CN117958193A

CN117958193A - Cultivation operation worker ship

Info

Publication number: CN117958193A
Application number: CN202410073383.9A
Authority: CN
Inventors: 郑荣才; 陈凯; 元轲新; 陈大勇; 郭燕萍; 张浩健
Original assignee: Southern Marine Science and Engineering Guangdong Laboratory Zhanjiang
Current assignee: Southern Marine Science and Engineering Guangdong Laboratory Zhanjiang
Priority date: 2024-01-17
Filing date: 2024-01-17
Publication date: 2024-05-03

Abstract

The application discloses a breeding operation ship, which relates to the technical field of ships, and comprises a ship body, a propulsion device, a ship-borne cruise control module, at least two feed cabins, a feeding device and a net cage identification module; the propulsion device is arranged on the ship body; the on-board cruise control module is used for acquiring the net cage coordinates sent by the shore-based monitoring system, converting the net cage coordinates into a cruise path, and controlling the propulsion device to drive the ship body to travel to each cultivation net cage according to the cruise path; the feed cabin is arranged on the ship body and respectively stores different kinds of fish feeds; the feeding end of the feeding device is connected with the feed cabin; the net cage identification module is used for identifying the type of the culture net cage and controlling the feeding device to deliver the fish feed in the corresponding feed cabin to the corresponding culture net cage. The breeding operation worker ship can solve the technical problems that the efficiency of the mode of feeding the breeding net cage with feed in a manual mode is low, the labor cost is consumed, and the automation and intelligent degree is low.

Description

Cultivation operation worker ship

Technical Field

The application relates to the technical field of ships, in particular to a cultivation operation worker ship.

Background

With the rapid development of the aquaculture industry in China, the cultivation scale of the fishing farm is gradually increased, and the fishing farm is gradually developed from offshore to deep and open sea. The culture net cage can construct a closed culture space in a culture water body, is widely applied to the field of fishery culture and shows good economic and social benefits.

However, in the conventional cage culture mode, when the fishes in the culture cage are required to be fed with feed, the farmers often need to drive ships into the culture sea area and travel to each culture cage one by one to perform manual feeding operation, the feeding efficiency is low, the labor cost is consumed, and the automation and intelligence degree is low.

Disclosure of Invention

The application aims to provide a cultivation operation ship, and aims to solve the technical problems that the efficiency of a mode of feeding feed to a cultivation net cage in a manual mode is low, the labor cost is consumed, and the automation and the intelligent degree are low.

The application adopts the following technical scheme to achieve the aim of the application:

a farming worker vessel, the farming worker vessel comprising:

A hull;

The propulsion device is arranged on the ship body;

The on-board cruise control module is electrically connected with the propulsion device; the shipborne cruise control module is used for acquiring the net cage coordinates sent by the shore-based monitoring system and converting the net cage coordinates into a cruise path based on a path planning algorithm; the on-board cruise control module is used for converting the angle parameter corresponding to the cruise path into a power parameter and sending the power parameter to the propulsion device so as to control the propulsion device to drive the ship body to travel to each cultivation net cage according to the cruise path;

at least two feed cabins arranged on the ship body; the at least two feed cabins are respectively used for storing different types of fish feeds;

The feeding end of the feeding device is connected with the at least two feed cabins;

The net cage identification module is electrically connected with the feeding device; the net cage identification module is used for identifying the type of the culture net cage based on an image identification algorithm and sending a corresponding feeding signal to the feeding device so as to control the feeding device to absorb the corresponding fish feed in the feed cabin and send the fish feed to the corresponding culture net cage through a feeding end of the feeding device.

Further, the breeding operation ship comprises an underwater inspection device and a prompting device; the underwater patrol device is arranged on the ship body, a target netting image is prestored in the underwater patrol device, and the underwater patrol device is electrically connected with the prompting device;

the underwater inspection device is used for moving around the culture net cage along with the movement of the ship body; the underwater inspection device is provided with a shooting end, and the shooting end is used for acquiring a current netting image of the cultivation net cage; the underwater inspection device is used for comparing the acquired current netting image with the target netting image to obtain a comparison result; and the underwater inspection device is used for sending a first alarm signal to the prompting device when the comparison result meets a first preset alarm condition so as to trigger the prompting device to execute alarm operation.

Further, the breeding operation ship comprises an underwater inspection device and a prompting device; the underwater patrol device is arranged on the ship body and is electrically connected with the prompting device;

The underwater inspection device is used for moving around the culture net cage along with the movement of the ship body; the underwater inspection device is provided with a laser emission end and a shooting end, wherein the laser emission end is used for emitting a linear laser beam to the aquaculture net cage, and the shooting end is used for acquiring a discrete light spot image formed by the linear laser beam on a net part of the aquaculture net cage; the underwater inspection device is used for sending a second alarm signal to the prompting device when the light spot gap existing in the discrete light spot image meets a second preset alarm condition so as to trigger the prompting device to execute alarm operation.

