driving route generation module is configured to obtain the driving route according to the control amount of the position deviation and the target turning angle control amount.

[0119] In one embodiment, the driving out of the production site of the animal to be identified according to the driving strategy and the driving route includes:

[0120] The driving module is configured to control the four-legged bionic structure designed inspection robot to move towards the target position according to the planned path and the positioning information.

[0121] The specific definition of the inspection robot can refer to the definition of the animal identification and driving method in the civil explosive industry production site in the above, and will not be described here. Each module in the above inspection robot can be realized by software, hardware and combination thereof in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so that the processor calls and executes the operations corresponding to each module. It should be noted that the division of the modules in the embodiments of the present application is illustrative, and is only a logical function division. In actual implementation, there can be another division manner.

[0122] In one of the embodiments, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in Figure 2 The computer device includes a processor, a memory, a network interface and a database connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store periodic task allocation data, such as configuration files, theoretical running parameters and theoretical deviation value ranges, task attribute information, etc. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a periodic task allocation method.

[0123] Those skilled in the art can understand that Figure 2 The structure shown in the above is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0124] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0125] Obtaining image information of an animal to be identified, the image information including animal features to be identified and animal position coordinates to be identified;

[0126] Obtaining animal species to be identified and corresponding animal position information in the production site according to the image information;

[0127] Determining a driving strategy according to the animal species to be identified and determining a driving route according to the animal position information;

[0128] According to the driving strategy and the driving route, the to-be-identified animal is driven out of the production site.

[0129] In one embodiment, the processor implements obtaining image information of the to-be-identified animal when executing the computer program, the image information includes characteristics of the to-be-identified animal and position coordinates of the to-be-identified animal, and includes:

[0130] The production site is monitored in real time by the image acquisition device which rotates continuously by 360° horizontally and rotates by +90° to -90° vertically.

[0131] The to-be-identified animal collected by the image acquisition device is positioned by the positioning module.

[0132] In one embodiment, the processor implements obtaining the species of the to-be-identified animal according to the image information when executing the computer program, and includes:

[0133] An initial frame image of the to-be-identified animal is obtained, the initial frame image includes key features of the to-be-identified animal and surrounding environment reference points, the key features are body shape, hair color and texture, and posture features, and the surrounding environment reference points are ground texture and vegetation distribution.

[0134] The region of the to-be-identified animal in the initial frame image is located by the image recognition module, the region of interest centered on the to-be-identified animal is taken as a matching template region, and the region where the animal is located in the animal image in the species database is located as a matching candidate region by a deep learning algorithm.

[0135] The region feature extraction of the template region and the candidate region is performed by the deep learning algorithm to obtain the gray scale feature and the texture feature.

[0136] The similarity measurement and matching calculation of the gray scale feature and the texture feature are performed to obtain the species of the to-be-identified animal.

[0137] In one embodiment, the processor implements obtaining the corresponding animal position information in the production site when executing the computer program, and includes:

[0138] Based on the lens parameters of the image acquisition device, the pixel coordinates are converted into the horizontal azimuth angle α and the vertical elevation angle β of the target relative to the camera.

[0139] According to the actual size of the to-be-identified animal, the pixel size of the to-be-identified animal in the image is measured by OpenCV, combined with the focal length f of the camera, and the straight line distance of the to-be-identified animal is obtained by using the similar triangle principle.

[0140] Taking the absolute coordinates Pcam of the camera of the image acquisition device as the origin, the straight line distance, the horizontal azimuth angle α and the vertical elevation angle β are obtained to obtain the corresponding animal position information in the production site.

[0141] In one embodiment, the processor implements determining a driving strategy according to the animal species to be identified when executing the computer program, including:

[0142] The animal species to be identified is a bird, the sound of its natural enemy is played, and the deterrent effect is obtained in combination with the intensity and frequency change of the sound;

[0143] The animal species to be identified is an animal sensitive to light, then a high-intensity, flickering light device is turned on to interfere with its vision and force it to leave;

[0144] The animal species to be identified is a small-sized and flexible animal, a wind power generation device generates directional wind to change its direction of action.

[0145] In one embodiment, the processor implements determining a driving route according to the animal position information when executing the computer program, including:

[0146] Determining the animal position information;

[0147] Calculating the distance between the point closest to the animal position information on the reference path and the animal position information, and taking the distance as the position deviation;

[0148] Calculating the control amount of the position deviation corresponding to the position deviation;

[0149] Determining the weight of each independent corner control amount according to the control amount;

[0150] Calculating the target corner control amount according to the independent corner control amount and the corresponding weight;

[0151] Obtaining the driving route according to the control amount of the position deviation and the target corner control amount.

[0152] In one embodiment, the processor implements driving the animal to be identified out of the production site according to the driving strategy and the driving route when executing the computer program, including:

[0153] The motion control module controls the four-legged bionic structure designed inspection robot to move towards the target position according to the planned path and positioning information.

