Method and device for monitoring growth state of lobsters in rice field through Internet of Things

Abstract

The invention belongs to the technical field of the Internet of Things, and discloses an Internet of Things monitoring method and device for the growth state of lobsters in a rice field, and the device comprises a lobster collecting module, a solar charging module and a telescopic image recognition module, and the telescopic image recognition module obtains a lobster growth condition monitoring original image collected by the lobster collecting module; and the obtained lobster growth condition monitoring source is transmitted to the cloud data server through the 4G channel, so that the growth change of the lobster can be monitored. According to the method, the growth state of the lobsters in the rice field is monitored in an internet-of-things mode, and yield and productivity prediction in the lobster breeding process is achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of Internet of Things, and specifically relates to an Internet of Things monitoring method and device for the growth status of rice field lobsters. Background Art

[0002] As an alien species, crayfish was introduced into my country in the 1930s and gradually developed into an industrialized species. The species of crayfish cultured in my country is generally Procambarus clarkii, a high-value aquatic species. It is high in protein and low in fat, and is a high-quality aquatic delicacy. It is in great demand in the international market, especially in Europe and the United States. In addition, lobsters also have a high added value in the industrial market. The shell of Procambarus clarkii contains substances such as astaxanthin, astaxanthin, and chitin, which are widely used in food, medicine, industry, printing and dyeing, daily chemicals, and agriculture.

[0003] The main method of raising crayfish in my country is rice-shrimp farming, in addition to pond farming, lotus field farming and wild rice field farming. However, lobster farming is still in an extensive mode, and the growth status of the breeding objects needs to be manually placed in cages, and then the status of the lobsters in the cages is evaluated. On the one hand, manual evaluation requires sending people to patrol the ponds and salvage, and on the other hand, the shrimps in the cages will kill each other, resulting in reduced production. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an Internet of Things monitoring device and a monitoring method for the growth status of rice field lobsters. The Internet of Things device system includes a shrimp collecting module, a solar charging module, a telescopic image recognition module, and a cloud data server. By regularly identifying the area of ​​shrimp heads and shrimp bodies in the shrimp collector, the growth status of lobsters in the rice field is monitored using mathematical statistics methods, thereby realizing production capacity prediction, growth status monitoring, and feeding and medication decision-making in the lobster farming process.

[0005] In order to solve the above problems, the present invention is achieved through the following technical solutions:

[0006] The present invention discloses an Internet of Things monitoring method for the growth status of rice field lobsters. The Internet of Things monitoring method is implemented by an Internet of Things monitoring device, and specifically comprises the following steps:

[0007] Step 1: Collect lobster images through a camera and learn the statistics of lobster pixel features;

[0008] Step 2: Collect the original images of lobster growth status monitoring through the lobster collection module, and remove the background after collection;

[0009] Step 3: Perform feature recognition on the original lobster growth status monitoring image obtained in step 2 to monitor the growth changes of the lobster.

[0010] A further improvement of the present invention is that the step 1 of acquiring and processing the lobster image specifically includes the following steps:

[0011] Step 1.1, select samples evenly according to the size of lobsters from 1 qian to 10 qian, use the camera to collect photos of lobsters ranging from green shrimp to dark red shrimp, manually process the collected lobster photos, replace the background pixels R, G, B with (0,0,0), count the lobster foreground RGB three-color pixels according to the pixel (x,y) sequence number, and use the sum of the R, G, B three-color distance as the partition distance, then the total distance D of the lobster sample foreground pixels is:

[0012]

[0013] Among them, (x, y) is the horizontal and vertical sequence number of the pixel;

[0014] Step 1.2, perform Gaussian interval statistics on the total distance of foreground pixels of lobster samples;

[0015] Step 1.3, count the foreground pixels of the lobster samples, and sort them from large to small according to the percentage of pixel intervals in the total pixels, and count them as Q1, Q2, Q3, Q4, and Q5 respectively;

[0016] Step 1.4, lift the shrimp heads to the surface of the water, and count the background pixels of the shrimp cage and the water surface. If the total distance D of the foreground pixels of the lobster sample cannot be distinguished from Q1, Q2, Q3, Q4, and Q5, then change the color of the shrimp head screen until it can be clearly distinguished from Q1, Q2, Q3, Q4, and Q5.

[0017] A further improvement of the present invention is that the step 2 specifically comprises the following steps:

[0018] Step 2.1, the shrimp collecting module is timed, the telescopic motor (301) is reversed, and the RS485 laser distance measurement (302) reaches the camera height;

[0019] Step 2.2, turn on the camera to obtain the original image of the lobster growth status monitoring of the lobsters in the lobster collection module;

[0020] Step 2.3: The original image of lobster growth status monitoring obtained in step 2.2 is transmitted to the cloud server through the 4G module, and the image is pixel filtered to only keep the pixels within the Q1, Q2, Q3, Q4, and Q5 pixel distances.