Further, the breeding worker ship comprises a fish shooting device, a fish image analysis module, a counting module and a prompting device; the fish image analysis module is provided with a GMM Gaussian mixture model and a YOLO neural network model, and is electrically connected with the fish shooting device and the counting module, and the counting module is electrically connected with the prompting device;

The fish shooting device is used for acquiring a fish key frame image set of a water surface part of the culture net cage, and inputting the acquired fish key frame image set into the GMM Gaussian mixture model and the YOLO neural network model respectively for moving target detection operation so as to obtain a GMM fish target frame and a YOLO fish target frame respectively;

The fish image analysis module is used for selecting the GMM fish target frame or the YOLO fish target frame as a dead fish target frame; the counting module is used for counting the number of the dead fish target frames, and the counting module is used for sending a third alarm signal to the prompting device when the number of the dead fish target frames reaches a preset number threshold value so as to trigger the prompting device to execute alarm operation.

Further, the hull is provided with a breeding cabin, and the breeding operation ship comprises a suction device; one end of the suction device is communicated with the culture cabin, and the suction device is used for sucking the living fish in the culture net cage into the culture cabin.

Further, at least two pumping pipelines with height difference are arranged in the culture cabin; the pumping and draining pipeline is used for carrying out water supply operation or water draining operation on the water body in the culture cabin so as to form a simulated flow field in the culture cabin.

Further, the breeding operation work ship comprises a satellite positioning module and a prompting device; the satellite positioning module is electrically connected with the shipborne cruise control module, and the shipborne cruise control module is electrically connected with the prompting device;

The satellite positioning module is used for acquiring current position information of the corresponding culture net boxes when the ship body moves to each culture net box and sending the current position information to the ship-borne cruise control module; the on-board cruise control module is used for correcting the net cage coordinates according to the current position information, and is used for sending a fourth alarm signal to the prompting device when the difference value between the current position information and the net cage coordinates reaches a preset distance threshold value so as to trigger the prompting device to execute alarm operation.

Further, the breeding worker ship comprises a flow direction detection device and a flow speed detection device; the flow direction detection device and the flow speed detection device are electrically connected with the shipborne cruise control module;

the flow direction detection device is used for acquiring flow direction information in a preset area taking the culture net cage as a center and sending the flow direction information to the shipborne cruise control module, and the flow speed detection device is used for acquiring flow speed information in the preset area taking the culture net cage as a center and sending the flow speed information to the shipborne cruise control module;

the on-board cruise control module is used for taking a path of which the water flow direction points to the aquaculture net cage as a feeding path, and taking a region of which the flow speed on the feeding path reaches a preset flow speed threshold as a feeding region; the ship-borne cruise control module is used for converting the position parameters corresponding to the feeding area into power parameters and sending the power parameters to the propulsion device so as to control the propulsion device to drive the ship body to travel to the feeding area and perform feeding operation.

Further, the farming worker vessel includes an offshore communication module; the offshore communication module is electrically connected with the propulsion device and the feeding device;

The offshore communication module is used for receiving real-time remote control instructions sent by the shore-based monitoring system; the marine communication module is used for converting the real-time remote control instruction into a power parameter and sending the power parameter to the propulsion device so as to control the propulsion device to drive the ship body to travel; and the offshore communication module is used for converting the real-time remote control instruction into a driving parameter and sending the driving parameter to the feeding device so as to control the feeding device to execute feeding operation.

Further, the ship body is provided with a radar monitoring device and a visual monitoring device, and the breeding operation worker ship comprises a prompting device; the radar monitoring device and the visual monitoring device are electrically connected with the prompting device; the radar monitoring device and the visual monitoring device are used for sending a fifth alarm signal to the prompting device when an unknown target object is detected, so as to trigger the prompting device to execute alarm operation.

Compared with the prior art, the application has the beneficial effects that:

According to the cultivation operation worker ship provided by the application, the on-board cruise control module is used for acquiring the net cage coordinates sent by the shore-based monitoring system, converting the net cage coordinates into the cruise path based on the path planning algorithm, converting the angle parameters corresponding to the cruise path into the power parameters and sending the power parameters to the propulsion device, so that the propulsion device can be controlled to drive the ship body to travel to each cultivation net cage according to the cruise path; when the ship body reaches any cultivation net cage, the type of the cultivation net cage is identified through the net cage identification module, and a corresponding feeding signal is sent to the feeding device, so that the feeding device is controlled to absorb fish feed in a corresponding feed cabin and send the fish feed to the cultivation net cage through a feeding end of the feeding device, and the function of automatically and selectively feeding according to cultivation types of different cultivation net cages is realized; based on the setting, automatic feeding of the culture net cage is realized, feeding efficiency is improved, labor cost is saved, and the intelligent degree of feeding operation of the culture operation ship is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic view of a modular connection structure of an embodiment of a farming vessel according to the present application;

FIG. 2 is a schematic view showing the overall structure of an embodiment of the cultivation operation ship of the present application.