[0154] In one embodiment, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement:

[0155] Obtaining image information of the animal to be identified, the image information including animal features to be identified and animal position coordinates to be identified;

[0156] Obtaining the animal species to be identified and the corresponding animal position information in the production site according to the image information;

[0157] According to the animal species to be identified, a driving strategy is determined, and according to the animal position information, a driving route is determined.

[0158] According to the driving strategy and the driving route, the animal to be identified is driven out of the production site.

[0159] In one embodiment, the computer program is executed by the processor to realize obtaining image information of the animal to be identified, the image information including animal characteristics and animal position coordinates of the animal to be identified, including:

[0160] Through the image acquisition device rotating continuously by 360° horizontally and +90° to -90° vertically, the production site is monitored in real time.

[0161] The animal to be identified collected by the image acquisition device is positioned by the positioning module.

[0162] In one embodiment, the computer program is executed by the processor to realize obtaining the animal species to be identified according to the image information, including:

[0163] An initial frame image of the animal to be identified is obtained, the initial frame image including key features of the animal to be identified and surrounding environment reference points, the key features being body shape, hair color and texture, and posture features, and the surrounding environment reference points being ground texture and vegetation distribution.

[0164] The region of the animal to be identified in the initial frame image is located by the image recognition module, and the region of interest centered on the animal to be identified is taken as a matching template region, and the region where the animal is located in the animal image in the species database is located as a matching candidate region through a deep learning algorithm.

[0165] The region feature extraction of the template region and the candidate region is performed through the deep learning algorithm to obtain the gray feature and the texture feature.

[0166] The similarity measurement and matching calculation of the gray feature and the texture feature are performed to obtain the animal species to be identified.

[0167] In one embodiment, the computer program is executed by the processor to realize the corresponding animal position information in the production site, including:

[0168] Based on the lens parameters of the image acquisition device, the pixel coordinates are converted into the horizontal azimuth angle α and the vertical elevation angle β of the target relative to the camera.

[0169] According to the actual size of the animal to be identified, the pixel size of the animal to be identified in the image is measured through OpenCV, and the straight-line distance of the animal to be identified is obtained by combining the focal length f of the camera and using the similar triangle principle.

[0170] Taking the camera absolute coordinate Pcam of the image acquisition device as the origin, according to the straight line distance, horizontal azimuth angle alpha and vertical elevation angle beta, the corresponding animal position information in the production site is obtained.

[0171] In one embodiment, the computer program is executed by the processor to determine the driving strategy according to the animal species to be identified, including:

[0172] The animal species to be identified is a bird, the sound of its natural enemy is played, and the deterrent effect is obtained by combining the intensity and frequency change of the sound;

[0173] The animal species to be identified is an animal sensitive to light, then a high-intensity, flickering light device is turned on to interfere with its vision and force it to leave;

[0174] The animal species to be identified is a small-sized and flexible animal, a wind power generation device generates directional wind to change its direction of action.

[0175] In one embodiment, the computer program is executed by the processor to determine the driving route according to the animal position information, including:

[0176] Determine the animal position information;

[0177] Calculate the distance between the nearest point on the reference path to the animal position information and the animal position information, and take the distance as the position deviation;

[0178] Calculate the control amount of the position deviation corresponding to the position deviation;

[0179] Determine the weight of each independent corner control amount according to the control amount;

[0180] According to the independent corner control amount and the corresponding weight, the target corner control amount is calculated;

[0181] According to the control amount of the position deviation and the target corner control amount, the driving route is obtained.

[0182] In one embodiment, the computer program is executed by the processor to drive the animal to be identified out of the production site according to the driving strategy and the driving route, including:

[0183] The motion control module controls the four-legged bionic structure designed inspection robot to move towards the target position according to the planned path and positioning information.

[0184] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, databases, or other media in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0185] Any of the technical features of the above embodiments can be coupled, and in order to make the description simple, not all possible couplings of each technical feature in the above embodiments are described, but as long as the coupling of these technical features does not exist contradictory, it should be considered as the scope of the present application.