[0021] A further improvement of the present invention is that the step 3 specifically comprises the following steps:

[0022] Step 3.1, binarizing the original lobster growth status monitoring image after background removal to obtain a binarized lobster growth status monitoring image;

[0023] Step 3.2, the collected binarized lobster growth status monitoring image is rotated by 5°-90°, with a unit interval of 5°, and the rotated images are marked as G1, G2, ..., G18, and one of the rotated images is marked as Gi, where i = 1-18;

[0024] Step 3.3, taking 10×10 pixels as intervals, perform horizontal pixel block statistics on Gi. When the black pixels in a pixel block account for 30%, the pixel block is counted as a lobster foreground pixel. The number of lobster background blocks in each row is counted and marked as Wij, where i is the image number and j is the row number.

[0025] Step 3.4: The sum of the number of lobster background pixel blocks in each binarized lobster growth status monitoring image is marked as the lobster area Si.

[0026]

[0027] Among them, i is the rotated image number, and the pixel is 1920×1080;

[0028] Step 3.5, count the number of lobster foreground pixel blocks in each column according to 10×10 pixels, marked as Nik, where i is the image number and k is the column number;

[0029] Step 3.6, count the maximum value of the number of lobster foreground pixel blocks Nik in each column and the number of lobster background blocks Wij in each row in the rotated image Gi, mark them as the lobster length Li and lobster width Wi of the rotated image respectively, and mark the aspect ratio as Yi:

[0030] Li=(Nik)max

[0031] Wi=(Wij)max

[0032] Yi=Li / Wi;

[0033] Where i = 1-18

[0034] Step 3.7, sort the aspect ratios Yi, take the maximum value, and use it as the image for calculating the width and height of the lobster. Mark it with the serial number n, and then obtain the characteristic value information of the lobster, which is the area S, width W, and length L:

[0035] S=Sn

[0036] L=Ln

[0037] W=Wn

[0038] Step 3.8, statistically analyze the lobster characteristic values ​​obtained in step 3.7 within one day, mark them as Sm, Lm, Wm, where m is a sequence number, control the lifting time of the shrimp cage to be 15 minutes, and obtain 0 or 1 shrimp with a high probability, calculate the mean of the aspect ratio Ym, and eliminate the sample data that deviates from the mean difference by 50%;

[0039] Step 3.9, calculate the mean values ​​of the area, width, and length of the lobster samples collected in one day and excluding those deviating from the mean, and record the breeding time;

[0040] Step 3.10: Compare Su, Wu, and Lu longitudinally according to the culture time to monitor the growth changes of lobsters.

[0041] The present invention provides an Internet of Things monitoring device for an Internet of Things monitoring method of the growth status of rice field lobsters. The Internet of Things monitoring device comprises a lobster collecting module which is placed on a lobster feeding platform and collects original images for monitoring the growth status of lobsters. A solar charging module is arranged above the lobster collecting module. A telescopic motor in a telescopic image recognition module is installed on a mounting platform in the solar charging module. The telescopic image recognition module acquires the original images for monitoring the growth status of lobsters collected by the lobster collecting module. The acquired original images for monitoring the growth status of lobsters are transmitted to a cloud data server through a 4G channel to monitor the growth changes of lobsters.

[0042] A further improvement of the present invention is that the shrimp collecting module is a shrimp cage structure, including a net cage placed in the lobster foraging area, a shrimp collecting head, an escape plate, an escape spring, a lifting screw hole and an escape column hole. The net cage is an ingress-ingress-ingress structure. The shrimp collecting head and the net cage are softly connected through a channel composed of an elastic connecting net, and the shrimp collecting head and the net cage are correspondingly provided with inlets and outlets, so that the lobster enters the shrimp collecting head from the net cage. After the shrimp collecting head rises, it is retracted and extended through the elastic connecting net without causing damage to the channel. The escape plate is connected to the bottom of the shrimp collecting head through an escape spring. When the escape rod in the telescopic image recognition module descends to the escape column hole, the escape plate is propped open, and the lobster falls into the water to escape. The lifting screw hole is at the top of the shrimp collecting module, and the lifting rod of the telescopic image recognition module is threadedly connected to make the shrimp collecting module leave the water surface. The telescopic image recognition module identifies the size and number of lobsters. The escape column hole is at the top of the shrimp collecting module, and the escape plate is propped open by the escape rod.