Reference numerals illustrate:

Reference numerals	Name of the name	Reference numerals	Name of the name
				1	Ship body	12	Cultivation cabin
2	Propelling device	13	Suction device
				3	Shipborne cruise control module	14	Satellite positioning module
4	Feed cabin	15	Flow direction detection device
				5	Feeding device	16	Flow velocity detection device
6	Net cage identification module	17	Offshore communication module
				7	Underwater inspection device	18	Radar monitoring device
8	Prompting device	19	Visual monitoring device
				9	Fish shooting device	20	Shore-based monitoring system
10	Fish image analysis module	1201	Pumping pipeline
				11	Counting module

The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

It should be noted that, if directional indications (such as up, down, left, right, front, and rear … …) are included in the embodiments of the present application, the directional indications are merely used to explain the relative positional relationship, movement conditions, etc. between the components in a specific posture, and if the specific posture is changed, the directional indications are correspondingly changed.

In addition, if there is a description of "first", "second", etc. in the embodiments of the present application, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, if "and/or" and/or "are used throughout, the meaning includes three parallel schemes, for example," a and/or B "including a scheme, or B scheme, or a scheme where a and B are satisfied simultaneously. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present application.

The embodiment of the application provides a breeding operation ship, referring to fig. 1 and 2, which comprises a ship body 1, a propulsion device 2, a ship-borne cruise control module 3, at least two feed cabins 4, a feeding device 5 and a net cage identification module 6; wherein:

The propulsion device 2 is arranged on the ship body 1;

The shipborne cruise control module 3 is electrically connected with the propulsion device 2; the shipborne cruise control module 3 is used for acquiring the net cage coordinates sent by the shore-based monitoring system 20 and converting the net cage coordinates into a cruise path based on a path planning algorithm; the shipborne cruise control module 3 is used for converting angle parameters corresponding to the cruise path into power parameters and sending the power parameters to the propulsion device 2 so as to control the propulsion device 2 to drive the ship body 1 to travel to each cultivation net cage according to the cruise path;

At least two feed tanks 4 are arranged on the hull 1; at least two feed compartments 4 are respectively used for storing different kinds of fish feed;

the feeding end of the feeding device 5 is connected with at least two feed cabins 4;

The net cage identification module 6 is electrically connected with the feeding device 5; the net cage identification module 6 is used for identifying the type of the culture net cage based on an image identification algorithm and sending a corresponding feeding signal to the feeding device 5 so as to control the feeding device 5 to absorb the fish feed in the corresponding feed cabin 4 and send the fish feed to the corresponding culture net cage through the feeding end of the feeding device 5.

In this embodiment, the propulsion device 2 may include a plurality of propellers for omni-directional propulsion and/or lateral propulsion, specifically, for omni-directional propulsion, two full-rotation propellers capable of rotating by 360 ° may be installed in a propeller cabin at the stern of the hull 1, and driven by a marine propulsion motor, and the marine propulsion motor and the frequency converter adopt a frequency conversion control manner, so as to ensure that the propellers can work normally in both the highest draft state of the hull 1 and the lowest draft state of the hull 1; for side propulsion, two propellers can be arranged in a propeller cabin of the bow of the ship body 1 and driven by a ship propulsion motor, and a variable frequency control mode is adopted to ensure that the propellers can work normally in the highest draft state of the ship body 1 and in the lowest draft state of the ship body 1.

The cultivation cages are distributed in the target cultivation sea area, and the coordinates of each cultivation cage can be acquired in advance and stored in the shore-based monitoring system 20. After the shipborne cruise control module 3 obtains the cage coordinates of each cultivation cage sent by the shore-based monitoring system 20 through wireless communication and the like, the shipborne cruise control module 3 can obtain an optimal cruise path by adopting path planning algorithms such as an a-search algorithm, a D-Lite route searching algorithm and the like, wherein the optimal cruise path can be formed by connecting a plurality of cage coordinates in series; the cruising path contains an angle parameter, i.e. the angle of orientation of the hull 1 at each moment in time during travel along the cruising path; the shipborne cruise control module 3 can convert the angle parameter into a power parameter, wherein the power parameter is an operation parameter of the propulsion device 2 at each moment, specifically can be a start-stop condition, a gear, a propulsion speed and other parameters of each propeller, so that the straight running and steering of the ship body 1 are realized through the cooperation among the propellers, and the ship body 1 can independently run along a cruise path and sequentially reach each aquaculture net cage.

The feeding device 5 may include a power device and a corresponding pipe, where one end of the pipe is connected to the feed cabin 4 and forms a feeding end, and the other end of the pipe forms a feeding end and faces the direction of the aquaculture net cage, and the power device may mechanically push, negative pressure suck, etc. fish feed in the feed cabin 4 to the feeding end from the feeding end and deliver the fish feed to the aquaculture net cage from the feeding end, so as to realize feeding of fish in the aquaculture net cage.