[0186] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. A method for animal identification and repelling at production sites in the civil explosives industry, characterized in that, include: Acquire image information of the animal to be identified, the image information including the animal's features and its location coordinates; Based on the image information, the species of animal to be identified and the corresponding animal location information within the production site are obtained; Determine a driving strategy based on the species of the animal to be identified, and determine a driving route based on the animal's location information; According to the driving strategy and the driving route, the animals to be identified are driven out of the production site; The step of obtaining the species of the animal to be identified based on the image information includes: Acquire an initial frame image of the animal to be identified. The initial frame image includes key features of the animal to be identified and reference points of the surrounding environment. The key features are body shape outline, fur color texture and posture features. The reference points of the surrounding environment are ground texture and vegetation distribution. The image recognition module locates the region of the animal to be identified in the initial frame image, and the region of interest centered on the animal to be identified is used as the matching template region. The region of interest includes a rectangular region of the animal's whole body and its surrounding environment. The deep learning algorithm locates the region of the animal in the animal image in the species database as the matching candidate region. Deep learning algorithms are used to extract region features from template regions and candidate regions to obtain grayscale features and texture features. The grayscale features and texture features are subjected to similarity measurement and matching calculation to obtain the species of the animal to be identified; The step of determining a driving-away strategy based on the species of the animal to be identified includes: The animal species to be identified is a bird. The sounds of its natural enemies are played, and the intensity and frequency of the sounds are combined to achieve a deterrent effect. If the animal species to be identified is a light-sensitive animal, then a high-intensity, flashing light device will be turned on to interfere with its vision and force it to leave. The animal species to be identified is a small and agile animal, and the wind generator will produce directional wind to change its direction of movement.

2. The method for animal identification and repelling in civil explosives production sites according to claim 1, characterized in that, The process of acquiring image information of the animal to be identified, wherein the image information includes the animal's features and its location coordinates, includes: The production site is monitored in real time by an image acquisition device that rotates continuously 360° horizontally and +90 to -90° vertically. The positioning module locates the animal to be identified, as captured by the image acquisition device.

3. The method for animal identification and repelling in civil explosives production sites according to claim 1, characterized in that, The animal location information within the production site includes: Based on the lens parameters of the image acquisition device, the pixel coordinates are converted into the horizontal azimuth angle α and the vertical elevation angle β of the target relative to the camera; Based on the actual size of the animal to be identified, its pixel size in the image is measured using OpenCV. Combined with the camera focal length f, the straight-line distance of the animal to be identified is obtained using the principle of similar triangles. Using the absolute coordinates Pcam of the camera in the image acquisition device as the origin, the animal location information within the production site is obtained based on the straight-line distance, the horizontal azimuth angle α, and the vertical elevation angle β.

4. A method for animal identification and repelling in a civil explosives production site according to claim 1, characterized in that, Determining the herding route based on the animal's location information includes: Determine the animal's location information; Calculate the distance between the point on the reference path closest to the animal's location information and the animal's location information, and use the distance as the position deviation; Calculate the control amount of the position deviation corresponding to the position deviation; The weights of each independent corner control quantity are determined based on the control quantity; The target angle control value is calculated based on the independent angle control value and the corresponding weight; The driving line is obtained based on the control amount of the position deviation and the control amount of the target turning angle.

5. A method for animal identification and repelling at production sites in the civil explosives industry according to claim 1, characterized in that, The step of driving the animal to be identified out of the production site according to the driving strategy and the driving route includes: Based on the planned path and positioning information, the motion control module controls the inspection robot with a quadrupedal bionic structure to move and drive it toward the target location.

6. An inspection robot, characterized in that, The inspection robot includes: The acquisition module is used to acquire image information of the animal to be identified, the image information including the features of the animal to be identified and the location coordinates of the animal to be identified; The image information analysis module is used to obtain the species of the animal to be identified and the corresponding animal location information within the production site based on the image information. The module for generating a driving strategy and a driving route is used to determine a driving strategy based on the species of the animal to be identified and to determine a driving route based on the animal's location information. The driving module is used to drive the animal to be identified out of the production site according to the driving strategy and the driving route; The step of obtaining the species of the animal to be identified based on the image information includes: The initial frame image acquisition module is used to acquire the initial frame image of the animal to be identified. The initial frame image includes the key features of the animal to be identified and the surrounding environment reference points. The key features are body shape outline, fur color texture and posture features. The surrounding environment reference points are ground texture and vegetation distribution. The candidate region analysis module is used to locate the region of the animal to be identified in the initial frame image through the image recognition module, and to use the region of interest centered on the animal to be identified as the matching template region. The region of interest includes a rectangular region of the animal's whole body and its surrounding environment. The region where the animal is located in the animal image in the species database is located as the matching candidate region through a deep learning algorithm. The feature extraction module is used to extract regional features from the template region and candidate region using deep learning algorithms to obtain grayscale features and texture features. The animal species analysis module is used to perform similarity measurement and matching calculation on the grayscale features and texture features to obtain the animal species to be identified; The step of determining a driving-away strategy based on the species of the animal to be identified includes: The bird deterrence module is used when the animal to be identified is a bird. It plays the sounds of its natural enemies and combines the intensity and frequency changes of the sounds to achieve a deterrent effect. The light-sensitive animal repelling module is used when the animal to be identified is a light-sensitive animal. In this case, a high-intensity, flashing light device is activated to interfere with its vision and force it to leave. A small and agile animal deterrence module is used when the animal to be identified is a small and agile animal. The wind generator produces directional wind to change its direction of movement.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images