[0043] A further improvement of the present invention is that the release of the lobsters is achieved by the structure of the net cage that allows entry but not exit, specifically: the telescopic motor is reversed, the RS485 laser distance measurement reaches the release height, the escape plate is opened, the lobsters enter the water, the telescopic motor is forward rotated, the RS485 laser distance measurement reaches the height for placement into the water, and the collected shrimp heads enter the water.

[0044] The further improvement of the present invention is that the solar charging module includes a solar panel, an Internet of Things module with an RS485 interface, an installation platform, a controller and a column. The solar panel is installed on the top of the column. The solar panel is 100W. The installation platform is fixed on the upper part of the column and close to the solar panel. The camera, controller, telescopic motor and Internet of Things module are installed on the installation platform. The controller controls the charging and discharging of the solar panel and has an RS485 communication interface and a 4-way DI / DO interface, which are respectively used to transmit remote control commands, lithium battery status collection and telescopic motor lifting control. The controller adopts Kunlun Tongtai's 7-inch MCGS PRO equipped with a 4G Internet of Things card as the main control unit, and has an external RS485 voltage collection module, a solar charging and discharging management module, and an RS485 DI / DO acquisition module, voltage acquisition module voltage acquisition range 0V-36V, main control unit built-in 4G traffic card, with RS485 interface and RJ45 interface, solar charge and discharge management module works independently, with voltage monitoring output port, connected to voltage acquisition module, RS485 interface adopts MODBUSRTU protocol, through RS485 voltage acquisition module to collect lithium battery voltage, through RS485 DI / DO acquisition module to collect 4-way DI information, control 4-way DO output, RJ45 interface is TCP / IP protocol, used to transmit camera image signal, image acquisition mode includes video mode and photo transmission mode, through MCGS PRO video acquisition interface program to achieve, using photo transmission mode can save traffic in traffic card.

[0045] The further improvement of the present invention is that the telescopic image recognition module includes a telescopic motor, an RS485 laser ranging, a camera with an RJ45 interface, a lifting rod, and an escape rod. The telescopic motor is installed on a mounting platform, the lifting rod is connected to the telescopic motor through a coupling, the escape rod is installed below the mounting platform, and the escape rod and the lifting rod are aligned with the escape column hole and the lifting screw hole in the shrimp collecting module. The controller controls the forward and reverse rotation and power-on of the motor respectively by controlling 2-way DO to realize the forward and reverse control of the motor. The camera is installed below the mounting platform, the RS485 laser ranging is installed on the top of the mounting platform by drilling a hole to expose the laser probe, the RS485 laser ranging and the RS485 communication interface of the controller are connected to the RS485 interface of the Internet of Things module, and the RJ45 interface of the camera is connected to the RJ45 interface of the Internet of Things module.

[0046] A further improvement of the present invention is that the cloud data server includes an Internet of Things terminal and a cloud data server, and the photos taken by the camera in the solar charging module interact with the Internet of Things terminal and the cloud data server through a 4G channel. At the same time, the Internet of Things terminal publishes controls through the MQTT protocol and is accessed by a mobile phone to achieve mobile phone control and viewing.

[0047] The beneficial effects of the present invention are:

[0048] The Internet of Things monitoring device of the present invention uses a telescopic structure to lift the lobster cage out of the water, and performs lobster imaging in the lobster collection module. The lobster image is collected once every 15 minutes, which can basically ensure that there is only one lobster in the lobster collection module most of the time. By counting the lobster foreground pixels and setting a black imaging background, a mixed background elimination algorithm is implemented to eliminate the lobster background and leave only the lobster image; the binary lobster foreground image basically maintains the lobster shape, but because the imaging angle is uncertain, the present invention adopts a fixed rotation method. When the acquired length value and width ratio are the maximum, the imaging is expressed as standard imaging. At this time, the number of horizontal black pixel blocks and the number of vertical black pixels can basically express the length and width of the lobster, and the number of black pixel blocks can also express the individual size of the lobster.

[0049] The image processing method of the present invention is simple and has clear principles. It monitors the growth status in proportion to the growth period without the need for data calibration and is easy to use. At the same time, a certain range of lobster collection frequency also expresses the lobster production in the breeding area. Since the entire device collects data outside the water, the cost is much lower than that of collecting data inside the water, which is suitable for large-scale promotion.

[0050] In summary, the method of the present invention uses the Internet of Things to monitor the growth status of rice field lobsters, and realizes the production capacity prediction during lobster farming. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the overall structure of the Internet of Things monitoring device of the present invention.

[0052] Figure 2 It is a structural schematic diagram of the shrimp collecting module of the present invention.