In practical application, the types, the numbers and the growth periods of the fishes in different culture net boxes are different, so that the requirements of the fishes in different culture net boxes on feed feeding are different. Based on the above, in the embodiment, a plurality of feed cabins 4 are arranged, and different feed cabins 4 are used for storing different kinds of fish feeds so as to meet different requirements of fish in different cultivation net cages; meanwhile, each culture net cage is set to have a certain difference so as to be convenient for corresponding to different fish conditions. Specifically, the cage identification module 6 may acquire an image of the culture cage by means of the camera device, extract and identify the appearance or local features of the culture cage by means of an image identification algorithm, for example, may extract and identify the overall outline border, local structure border, code, color, etc. of the culture cage, compare with the preset cage image stored in the system, so as to determine the type of the culture cage and the fish condition in the culture cage, and send a corresponding feeding signal to the feeding device 5, so as to trigger the feeding device 5 to adjust the feeding parameters according to the preset program of the system, and control the feeding device 5 to suck the fish feed in the corresponding feed cabin 4 to feed the culture cage. Wherein, each pipeline of the feeding device 5 can be respectively provided with an electric control valve, and each electric control valve is respectively and electrically connected with the net cage identification module 6; different feeding signals sent by the net cage identification module 6 can correspondingly trigger the opening or closing of different electric control valves, so that the conduction or blocking of different pipelines is realized, the power device can absorb fish feed in the corresponding feed cabin 4 through the conducted pipelines and deliver the fish feed to the corresponding culture net cage, and the function of automatically and selectively feeding the fish feed according to the culture types of different culture net cages is realized.

It can be understood that the on-board cruise control module 3 and the net cage identification module 6 in the present embodiment may be corresponding functional modules in a controller having functions of data storage, read-write, program execution, signal input and output, etc.; the functional modules mentioned in the following embodiments are understood in this manner, and will not be described in detail.

Therefore, the cultivation operation ship provided by the embodiment obtains the net cage coordinates sent by the shore-based monitoring system 20 through the shipborne cruise control module 3, converts the net cage coordinates into a cruise path based on the path planning algorithm, converts the angle parameters corresponding to the cruise path into power parameters and sends the power parameters to the propulsion device 2, so that the propulsion device 2 can be controlled to drive the ship body 1 to travel to each cultivation net cage according to the cruise path; when the ship body 1 reaches any cultivation net cage, the type of the cultivation net cage is identified through the net cage identification module 6, and a corresponding feeding signal is sent to the feeding device 5, so that the feeding device 5 is controlled to absorb fish feed in the corresponding feed cabin 4 and feed the fish feed into the cultivation net cage through the feeding end of the feeding device 5, and the function of automatically and selectively feeding the fish feed according to the cultivation types of different cultivation net cages is realized; based on the setting, automatic feeding of the culture net cage is realized, feeding efficiency is improved, labor cost is saved, and the intelligent degree of feeding operation of the culture operation ship is improved.

Further, with reference to fig. 1 and 2, in one embodiment, the farming worker's vessel comprises an underwater inspection device 7 and a prompting device 8; the underwater patrol device 7 is arranged on the ship body 1, the underwater patrol device 7 pre-stores target netting images, and the underwater patrol device 7 is electrically connected with the prompting device 8;

The underwater inspection device 7 is used for moving around the aquaculture net cage along with the movement of the ship body 1; the underwater inspection device 7 is provided with a shooting end, and the shooting end is used for acquiring a current netting image of the aquaculture net cage; the underwater inspection device 7 is used for comparing the acquired current netting image with the target netting image to obtain a comparison result; and the underwater inspection device 7 is used for sending a first alarm signal to the prompting device 8 when the comparison result meets a first preset alarm condition so as to trigger the prompting device 8 to execute alarm operation.

In this embodiment, the underwater inspection device 7 may be a detection end connected to the hull 1 and extending to the underwater, where the detection end has a certain degree of freedom of movement with respect to the hull 1, and is capable of moving independently in the height direction while moving around the aquaculture net cage along with the movement of the hull 1, and adjusting the angle in real time, so as to ensure that the detection end is always opposite to the netting portion of the aquaculture net cage, and obtain current netting images at different depth positions of the aquaculture net cage. In addition, the underwater inspection device 7 can also directly adopt an underwater robot which is towed on the hull 1 through a cable and is communicated with corresponding functional modules on the hull 1 through the cable, the underwater robot has an autonomous cruising function, can autonomously encircle the aquaculture net cage to move, and can realize underwater positioning by means of the cooperation between the laser sensor and the netting part of the aquaculture net cage, so that the moving direction of the underwater robot can be corrected, and the accuracy of the moving path of the underwater robot is ensured; regarding the working principle and the specific implementation manner of the underwater robot, reference may be made to the relevant contents described in the patent application No. 202111298517.X entitled "positioning method, system and storage medium of underwater netting inspection robot", which are not described in detail herein.

In the process that the underwater inspection device 7 moves around the netting part of the aquaculture net cage, the shooting end continuously acquires the current netting image of the aquaculture net cage, and can continuously extract relevant features in the current netting image and compare the pre-stored target netting image through an image recognition algorithm, for example, the current netting image can be subjected to denoising treatment, and then the coverage area of each netting wire in the current netting image is recognized by adopting a Canny edge detection algorithm, so that a contour frame of the netting wire is formed, and the contour frame is compared with the contour frame of the netting wire in the target netting image by utilizing the contour frame, if the deviation amount of the contour frame and the contour frame is large, the condition that the netting wire breaks can be considered to exist, and the problem of netting wire breakage of the aquaculture net cage can be judged at the moment; when a first preset alarm condition is reached, a first alarm signal can be sent to the prompting device 8 to trigger the prompting device 8 to execute alarm operation, so as to inform an operator to process the aquaculture net cage and remove hidden danger in time; the first preset warning condition may be set such that the deviation amount of the current web image and the target web image exceeds a preset threshold. The alarm operation of the prompting device 8 can be realized by an audible and visual alarm (such as an alarm lamp, a buzzer and the like), and prompting information can be output and presented to an operator in the forms of voice, characters, images and the like so as to inform of abnormality; the following explanation will be referred to for the prompting device 8 and the alarm operation, and will not be repeated.

