Control method, device, equipment and storage medium for unmanned feeding equipment

Abstract

The embodiments of the present application relate to a control method, device, equipment and storage medium for unmanned feeding equipment, which are applied to unmanned feeding equipment. The unmanned feeding equipment includes a feeding device and a first camera. The first camera is located on the feeding device. The above method includes: acquiring multiple frames of first images obtained by collecting images of the net cage through the first camera; extracting features of the net cage contained in each frame of the first image to obtain the net cage features corresponding to each frame of the first image; calculating the first relative position relationship between the net cage center corresponding to the net cage and the feeding device according to the net cage features corresponding to the multiple frames of the first image; controlling the movement of the feeding device according to the first relative position relationship so that the feeding device is aligned with the center of the net cage to feed. The control method, device, equipment and storage medium of the above unmanned feeding equipment improves the accuracy of the unmanned feeding equipment in feeding and improves the breeding efficiency.

Description

Technical Field

[0001] The present application relates to the field of aquaculture technology, and in particular to a control method, device, equipment and storage medium for unmanned feeding equipment. Background Art

[0002] The ocean is a strategic place for high-quality development. The development of the marine economy is an important pillar of my country's economic development. Using deep-sea cages for fish farming is a major strategy for developing the marine economy. With the rapid expansion of the scale of the marine aquaculture industry, feeding in aquaculture has also become a key development direction. At present, feeding is usually done by manual experience, which has low feeding efficiency, high safety risks at sea, and inaccurate feeding of fish. Summary of the invention

[0003] The embodiments of the present application disclose a control method, device, equipment and storage medium for unmanned feeding equipment, which improve the accuracy of bait delivery by the unmanned feeding equipment and improve the breeding efficiency.

[0004] In a first aspect, an embodiment of the present application discloses a control method for an unmanned feeding device, which is applied to the unmanned feeding device, wherein the unmanned feeding device includes a feeding device and a first camera, wherein the first camera is located on the feeding device, and the method includes:

[0005] Acquire multiple frames of first images obtained by capturing images of the cage through the first camera;

[0006] Extracting features of the mesh boxes contained in each frame of the first image to obtain mesh box features corresponding to each frame of the first image;

[0007] Calculating a first relative position relationship between a net cage center corresponding to the net cage and the feeding device according to the net cage features respectively corresponding to the multiple frames of first images;

[0008] The feeding device is controlled to move according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait.

[0009] As an optional implementation, in the first aspect of the embodiment of the present application, the unmanned feeding device further includes a second camera, and the field of view of the second camera is greater than the field of view of the first camera; before acquiring the first frame of the first image obtained by capturing the image of the cage through the first camera, the method further includes:

[0010] Acquire multiple frames of second images obtained by collecting images of the cage through the second camera;

[0011] Performing target recognition on the net cages included in the plurality of frames of second images, and determining a second relative position relationship between the net cages and the unmanned feeding device according to the target recognition result;

[0012] According to the second relative position relationship, the unmanned feeding equipment and / or the feeding device are controlled to move so that the feeding device is aligned with the cage.

[0013] As an optional implementation manner, in the first aspect of the embodiment of the present application, after controlling the feeding device to move according to the first relative position relationship, the method further includes:

[0014] Calculating the relative distance between the discharge port of the feeding device after the movement and the center of the net cage;

[0015] Determine the bait launching force corresponding to the feeding device according to the relative distance and the discharge amount;

[0016] The feeding device is controlled to launch the bait according to the bait launching force so that the bait is dropped into the center of the net cage.

[0017] As an optional implementation, in the first aspect of the embodiment of the present application, the unmanned feeding device further includes a wind speed and direction sensor, and the wind speed and direction sensor is located on the feeding device;

[0018] Determining the bait launching force corresponding to the feeding device according to the relative distance and the discharge amount includes:

[0019] Obtaining the sea surface wind speed and sea surface wind direction collected by the wind speed and wind direction sensor;

[0020] The bait launching force corresponding to the feeding device is calculated according to the sea surface wind speed, the sea surface wind direction, the relative distance and the discharge amount.

[0021] As an optional implementation, in the first aspect of the embodiment of the present application, the cage includes a plurality of floating body components; the feature extraction of the cage contained in each frame of the first image to obtain the cage features corresponding to each frame of the first image includes:

[0022] Identify the mesh cage contained in each frame of the first image, and determine the target image area corresponding to the mesh cage in each frame of the first image;

[0023] The floating body components contained in each target image area are identified and features are extracted to obtain the floating body component features corresponding to each target image area.