[0053] Figure 3 It is a structural schematic diagram of the solar charging module of the present invention.

[0054] Figure 4 Schematic diagram of the control of the controller of the present invention.

[0055] Figure 5 It is a structural schematic diagram of the telescopic image recognition module of the present invention.

[0056] Figure 6 It is a structural diagram of the cloud data server of the present invention.

[0057] Figure 7 1 is a binary rotation diagram of a mature lobster in an embodiment of the present invention.

[0058] Figure 8 It is a schematic diagram of the binary rotation of the lobster larvae in an embodiment of the present invention.

[0059] Fig. 9 Schematic diagram of the principle of extracting the length, width and area morphological parameters of lobsters in an embodiment of the present invention.

[0060] Fig.10 It is a flow chart of the monitoring method of the present invention. DETAILED DESCRIPTION

[0061] The following will disclose the embodiments of the present invention with drawings. For the purpose of clear description, many practical details will be described together in the following description. However, it should be understood that these practical details should not be used to limit the present invention. That is, in some embodiments of the present invention, these practical details are not necessary. In addition, for the purpose of simplifying the drawings, some conventional structures and components will be depicted in a simple schematic manner in the drawings.

[0062] like Figure 1 As shown, the present invention is an Internet of Things monitoring device for the growth status of rice field lobsters, the device includes a lobster collecting module 4 placed on a lobster feeding platform and collecting original images for monitoring the growth status of lobsters, a solar charging module 1 is arranged above the lobster collecting module 4, the telescopic motor 103 in the telescopic image recognition module 3 is installed on the installation platform 103 in the solar charging module 1, the telescopic image recognition module 3 obtains the original image of lobster growth status monitoring collected by the lobster collecting module 4, and transmits the obtained original image of lobster growth status monitoring to the cloud data server 2 through the 4G channel to monitor the growth changes of lobsters.

[0063] Specifically, Figure 2As shown, the shrimp collecting module 4 is a shrimp cage structure, including a net cage 401 placed in the lobster foraging area, a shrimp collecting head 404, an escape plate 402, an escape spring 403, a lifting screw hole 405 and an escape column hole 406. The net cage 401 is an enterable but not exitable structure. The release of the lobsters is achieved through the enterable but not exitable structure of the net cage 401 as follows: the telescopic motor 301 reverses, the RS485 laser rangefinder 302 measures the distance to reach the release height, the escape plate 402 opens, the lobsters enter the water, the telescopic motor 301 rotates forward, the RS485 laser rangefinder 302 measures the distance to reach the height for placing in the water, and the shrimp collecting head 404 enters the water. The shrimp collecting head 404 is softly connected to the net cage 401 through a channel composed of an elastic connection net, and the shrimp collecting head 404 and the net cage 401 are correspondingly provided with an inlet and outlet, so that the lobster enters the shrimp collecting head 404 from the net cage 401. After the shrimp collecting head 404 rises, it is retracted and extended through the elastic connection net without causing damage to the channel. The escape plate 402 is connected to the bottom of the shrimp collecting head 404 through an escape spring 403. When the escape rod 305 in the telescopic image recognition module 3 descends to the escape column hole 406, the escape plate 402 is opened, and the lobster falls into the water to escape. The lifting screw hole 405 is at the top of the shrimp collecting module 4, and the lifting rod 304 of the telescopic image recognition module 3 is threadedly connected, so that the shrimp collecting module 4 leaves the water surface. The telescopic image recognition module 3 identifies the size and number of lobsters. The escape column hole 406 is at the top of the shrimp collecting module 4, and the escape plate 402 is opened through the escape rod (305).

[0064] like Figure 3-4As shown, the solar charging module 1 includes a solar panel 101, an Internet of Things module 102 with an RS485 interface, a mounting platform 103, a controller 104 and a column 105. The solar panel 101 is mounted on the top of the column 105. The solar panel 101 is 100W. The mounting platform 103 is fixed on the upper part of the column 105 and close to the solar panel 101. On the one hand, it has a certain rainproof effect, so that the waterproof and dustproof level of the camera, controller, telescopic motor and Internet of Things module can be reduced. The installation platform 103 is used to install a camera 303, a controller 104, a telescopic motor, and an Internet of Things module 102. The controller 104 controls the charging and discharging of the solar panel and has an RS485 communication interface and a 4-way DI / DO interface, which are respectively used to transmit remote control commands, lithium battery status collection, and telescopic motor lifting control. The controller 104 uses Kunlun Tongtai's 7-inch MCGS PRO equipped with a 4G Internet of Things card as the main control unit, with an external RS485 voltage collection module, a solar charge and discharge management module, and an RS485 DI / DO collection module. The voltage collection module has a voltage collection range of 0V-36V. The main control unit has a built-in 4G traffic card, an RS485 interface and an RJ45 interface. The solar charge and discharge management module works independently and has a voltage monitoring output port to connect the voltage collection module. The RS485 interface uses MODBUS RTU protocol, collects lithium battery voltage through RS485 voltage acquisition module, collects 4-way DI information through RS485 DI / DO acquisition module, controls 4-way DO output, RJ45 interface is TCP / IP protocol, used to transmit camera image signal, image acquisition mode includes video mode and photo transmission mode, which is realized through MCGS PRO video acquisition interface program, and photo transmission mode can save traffic in traffic card.