Further, with reference to fig. 1 and 2, in another embodiment, the farming worker's vessel comprises an underwater inspection device 7 and a prompting device 8; the underwater inspection device 7 is arranged on the ship body 1, and the underwater inspection device 7 is electrically connected with the prompting device 8;

The underwater inspection device 7 is used for moving around the aquaculture net cage along with the movement of the ship body 1; the underwater inspection device 7 is provided with a laser emission end and a shooting end, wherein the laser emission end is used for emitting a linear laser beam to the aquaculture net cage, and the shooting end is used for acquiring a discrete light spot image formed by the linear laser beam on a net part of the aquaculture net cage; the underwater inspection device 7 is configured to send a second alarm signal to the prompting device 8 when the light spot gap existing in the discrete light spot image meets a second preset alarm condition, so as to trigger the prompting device 8 to execute an alarm operation.

The present embodiment provides another scheme for determining the damage of the netting, and specifically, the specific setting manner of the underwater inspection device 7 can refer to the previous embodiment, which is not described herein again. The netting part of the aquaculture net cage is a flexible piece which is woven by net wires and provided with a plurality of meshes, the line laser beam can be regarded as a continuous line segment which consists of a plurality of laser points, no matter how the netting part of the aquaculture net cage fluctuates in water, the continuous line laser beam always has laser points which can be beaten on the net wires of the netting part and collected by a shooting end; after capturing a plurality of laser points striking the netting part, the shooting end will appear on the imaging surface in the form of discrete light spot images, wherein the discrete light spots in the discrete light spot images are the collection of the laser points striking the netting wire on the images.

After the discrete light spot images are obtained, the discrete light spot images can be extracted and identified based on an image identification algorithm; because the sizes of all the mesh holes of the mesh part are basically consistent, under the condition that the mesh part is not broken and damaged, the discrete light spots in the discrete light spot images are approximately in a continuous straight line shape, if the light spots in the discrete light spots are obviously oversized in interval, namely when light spot gaps exist in the discrete light spot images, the problem that the mesh wire breaks at the light spot gaps can be judged, so that the wire laser beam cannot strike the part to form light spots; when the spot gap meets a second preset alarm condition, for example, when the size and/or the number of the spot gap reach a preset threshold condition, it can be considered that the breaking condition of the net wire reaches a degree sufficient to influence the overall performance of the aquaculture net cage, and the fish may escape, at this time, a second alarm signal needs to be sent to the prompting device 8 to trigger the prompting device 8 to execute an alarm operation, so as to inform an operator to process the aquaculture net cage and timely eliminate hidden danger.

Optionally, referring to fig. 1 and 2, the farming worker ship includes a fish photographing device 9, a fish image analysis module 10, a counting module 11, and a prompting device 8; the fish image analysis module 10 is provided with a GMM Gaussian mixture model and a YOLO neural network model, the fish image analysis module 10 is electrically connected with the fish shooting device 9 and the counting module 11, and the counting module 11 is electrically connected with the prompting device 8;

The fish shooting device 9 is used for acquiring a fish key frame image set of a water surface part of the cultivation net cage, and respectively inputting the acquired fish key frame image set into the GMM Gaussian mixture model and the YOLO neural network model for moving target detection operation so as to respectively obtain a GMM fish target frame and a YOLO fish target frame;

The fish image analysis module 10 is used for selecting a GMM fish target frame or a YOLO fish target frame as a dead fish target frame; the counting module 11 is used for counting the number of the frames of the dead fish targets, and the counting module 11 is used for sending a third alarm signal to the prompting device 8 when the number of the frames of the dead fish targets reaches a preset number threshold value so as to trigger the prompting device 8 to execute alarm operation.

When the death condition of fish occurs in the culture net cage, the dead fish floats on the water surface part; acquiring video images of the water surface part through the fish shooting device 9 to form a fish key frame image set, wherein the fish key frame image set is extracted from the video shot by the fish shooting device 9 through a key frame technology and subjected to denoising treatment, and comprises a plurality of frames of images capable of representing the distribution condition of fish appearing on the water surface; the fish key frame image set is input into the fish image analysis module 10 for image analysis, the dead fish target frames floating on the water surface due to death can be extracted from the fish key frame image set, the number of the dead fish target frames is counted through the counting module 11, the current dead fish number in the aquaculture net cage can be judged, and a third warning signal is sent to the prompting device 8 when the number reaches a preset number threshold value, so that the prompting device 8 is triggered to execute warning operation, operators are informed of cleaning the dead fish, and the situation that the number of the dead fish is too large to pollute the aquaculture water body and influence the normal activities of the living fish is avoided.