[0024] As an optional implementation, in the first aspect of the embodiment of the present application, the identifying and feature extraction of the floating body components contained in each target image area to obtain the floating body component features corresponding to each target image area includes:

[0025] Identify and extract features of the floating body components contained in the first target image area to obtain a plurality of target detection frames contained in the first target image area; the first target image area is a target image area in any frame of the first image;

[0026] Determine the floating component features corresponding to the first target image area according to the feature information corresponding to the multiple target detection frames; each of the target detection frames corresponds to at least one floating component in the cage;

[0027] The calculating, based on the cage features respectively corresponding to the multiple frames of first images, a first relative position relationship between the cage center corresponding to the cage and the feeding device comprises:

[0028] Determine the image coordinates of the cage center corresponding to the cage in each frame of the first image according to the floating body component features corresponding to the target image area in each frame of the first image;

[0029] The first relative position relationship between the center of the cage and the feeding device is calculated according to the image coordinates corresponding to the center of the cage in each frame of the first image.

[0030] As an optional implementation, in the first aspect of the embodiment of the present application, controlling the feeding device to move according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait includes:

[0031] According to the first relative position relationship, determining the moving direction and moving angle corresponding to the feeding device;

[0032] According to the moving direction and the moving angle, the feeding device is controlled to move so that the feeding device continuously aims at the center of the cage to feed the bait.

[0033] In a second aspect, an embodiment of the present application discloses a control device for an unmanned feeding device, which is applied to an unmanned feeding device, wherein the unmanned feeding device includes a feeding device and a first camera, wherein the first camera is located on the feeding device, and the control device for the unmanned feeding device includes:

[0034] An image acquisition module, used to acquire multiple frames of first images obtained by collecting images of the cage through the first camera;

[0035] A feature extraction module, used to extract features of the cage contained in each frame of the first image, to obtain features of the cage corresponding to each frame of the first image;

[0036] A center calculation module, used for calculating a first relative position relationship between a net box center corresponding to the net box and the feeding device according to the net box features respectively corresponding to the multiple frames of first images;

[0037] The rotating feeding module is used to control the movement of the feeding device according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait.

[0038] In a third aspect, an embodiment of the present application discloses an unmanned feeding device, including a feeding device, a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor implements the method described in any of the above embodiments.

[0039] In a fourth aspect, an embodiment of the present application discloses a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.

[0040] The control method, apparatus, device and storage medium of the unmanned feeding equipment disclosed in the embodiments of the present application obtain multiple frames of first images by acquiring images of the net cage through a first camera; extracting features of the net cage contained in each frame of the first image to obtain net cage features corresponding to each frame of the first image; calculating the first relative position relationship between the net cage center corresponding to the net cage and a feeding device based on the net cage features corresponding to the multiple frames of the first images; and controlling the movement of the feeding device based on the first relative position relationship so that the feeding device is aligned with the center of the net cage to feed bait. In an embodiment of the present application, the unmanned feeding equipment can accurately obtain the first relative position relationship between the center of the cage and the feeding device according to the cage features corresponding to the multiple frames of first images collected by the camera, and can control the unmanned feeding equipment to move the feeding device so that the feeding device can accurately deliver bait to the center of the cage, thereby achieving precise control of bait delivery and allowing the bait to fall into the center of the cage, thereby achieving precise feeding operation of the unmanned feeding equipment, improving the accuracy of bait delivery by the unmanned feeding equipment, and the bait delivered to the center of the cage can be evenly distributed in the cage, reducing bait waste and the time it takes for fish to obtain bait, thereby improving breeding efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1a This is an application scenario diagram of a control method for an unmanned feeding device in one embodiment;

[0043] Figure 1b In one embodiment, a frame of a first image captured by a camera is defined;

[0044] Figure 2 It is a flow chart of a control method of an unmanned feeding device in one embodiment;

[0045] Figure 3 A flow chart of a control method for an unmanned feeding device in another embodiment;

[0046] Figure 4 A flow chart of a control method for an unmanned feeding device in another embodiment;

[0047] Figure 5 A flow chart of a control method for an unmanned feeding device in another embodiment;

[0048] Figure 6 A block diagram of a control device for an unmanned feeding device in one embodiment;

[0049] Figure 7 It is a structural block diagram of an unmanned feeding device in one embodiment. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0051] It should be noted that the terms "including" and "having" and any variations thereof in the embodiments of the present application and the accompanying drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0052] It is understood that the terms "first", "second", etc. used in this application can be used to describe various elements in this article, but these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, the first relative position relationship can be referred to as the second relative position relationship, and similarly, the second relative position relationship can be referred to as the first relative position relationship. Both the first relative position relationship and the second relative position relationship are relative position relationships, but they can be the same relative position relationship or different relative position relationships.