[0065] like Figure 5As shown, the telescopic image recognition module 3 includes a telescopic motor 301, an RS485 laser rangefinder 302, a camera 303 with an RJ45 interface, a lifting rod 304, and an escape rod 305. The telescopic motor 301 is installed on the installation platform 103, and the lifting rod 304 is connected to the telescopic motor 301 through a coupling. The escape rod 305 is installed below the installation platform 103 and the escape rod 305 and the lifting rod 304 are aligned with the escape column hole 406 and the lifting screw hole 405 in the shrimp collecting module 4. The controller 104 controls the forward and reverse rotation and power-on of the motor by controlling two DOs respectively, thereby realizing the control of the forward and reverse rotation of the motor. The camera 303 is installed below the mounting platform 103. The RS485 laser rangefinder 302 is installed on the top of the mounting platform 103 by drilling a hole to expose the laser probe. The RS485 communication interface of the RS485 laser rangefinder 302 and the controller 104 is connected to the RS485 interface of the Internet of Things module 102, and the RJ45 interface of the camera 303 is connected to the RJ45 interface of the Internet of Things module 102.

[0066] like Figure 6 As shown, the cloud data server 2 includes an Internet of Things terminal 201 and a cloud data server 2. The photos taken by the camera 303 in the solar charging module 1 interact with the Internet of Things terminal 201 and the cloud data server 2 through the 4G channel. At the same time, the Internet of Things terminal 201 publishes controls through the MQTT protocol and is accessed by a mobile phone to achieve mobile phone control and viewing.

[0067] like Fig.10 As shown, based on the above-mentioned Internet of Things monitoring device, the present invention also provides an Internet of Things monitoring method for the growth status of rice field lobsters, which specifically includes the following steps:

[0068] Step 1: Collect lobster images through the camera and learn the statistics of lobster pixel features.

[0069] The specific steps include:

[0070] Step 1.1, according to the size of lobsters from 1 qian to 10 qian, evenly select 140 samples, use the camera to collect photos of lobsters from green shrimp to dark red shrimp, manually process the collected lobster photos, replace the background pixels R, G, B with (0,0,0), count the lobster foreground RGB three-color pixels according to the pixel (x,y) sequence number, and use the sum of the R, G, B three-color distance as the partition distance, then the total distance D of the lobster sample foreground pixels is 443, and the calculation formula is:

[0071]

[0072] Among them, (x, y) is the horizontal and vertical sequence number of the pixel;

[0073] Step 1.2: Perform Gaussian interval statistics on the total distance of foreground pixels of the 140 lobster samples, with every 20 being a pixel distance, i.e. [0,20], (20,40], ... (420,443], for a total of 22 intervals;

[0074] Step 1.3: Count the foreground pixels of the 140 lobster samples. The first 5 pixel intervals account for 90% of the total pixels. Sort them from large to small according to the percentage of pixel intervals to the total pixels, and count them as Q1, Q2, Q3, Q4, and Q5 respectively;

[0075] Step 1.4, lift the shrimp head 300mm above the water surface, count the background pixels of the shrimp cage and the water surface, if the total distance D of the foreground pixels of the lobster sample cannot be distinguished from Q1, Q2, Q3, Q4, Q5, then change the color of the shrimp head screen until it is significantly distinguished from Q1, Q2, Q3, Q4, Q5. The top grid of the shrimp head is white with 10 mesh, and the bottom grid of the shrimp head is black with 100 mesh, which has a better effect of distinguishing the background.

[0076] Step 2: Collect the original images of lobster growth status monitoring through the lobster collection module, and remove the background after collection, including the following steps:

[0077] Step 2.1, the shrimp collecting module is timed for 15 minutes, the telescopic motor 301 is reversed, and the RS485 laser distance measurement 302 reaches the camera height;

[0078] Step 2.2, turn on the camera to obtain the original image of the lobster growth status monitoring of the lobsters in the lobster collection module;

[0079] Step 2.3: The original image of lobster growth status monitoring obtained in step 2.2 is transmitted to the cloud server through the 4G module, and the image is pixel filtered in the intervals of [0,20], (20,40], ... (420,443], leaving only the pixels within the pixel distances of Q1, Q2, Q3, Q4, and Q5.