Regarding the process of the fish image analysis module 10 performing image analysis on the fish keyframe image set, specifically, the YOLO neural network model based on deep learning may label and train the image texture features such as the color, the shape, the outline, etc. of the fish target in the video sequence image, so as to form a feature extraction convolutional network, thereby realizing detection of the fish target and outputting the distribution condition of the fish target. However, the YOLO neural network model usually sets a detection threshold to avoid false detection of a stationary object that forms a fish target, but if the threshold is set too high, a problem of missed detection occurs, in other words, a detection algorithm of a single YOLO neural network model cannot well mine fish target motion information for fish transient and fish shielding situations to avoid missed detection, and cannot adapt to target feature extraction for changes of fish orientations to avoid false detection. Therefore, the GMM Gaussian mixture model (Gaussian Mixture Model) is correspondingly introduced, and can fully mine the characteristics of the fish target motion information, and the GMM Gaussian mixture model is used for replacing the detection threshold value set by the YOLO neural network model necessarily, so that false detection and omission detection are effectively avoided. In the specific implementation process, after a fish key frame image set is input into a YOLO neural network model, each frame image is divided into S multiplied by S grids, a YOLO detection algorithm detects a fish target frame aiming at each grid, RGB frame spots in the grids are extracted and transferred to a fish YOLO classifier, and therefore the fish target frame in each grid is detected; the fish object frames refer to fish outline frames identified by a corresponding detection algorithm, the quantity, distribution and other information of fish objects in any grid can be determined according to the identified fish object frames, all the fish object frames in one frame of picture can be obtained after combining the fish object frames of all the grids, and then the fish object distribution condition of the water surface part in a certain time period can be obtained after combining the fish object frames of each frame of picture. It can be understood that the YOLO fish target frame may refer to a single fish target frame or a set of fish target frames in any grid, all grids of any frame of picture, and multiple frames of pictures detected by the YOLO neural network model, and the definition of the single fish target frame or the set of fish target frames may be flexibly adjusted according to an application scene, which is not limited herein. Similarly, after a fish key frame image set is input into a GMM Gaussian mixture model, each frame image is divided into S multiplied by S grids, a GMM detection algorithm detects a fish target frame aiming at each grid, RGB frame spots in the grids are extracted and transferred to a fish ResNet-50 classifier, and therefore the fish target frame in each grid is detected; according to the identified fish object frames, the quantity, distribution condition and other information of fish objects in any grid can be determined, all the fish object frames in the frame can be obtained after combining the fish object frames of all the grids in one frame of picture, and then the fish object distribution condition of the water surface part in a certain time period can be obtained after integrating the fish object frames of each frame of picture. It can be understood that the GMM fish target frame may refer to a single fish target frame or a set of fish target frames in any grid detected by the GMM gaussian mixture model, in all grids of any frame of picture, in multiple frames of picture, and the definition of the single fish target frame or the set of fish target frames may be flexibly adjusted according to the application scenario, which is not limited herein. For one of the fish target rims, it may be at least the following: 1. is detected by the GMM gaussian mixture model but not by the YOLO neural network model; 2. detected by the YOLO neural network model but not by the GMM gaussian mixture model; and is detected by the GMM Gaussian mixture model and the YOLO neural network model. Aiming at the various conditions, the judgment and selection can be carried out through preset judgment logic; specifically, the judging and selecting operations can be automatically executed through a judging device, and aiming at the same fish target frame, the judging device is used for analyzing all detection parameters of the GMM fish target frame and the YOLO fish target frame, and comparing all detection parameters with a preset threshold value to finally determine whether the accuracy of the detected GMM fish target frame and/or YOLO fish target frame meets the preset requirement or not, and selecting the GMM fish target frame or YOLO fish target frame with the accuracy meeting the preset requirement as a dead fish target frame to output; the same operation is performed on other fish target frames, so that a dead fish target frame which is more accurate than a single data source can be obtained.

Alternatively, referring to fig. 1 and 2, the hull 1 is provided with a farming compartment 12, the farming worker's vessel comprising suction means 13; one end of the suction device 13 is communicated with the culture cabin 12, and the suction device 13 is used for sucking the living fish in the culture net cage into the culture cabin 12.

Specifically, the suction device 13 may include a power device and a corresponding pipe, one end of the pipe is communicated with the culture cabin 12, the other end of the pipe is used for being in butt joint with the culture net cage, the power device can transfer the live fish in the culture net cage into the culture cabin 12 through mechanical pushing, negative pressure suction and other modes, for example, the live fish in the culture net cage can be sucked into the culture cabin 12 through a fish suction pump, so that the fishing of the live fish is realized, the manual fishing operation is omitted, the labor cost is saved, and the fishing efficiency is improved.

Preferably, various auxiliary devices for living the living fish in a short time, such as water quality monitoring devices, oxygen cones and the like, are arranged in the culture cabin 12 so as to avoid death of the living fish in the culture cabin 12.

Optionally, referring to fig. 1 and 2, at least two pumping pipes 1201 having a height difference are provided in the culture compartment 12; the pumping duct 1201 is used to supply or discharge water to the body of water in the habitat 12 to create a simulated flow field within the habitat 12.