[0053] At present, in large-scale deep-sea aquaculture, there are a large number of deep-sea cages, and in order to ensure the growth of fish in the cages, fishermen need to manually drive large boats to the approximate location of the cages, unload a few bags of feed from the large boats and load them onto small boats, and feed the fish in the cages next to the deep-sea cages. The aquaculture efficiency is not high, but the labor cost is higher. Taking the operation volume of about 60 cage aquaculture waters as an example, feeding is required at least twice a day. The average amount of feed required for a base is 2,500 tons per day, and an average of 167 aquaculture workers are required for each base. There are great safety risks in going out to sea and feeding is not accurate. In most cases, people will feed too much or too little based on manual experience. Therefore, the bait and feeding of bait account for 60 to 70% of the current deep-sea aquaculture costs, affecting aquaculture efficiency and economic benefits.

[0054] The embodiments of the present application disclose a control method, device, equipment and storage medium for unmanned feeding equipment, which improve the accuracy of bait delivery by the unmanned feeding equipment and improve the breeding efficiency.

[0055] The control method of the unmanned feeding device provided in this application can be applied to Figure 1a In the application scenario shown in FIG.

[0056] The unmanned feeding device 110 may include a feeding device 120 and a first camera 130. Optionally, the first camera 130 may be arranged on the feeding device 120. The unmanned feeding device 110 may include but is not limited to electronic devices such as drones and unmanned ships equipped with intelligent unmanned systems. Optionally, the feeding device 120 may also have a discharge port so that the feeding device 120 can emit bait through the discharge port. Optionally, the unmanned feeding device 110 and / or the feeding device 120 may also have a storage bin for storing various baits. Optionally, the first camera 130 may include but is not limited to imaging devices such as infrared thermal imagers, digital cameras, digital imaging equipment, and X-ray computer tomography equipment; due to the different types of the first camera 130, the images or videos captured are also different, which may be RGB (red, green, and blue primary colors) color domain, CMY / CMYK (Cyan, Magenta, Yellow, Black) color domain, or grayscale imaging.

[0057] The unmanned feeding device 110 can feed fish and other organisms raised in the net cage 140. The net cage 140 may include a box body and a floating body assembly, the box body may be set under the water surface, the floating body assembly may float on the water surface, and the net cage 140 may be placed in a certain water area to breed fish. Optionally, the floating body assembly of the net cage 140 floating on the water surface may form various shapes such as round, square, polygonal, etc.

[0058] In some embodiments, Figure 1a The application scenario shown can be in lakes, rivers, reservoirs, oceans and even deep sea areas with certain water flow, fresh water quality and high dissolved oxygen content. Specifically, farmers can set up multiple cages 140 in the designated breeding area of ​​the deep sea, and place fish fry in the cages 140 for breeding. During the breeding process, unmanned feeding equipment 110 is used to go to the areas where the cages 140 are located in the deep sea, and the feeding device 120 is used to feed the fish farmed in the cages 140.

[0059] like Figure 2 As shown, in one embodiment, a control method for an unmanned feeding device is provided, which can be applied to the above-mentioned unmanned feeding device. The method may include the following steps:

[0060] Step 210, acquiring multiple frames of first images by capturing images of the cage through a first camera.

[0061] When the farmer needs to feed the fish in the cage, he can send out an unmanned feeding device, which is loaded with the bait needed by the fish in the cage, and control the unmanned feeding device to go to the location of the cage. The unmanned feeding device can analyze the first image captured by the first camera to accurately locate the location where feeding is required, so as to feed the fish in the cage. The cage may include multiple floating body components; the floating body components enable the cage to float on the water surface.

[0062] In some embodiments, the unmanned feeding equipment may include a feeding device and a first camera, and the first camera is located on the feeding device; during the operation of the unmanned feeding equipment (in the process of moving towards the location of the cage or in the process of searching for the cage at a specified location), the first camera can continuously capture images. If the unmanned feeding equipment does not detect the cage in the image captured by the first camera, the frame of image is discarded. If the unmanned feeding equipment detects the cage in the image captured by the first camera, the frame of image is used as the first image.