[0080] Step 3: Perform feature recognition on the original image of lobster growth status monitoring obtained in step 2 to monitor the growth changes of lobsters, such as Figure 7-8 The specific steps include:

[0081] Step 3.1, binarizing the original lobster growth status monitoring image after background removal to obtain a binarized lobster growth status monitoring image;

[0082] Step 3.2, the collected binarized lobster growth status monitoring image is rotated by 5°-90°, with a unit interval of 5°, and the rotated images are marked as G1, G2, ..., G18, and one of the rotated images is marked as Gi, where i = 1-18;

[0083] Step 3.3, taking 10×10 pixels as the interval, perform horizontal pixel block statistics on Gi. When the black pixels account for 30% of the pixel block, the pixel block is counted as a lobster foreground pixel. The number of lobster background blocks in each row is counted and marked as Wij, where i is the image number and j is the row number. Fig. 9 shown.

[0084] Step 3.4: The sum of the number of lobster background pixel blocks in each binarized lobster growth status monitoring image is marked as the lobster area Si.

[0085]

[0086] Among them, i is the rotated image number, and the pixel is 1920×1080;

[0087] Step 3.5, count the number of lobster foreground pixel blocks in each column according to 10×10 pixels, marked as Nik, where i is the image number and k is the column number;

[0088] Step 3.6, count the maximum value of the number of lobster foreground pixel blocks Nik in each column and the number of lobster background blocks Wij in each row in the rotated image Gi, mark them as the lobster length Li and lobster width Wi of the rotated image respectively, and mark the aspect ratio as Yi:

[0089] Li=(Nik)max

[0090] Wi=(Wij)max

[0091] Yi=Li / Wi;

[0092] Where i = 1-18

[0093] Step 3.7, sort the aspect ratios Yi, take the maximum value, and use it as the image for calculating the width and height of the lobster. Mark it with the serial number n, and then obtain the characteristic value information of the lobster, which is the area S, width W, and length L:

[0094] S=Sn

[0095] L=Ln

[0096] W=Wn

[0097] Step 3.8, statistically analyze the lobster characteristic values ​​obtained in step 3.7 within one day, mark them as Sm, Lm, Wm, where m is a sequence number, control the lifting time of the shrimp cage to be 15 minutes, and obtain 0 or 1 shrimp with a high probability, calculate the mean of the aspect ratio Ym, and eliminate the sample data that deviates from the mean difference by 50%;

[0098] Step 3.9, calculate the mean values ​​of the area, width, and length of the lobster samples collected in one day and excluding those deviating from the mean, and record the breeding time;

[0099] Step 3.10: Compare Su, Wu, and Lu longitudinally according to the culture time to monitor the growth changes of lobsters.

[0100] The present invention periodically identifies the area of ​​shrimp heads and shrimp bodies in a shrimp collector and uses mathematical statistics to monitor the growth status of lobsters in rice fields, thereby realizing production capacity prediction, growth status monitoring and feeding and medication decision-making in the lobster breeding process.

[0101] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An Internet of Things monitoring method for the growth status of rice field lobsters, characterized in that: The Internet of Things monitoring method is implemented by an Internet of Things monitoring device, and specifically includes the following steps: Step 1: Collect lobster images through a camera and learn the statistics of lobster pixel features; Step 2: Collect the original images of lobster growth status monitoring through the lobster collection module, and remove the background after collection; Step 3: Perform feature recognition on the original lobster growth status monitoring image obtained in step 2 to monitor the growth changes of the lobster.

2. The Internet of Things monitoring method for the growth status of rice field lobsters according to claim 1 is characterized in that: The step 1 of acquiring and processing the lobster image specifically includes the following steps: Step 1.1, select samples according to the size of lobsters from 1 qian to 10 qian, use the camera to collect photos of lobsters ranging from green shrimp to dark red shrimp, manually process the collected lobster photos, replace the background pixels R, G, B with (0,0,0), count the lobster foreground RGB three-color pixels according to the pixel (x, y) sequence number, and use the sum of the R, G, B three-color distance as the partition distance, then the total distance of the lobster sample foreground pixels is D: Where (x, y) is the horizontal and vertical serial number of the pixel; Step 1.2, perform Gaussian interval statistics on the total distance of foreground pixels of lobster samples; Step 1.3, count the foreground pixels of the lobster samples, and sort them from large to small according to the percentage of pixel intervals in the total pixels, and count them as Q1, Q2, Q3, Q4, and Q5 respectively; Step 1.4, lift the shrimp head to the surface of the water, and count the background pixels of the shrimp cage and the water surface. If the total distance D of the foreground pixels of the lobster sample cannot be distinguished from Q1, Q2, Q3, Q4, and Q5, then change the color of the shrimp head screen until it can be distinguished from Q1, Q2, Q3, Q4, and Q5.