Specifically, the flow-through ports of the pumping and draining pipelines 1201 may be disposed on the bottom wall and the side wall of the culture cabin 12, and each pumping and draining pipeline 1201 may supply external water into the culture cabin 12 and pump water in the culture cabin 12 to the outside, so as to form a circulating water system; wherein, parameters such as the opening and closing state, flow rate and the like of each pumping and discharging pipeline 1201 can be controlled by a valve arranged in the pumping and discharging pipeline 1201; by setting different pumping and draining pipelines 1201 to different opening and closing states, a plurality of different water changing combination modes can be formed, so that the flow state of the flow field in the culture cabin 12 can be adjusted according to different conditions, the real culture environment of deep and open sea can be simulated with higher reduction degree, and further, a flow field environment with better culture fish adaptability can be constructed in the culture cabin 12, and the growth activity of living fish in the culture cabin 12 is facilitated.

Optionally, referring to fig. 1 and 2, the farming worker vessel comprises a satellite positioning module 14 and a prompting device 8; the satellite positioning module 14 is electrically connected with the shipborne cruise control module 3, and the shipborne cruise control module 3 is electrically connected with the prompting device 8;

The satellite positioning module 14 is used for acquiring current position information of the corresponding cultivation net boxes when the ship body 1 moves to each cultivation net box and sending the current position information to the shipboard cruise control module 3; the on-board cruise control module 3 is used for correcting the net cage coordinates according to the current position information, and the on-board cruise control module 3 is used for sending a fourth alarm signal to the prompting device 8 when the difference between the current position information and the net cage coordinates reaches a preset distance threshold value so as to trigger the prompting device 8 to execute alarm operation.

Specifically, by arranging the satellite positioning module 14 (Global Positioning System, GPS) on the hull 1, when the hull 1 advances to any cultivation net cage, the real current position information of the cultivation net cage can be obtained, so that the current position information can be utilized to correct the historical net cage coordinate information obtained by the shipborne cruise control module 3, the precision of the cruise path can be continuously improved, the hull 1 can advance to a preset position more accurately later, and corresponding operations such as feed feeding, net inspection, live fish fishing and the like can be better performed.

When the difference between the current position information of the culture net cage and the coordinates of the net cage is larger and reaches the preset distance threshold, the anchoring system of the culture net cage can be considered to have a problem, so that the culture net cage has a larger degree of drift, and a fourth alarm signal is required to be sent to the prompting device 8 at the moment so as to trigger the prompting device 8 to execute alarm operation and inform operators of timely intervention treatment.

Alternatively, referring to fig. 1 and 2, the farming worker's vessel includes a flow direction detection device 15 and a flow rate detection device 16; the flow direction detection device 15 and the flow speed detection device 16 are electrically connected with the shipboard cruise control module 3;

the flow direction detection device 15 is used for acquiring flow direction information in a preset area with the culture net cage as a center and sending the flow direction information to the shipborne cruise control module 3, and the flow speed detection device 16 is used for acquiring flow speed information in the preset area with the culture net cage as a center and sending the flow speed information to the shipborne cruise control module 3;

The on-board cruise control module 3 is used for taking a path of which the water flow direction is directed to the aquaculture net cage as a feeding path, and the on-board cruise control module 3 is used for taking a region of which the flow speed on the feeding path reaches a preset flow speed threshold value as a feeding region; the on-board cruise control module 3 is used for converting the position parameters corresponding to the feeding area into power parameters and sending the power parameters to the propulsion device 2 so as to control the propulsion device 2 to drive the ship body 1 to travel to the feeding area and perform feeding operation.

Specifically, when the ship body 1 moves to any cultivation net box, the flow direction near the cultivation net box can be detected by means of the flow direction detection device 15, so that the upstream and downstream of the cultivation water area at the cultivation net box can be determined, and the downstream water flow route passing through the cultivation net box is used as a feeding path, so that the ship-borne cruise control module 3 can control the propulsion device 2 to drive the ship body 1 to move to the feeding path, and the ship body 1 is enabled to feed in the upstream position, and at the moment, the fish feed fed by the feeding device 5 can be accurately conveyed into the cultivation net box along the feeding path under the action of water flow, thereby improving the accuracy of feed feeding, and avoiding waste caused by excessive water flow flushing of the feed.

In a specific implementation process, a plurality of feeding paths may exist, at this time, the feeding paths are further screened by the flow velocity detection device 16, and a region with a faster flow velocity in the feeding paths is selected as a feeding region, so that the ship-borne cruise control module 3 can control the propulsion device 2 to drive the ship body 1 to move to the feeding region for feeding operation, fish feed can quickly reach the aquaculture net cage at a higher flow velocity, feeding efficiency is improved, and influence of tributaries on feed feeding is reduced to the greatest extent.

Preferably, after determining the feeding area, the on-board cruise control module 3 may send a corresponding adjusting signal to the feeding device 5, so as to control the feeding device 5 to adjust parameters such as a feeding angle, a height, a force, a feeding amount of the feeding end according to a preset program, so that a rated amount of fish feed can be accurately fed into the cultivation net cage.