[0063] In some embodiments, the unmanned feeding device can classify the images collected by the first camera through a classification model, and classify images containing cages into one category and images without cages into another category. The unmanned feeding device can use a large number of sample images and accurate classification results corresponding to the sample images to train the pre-trained classification model, and the accurate classification results indicate whether the sample images actually contain cages; multiple frames of sample images are input into the pre-trained classification model to obtain sample classification results corresponding to each frame of sample images; according to the sample classification results and accurate classification results corresponding to each frame of sample images, the accuracy rate corresponding to the pre-trained classification model is determined; the parameters of the pre-trained classification model are optimized until the accuracy rate of the classification model is greater than a preset threshold.

[0064] In some embodiments, even if the unmanned feeding device knows the geographical location corresponding to each cage in advance, such as longitude and latitude, etc., the positioning of the deep sea area is prone to deviation, or the movement of ocean currents causes the position of the cage to shift, etc. In these cases, even if the unmanned feeding device reaches the geographical location corresponding to the cage, it is usually impossible to immediately determine the position of the cage, resulting in the first camera being unable to directly capture images of the cage. Therefore, the unmanned feeding device can be provided with detection devices such as radar detection and / or ultrasonic detection, so that the unmanned feeding device can quickly obtain the actual position of the cage through the detection device when it reaches the geographical location corresponding to each cage, thereby controlling the first camera to capture the first image containing the cage.

[0065] In some embodiments, whether fish farming is carried out in rivers, lakes or deep seas, it is easily affected by factors such as water currents, ocean currents, and air currents. For example, in deep-sea farming mode, a strong sea breeze usually blows on the sea surface, and the sea breeze will affect the position of the bait falling on the net cage when the unmanned feeding device launches the bait; or, the strong ocean current changes caused by the sea breeze will also produce a force on the position of the net cage and the unmanned feeding device, thereby affecting the feeding, etc. Therefore, the unmanned feeding device may also include a wind speed and direction sensor and / or a water flow detection sensor. The wind speed and direction sensor may be located on the feeding device to detect the wind speed and direction on the water surface; the water flow detection sensor may be located below the horizontal plane to detect the water flow speed and direction under the water surface. While acquiring the first image, the unmanned feeding device may also obtain information such as the current wind speed and direction on the water surface, the water flow speed and direction under the water surface, etc. through the wind speed and direction sensor and / or the water flow sensor. The unmanned feeding device can place bait into the net cage according to each frame of the first image, the wind speed and direction on the water surface, and the water flow speed and direction under the water surface. When the unmanned feeding device obtains a frame of the first image, it can continue to capture images of the net cage along the wind direction on the water surface and / or the water flow direction under the water surface; the unmanned feeding device can determine the corresponding first shooting interval according to the wind speed on the water surface and / or the water flow speed under the water surface; control the first camera to capture the first image at the first shooting interval, and the shooting interval refers to the time interval between the first camera capturing two adjacent frames of the first image, wherein the greater the wind speed on the water surface and / or the water flow speed under the water surface, the shorter the corresponding shooting interval. By comprehensively considering the influence of various factors such as the environment, the unmanned feeding device can find a more accurate position, thereby performing more accurate feeding.

[0066] Step 220 , extracting features of the grid boxes contained in each frame of the first image to obtain grid box features corresponding to each frame of the first image.

[0067] In order to be able to feed accurately to the center of the cage, the unmanned feeding equipment needs to determine the center of the cage. The center of the cage refers to the center point of the cage on the water surface. Therefore, the unmanned feeding equipment can deduce the position of the center of the cage in the time coordinate system, or the relative position relationship between the center of the cage and the feeding device, by analyzing the cage features corresponding to the cages contained in the multiple frames of the first image. The cage features in the first image refer to specific features in the first image that can characterize the situation of the cage. Optionally, the cage features may include but are not limited to one or more of the shape of the cage, the image position of the cage in the first image, the size of the image area occupied by the cage in the first image, the shape of the floating assembly contained in the cage, the size of the image area occupied by the floating assembly in the first image, etc.

[0068] In some embodiments, the feature extraction algorithms that can be used by the unmanned feeding equipment include but are not limited to HOG (Histogram of Oriented Gradient), SIFT (Scale Invariant Feature Transform), SURF (Speeded Up Robust Features), ORB (Oriented FAST and Rotated BRIEF), etc.

[0069] In one embodiment, the unmanned feeding device can use a cage feature extraction model to extract features of the cages contained in each frame of the first image to obtain cage features corresponding to each frame of the first image. The cage feature extraction model refers to a model that can identify cage features contained in an input image; the cage feature extraction model can be constructed based on a deep learning algorithm, and through training with a large number of sample first images with cage feature labels, it can accurately identify and locate cage features in the first image.