3. The Internet of Things monitoring method for the growth status of rice field lobsters according to claim 1 is characterized in that: The step 2 specifically includes the following steps: Step 2.1, the shrimp collecting module is timed, the telescopic motor (301) is reversed, and the RS485 laser distance measurement (302) reaches the camera height; Step 2.2, turn on the camera to obtain the original image of the lobster growth status monitoring of the lobsters in the lobster collection module; Step 2.3: The original image of lobster growth status monitoring obtained in step 2.2 is transmitted to the cloud server through the 4G module, and the image is pixel filtered to only keep the pixels within the Q1, Q2, Q3, Q4, and Q5 pixel distances.

4. The Internet of Things monitoring method for the growth status of rice field lobsters according to claim 1 is characterized in that: The step 3 specifically includes the following steps: Step 3.1, binarizing the original lobster growth status monitoring image after background removal to obtain a binarized lobster growth status monitoring image; Step 3.2, the collected binarized lobster growth status monitoring image is rotated by 5°-90°, with a unit interval of 5°, and the rotated images are marked as G1, G2, ..., G18, and one of the rotated images is marked as Gi, where i = 1-18; Step 3.3, taking 10×10 pixels as intervals, perform horizontal pixel block statistics on Gi. When the black pixels in a pixel block account for 30%, the pixel block is counted as a lobster foreground pixel. The number of lobster background blocks in each row is counted and marked as Wij, where i is the image number and j is the row number. Step 3.4: The sum of the number of lobster background pixel blocks in each binarized lobster growth status monitoring image is marked as the lobster area Si. Among them, i is the rotated image number, and the pixel is 1920×1080; Step 3.5, count the number of lobster foreground pixel blocks in each column according to 10×10 pixels, marked as Nik, where i is the image number and k is the column number; Step 3.6, count the maximum value of the number of lobster foreground pixel blocks Nik in each column and the number of lobster background blocks Wij in each row in the rotated image Gi, mark them as the lobster length Li and lobster width Wi of the rotated image respectively, and mark the aspect ratio as Yi: Li=(Nik)max Wi=(Wij)max Yi=Li / Wi; Where i = 1-18 Step 3.7, sort the aspect ratios Yi, take the maximum value, and use it as the image for calculating the width and height of the lobster. Mark it with the serial number n, and then obtain the characteristic value information of the lobster, which is the area S, width W, and length L: S=Sn L=Ln W=Wn Step 3.8, statistically analyze the lobster characteristic values ​​obtained in step 3.7 within one day, mark them as Sm, Lm, Wm, respectively, where m is a sequence number, control the lifting time of the shrimp cage to be 15 minutes, obtain 0 or 1 shrimp, calculate the mean of the aspect ratio Ym, and eliminate the sample data that deviates from the mean difference by 50%; Step 3.9, calculate the mean values ​​of the area, width, and length of the lobster samples collected in one day and excluding those deviating from the mean, and record the breeding time; Step 3.10: Compare Su, Wu, and Lu longitudinally according to the culture time to monitor the growth changes of lobsters.

5. An Internet of Things monitoring device for implementing the Internet of Things monitoring method for the growth status of rice field lobsters according to any one of claims 1 to 4, characterized in that: The Internet of Things monitoring device comprises a shrimp collecting module (4) placed on a lobster feeding platform and collecting original images of lobster growth status monitoring. A solar charging module (1) is arranged above the shrimp collecting module (4). A telescopic motor (103) in a telescopic image recognition module (3) is installed on a mounting platform (103) in the solar charging module (1). The telescopic image recognition module (3) obtains the original images of lobster growth status monitoring collected by the shrimp collecting module (4). The obtained original images of lobster growth status monitoring are transmitted to a cloud data server (2) through a 4G channel to monitor the growth changes of the lobster.