Optionally, referring to fig. 1 and 2, the farming worker vessel comprises an offshore communication module 17; the offshore communication module 17 is electrically connected with the propulsion device 2 and the feeding device 5;

the offshore communication module 17 is used for receiving real-time remote control instructions sent by the shore-based monitoring system 20; the marine communication module 17 is used for converting the real-time remote control instruction into a power parameter and sending the power parameter to the propulsion device 2 so as to control the propulsion device 2 to drive the ship body 1 to travel; and the offshore communication module 17 is used for converting the real-time remote control instruction into a driving parameter and sending the driving parameter to the feeding device 5 so as to control the feeding device 5 to execute feeding operation.

In the present embodiment, shore-based remote control can be realized by providing the offshore communication module 17; specifically, the shore-based monitoring system 20 may use a large-bandwidth high-speed marine 5G communication technology, and transmit operation information and control commands to the aquaculture worker ship in real time through the marine communication module 17, so as to control the aquaculture worker ship to travel according to the instructions and perform operations such as feeding, netting inspection, dead fish detection, and the like.

Preferably, a switch can be arranged between the shipboard cruise control module 3 and the offshore communication module 17 to freely switch between two operation modes of autonomous cruise feeding and shore-based remote control sailing feeding according to actual needs.

Optionally, referring to fig. 1 and 2, the hull 1 is provided with radar monitoring means 18 and visual monitoring means 19, the farming worker vessel comprising prompting means 8; the radar monitoring device 18 and the visual monitoring device 19 are electrically connected with the prompting device 8; the radar monitoring device 18 and the vision monitoring device 19 are used for sending a fifth alarm signal to the prompting device 8 when an unknown target object is detected, so as to trigger the prompting device 8 to execute an alarm operation.

Specifically, the radar monitoring device 18 and the visual monitoring device 19 may be installed at a plurality of positions such as a top, a port, a starboard and the like of the hull 1, so as to realize real-time monitoring and patrol of the culture sea area around the hull 1, and when an unknown object such as a suspicious person, a ship and the like is found to be close to the hull 1 or the culture net cage, the warning device 8 may be triggered to execute the warning operation by sending a fifth warning signal, so as to inform personnel on the shore to intervene in the treatment, thereby eliminating the potential safety hazard in time.

It should be noted that, other contents of the cultivation operation ship disclosed in the present application can be referred to the prior art, and will not be described herein.

The foregoing is only an optional embodiment of the present application, and is not limited to the scope of the patent application, and all equivalent structural changes made by the description of the present application and the accompanying drawings or direct/indirect application in other related technical fields are included in the scope of the patent application.

Claims

1. A farming worker vessel, the farming worker vessel comprising:

A hull;

The propulsion device is arranged on the ship body;

2. The farming worker vessel according to claim 1, wherein the farming worker vessel comprises an underwater inspection device and a prompting device; the underwater patrol device is arranged on the ship body, a target netting image is prestored in the underwater patrol device, and the underwater patrol device is electrically connected with the prompting device;

3. The farming worker vessel according to claim 1, wherein the farming worker vessel comprises an underwater inspection device and a prompting device; the underwater patrol device is arranged on the ship body and is electrically connected with the prompting device;

4. The farming worker vessel according to claim 1, wherein the farming worker vessel comprises a fish photographing device, a fish image analysis module, a counting module, and a prompting device; the fish image analysis module is provided with a GMM Gaussian mixture model and a YOLO neural network model, and is electrically connected with the fish shooting device and the counting module, and the counting module is electrically connected with the prompting device;

5. A farming worker vessel according to claim 1, wherein the hull is provided with a farming tank, the farming worker vessel comprising suction means; one end of the suction device is communicated with the culture cabin, and the suction device is used for sucking the living fish in the culture net cage into the culture cabin.

6. The farming worker ship according to claim 5, wherein at least two suction pipes having a height difference are provided in the farming tanks; the pumping and draining pipeline is used for carrying out water supply operation or water draining operation on the water body in the culture cabin so as to form a simulated flow field in the culture cabin.

7. The farming worker vessel according to claim 1, wherein the farming worker vessel comprises a satellite positioning module and a prompting device; the satellite positioning module is electrically connected with the shipborne cruise control module, and the shipborne cruise control module is electrically connected with the prompting device;

8. The farming worker vessel according to claim 1, wherein the farming worker vessel comprises a flow direction detection device and a flow rate detection device; the flow direction detection device and the flow speed detection device are electrically connected with the shipborne cruise control module;

9. The farming worker vessel according to claim 1, wherein the farming worker vessel comprises an offshore communication module; the offshore communication module is electrically connected with the propulsion device and the feeding device;

10. A farming worker vessel according to claim 1, wherein the hull is provided with radar monitoring means and visual monitoring means, the farming worker vessel comprising prompting means; the radar monitoring device and the visual monitoring device are electrically connected with the prompting device; the radar monitoring device and the visual monitoring device are used for sending a fifth alarm signal to the prompting device when an unknown target object is detected, so as to trigger the prompting device to execute alarm operation.