6. The Internet of Things monitoring device for the growth status of rice field lobsters according to claim 5, characterized in that: The shrimp collecting module (4) is a shrimp cage structure, comprising a net cage (401) placed in a lobster foraging area, a shrimp collecting head (404), an escape plate (402), an escape spring (403), a lifting screw hole (405) and an escape column hole (406); the net cage (401) is a structure that allows entry but not exit; the shrimp collecting head (404) and the net cage (401) are softly connected via a channel composed of an elastic connection net, and the shrimp collecting head (404) and the net cage (401) are correspondingly provided with an inlet and outlet, so that the lobster enters the shrimp collecting head (404) from the net cage (401); after the shrimp collecting head (404) rises, it enters the net through the elastic connection net. The escape plate (402) is connected to the bottom of the shrimp collecting head (404) through an escape spring (403). When the escape rod (305) in the telescopic image recognition module (3) descends to the escape column hole (406), the escape plate (402) is opened, and the lobsters fall into the water to escape. The lifting screw hole (405) is at the top of the shrimp collecting module (4), and the lifting rod (304) of the telescopic image recognition module (3) is connected by thread, so that the shrimp collecting module (4) leaves the water surface. The telescopic image recognition module (3) identifies the size and number of lobsters. The escape column hole (406) is at the top of the shrimp collecting module (4), and the escape plate (402) is opened through the escape rod (305).

7. The Internet of Things monitoring device for the growth status of rice field lobsters according to claim 5, characterized in that: The release of the lobsters by the structure of the net cage (401) that allows entry but not exit is achieved by: the telescopic motor (301) rotates in the reverse direction, the RS485 laser distance measurement (302) measures the distance to reach the release height, the escape plate (402) opens, the lobsters enter the water, the telescopic motor (301) rotates forward, the RS485 laser distance measurement (302) measures the distance to reach the height for being put into the water, and the lobster heads (404) enter the water.

8. The Internet of Things monitoring device for the growth status of rice field lobsters according to claim 5, characterized in that: The solar charging module (1) comprises a solar panel (101), an Internet of Things module (102) with an RS485 interface, a mounting platform (103), a controller (104) and a column (105). The solar panel (101) is mounted on the top of the column (105). The mounting platform (103) is fixed on the upper part of the column (105) and close to the solar panel (101). The mounting platform (103) is used to install a camera (303), a controller (104), a telescopic motor and an Internet of Things module (102). The controller (104) is configured with a 7-inch MCGS PRO with a 4G Internet of Things card as a main control unit, an external RS485 voltage acquisition module, a solar charge and discharge management module and an RS485 DI / DO acquisition module. The main control unit has a built-in 4G traffic card and has an RS485 interface and an RJ45 interface. The solar charge and discharge management module works independently and has a voltage monitoring output port connected to the voltage acquisition module. The RS485 interface adopts MODBUS RTU protocol, collects lithium battery voltage through RS485 voltage acquisition module, collects 4-way DI information through RS485 DI / DO acquisition module, controls 4-way DO output, RJ45 interface is TCP / IP protocol, used to transmit camera image signal, image acquisition mode includes video mode and photo transmission mode, which is realized through MCGS PRO video acquisition interface program, and photo transmission mode can save traffic in traffic card.

9. The Internet of Things monitoring device for the growth status of rice field lobsters according to claim 8, characterized in that: The telescopic image recognition module (3) comprises a telescopic motor (301), an RS485 laser distance measurement (302), a camera (303) with an RJ45 interface, a lifting rod (304), and an escape rod (305); the telescopic motor (301) is mounted on a mounting platform (103); the lifting rod (304) is connected to the telescopic motor (301) via a coupling; the escape rod (305) is mounted below the mounting platform (103); and the escape rod (305) and the lifting rod (304) are aligned with the escape column hole (406) and the lifting screw hole (406) in the shrimp collecting module (4). 405), the controller (104) controls the forward and reverse rotation and power-on of the motor by controlling the two DOs respectively, thereby realizing the control of the forward and reverse rotation of the motor. The camera (303) is installed below the mounting platform (103), and the RS485 laser rangefinder (302) is installed on the top of the mounting platform (103) by drilling a hole to expose the laser probe. The RS485 communication interface of the RS485 laser rangefinder (302) and the controller (104) is connected to the RS485 interface of the Internet of Things module (102), and the RJ45 interface of the camera (303) is connected to the RJ45 interface of the Internet of Things module (102).

10. The Internet of Things monitoring device for the growth status of rice field lobsters according to claim 5, characterized in that: The cloud data server (2) comprises an Internet of Things terminal (201) and a cloud data server (2); the photos taken by the camera (303) in the solar charging module (1) interact with the Internet of Things terminal (201) and the cloud data server (2) via a 4G channel; at the same time, the Internet of Things terminal (201) publishes controls via the MQTT protocol and is accessed by a mobile phone, thereby realizing mobile phone control and viewing.

Images