[0070] Step 230, calculating the first relative position relationship between the mesh cage center corresponding to the mesh cage and the feeding device according to the mesh cage features corresponding to the plurality of frames of first images.

[0071] The feeding device refers to a device used by unmanned feeding equipment to feed bait into the cage; the first relative position relationship may include information such as the relative distance and relative direction between the center of the cage and the feeding device.

[0072] In some specific embodiments, the unmanned feeding device can determine the image position corresponding to the cage in each frame of the first image, the size of the image area occupied by the cage in the first image, and the shape of the floating body component contained in the cage by analyzing the corresponding cage features in each frame of the first image. The unmanned feeding device can determine the image coordinates of the center of the cage in each frame of the first image according to the image position corresponding to the cage in each frame of the first image, the size of the image area occupied by the cage in the first image, and the shape of the floating body component contained in the cage; convert the image coordinates to the world coordinate system to obtain the position of the center of the cage in the world coordinate system; according to the position of the feeding device in the world coordinate system and the position of the center of the cage in the world coordinate system, obtain the first relative position relationship between the center of the cage and the feeding device in the world coordinate system. Among them, the origin of the world coordinate system can be a preset fixed origin, or the position of the feeding device (the discharge port of the feeding device) can be the origin, or the designated position on the unmanned feeding device can be the origin, which is not limited here.

[0073] In some embodiments, the position of the cage center corresponding to the cage can be determined based on the cage features corresponding to multiple frames of first images; the first relative position relationship of the cage center with respect to the feeding device can be determined based on the position of the cage center, the position of the feeding device and the direction of the discharge port.

[0074] Step 240, controlling the feeding device to move according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait.

[0075] In some embodiments, the feeding device may include a discharge port, through which the feeding device discharges bait to the center of the cage, so that the bait falls accurately in the center of the cage. The bait is used to feed the fish; the bait can be stored in the feeding device, or there can be an independent storage bin on the unmanned feeding device for storing the bait. When the bait needs to be delivered, the unmanned feeding device controls the storage bin to move the bait to be delivered to the feeding device for delivery.

[0076] In some embodiments, the movement of the feeding device may be rotation and / or translation in various directions, for example, 360-degree rotation at the same height, up and down movement to change the height of the feeding device, parallel movement at the same height, etc., which are not limited here. Therefore, in some embodiments, the unmanned feeding device may determine the corresponding moving direction and moving angle of the feeding device according to the first relative position relationship; and control the feeding device to move according to the moving direction and moving angle, so that the feeding device continuously aligns with the center of the cage to feed.

[0077] In some embodiments, before the unmanned feeding device controls the feeding device to align with the center of the net cage to release the bait, it can also obtain the wind speed and wind direction detected by the wind speed and wind direction sensor, and obtain the water flow speed and water flow direction detected by the water flow sensor, and move the feeding device again by analyzing the influence of the wind speed, wind direction, water flow speed and water flow direction on the net cage and the unmanned feeding device. Specifically, the unmanned feeding device can move against the wind direction and / or against the water flow direction to avoid the unmanned feeding device being pushed by the thrust (wind force and / or water force), thereby causing position deviation, so that the feeding device cannot be aligned with the center of the net cage; the unmanned feeding device can calculate the magnitude of the thrust received according to the wind speed and water flow speed, and determine the position of the feeding device to move again according to the magnitude of the thrust. Thereby reducing the influence of the sea breeze and water flow on the water, so that the feeding device can accurately align with the center of the net cage to release the bait.

[0078] In an embodiment of the present application, the unmanned feeding equipment can accurately obtain the first relative position relationship between the center of the cage and the feeding device based on the cage features corresponding to the multiple frames of first images collected by the first camera, and can control the unmanned feeding equipment to move the feeding device, so that the feeding device can accurately deliver bait to the center of the cage, thereby achieving precise control of bait delivery, so that the bait can fall into the center of the cage, thereby realizing the precise feeding operation of the unmanned feeding equipment, improving the accuracy of bait delivery by the unmanned feeding equipment, and the bait delivered to the center of the cage can be evenly distributed in the cage, reducing the waste of bait and the time it takes for fish to obtain bait, thereby improving the breeding efficiency.

[0079] like Figure 3 As shown, in another embodiment, a control method for an unmanned feeding device is provided, which can be applied to the above-mentioned unmanned feeding device. The method may include the following steps:

[0080] Step 302, acquiring multiple frames of second images by capturing images of the cage through a second camera.

[0081] In order to enable the unmanned feeding device to obtain the first image faster and thus increase the feeding rate, a second camera can be set on the unmanned feeding device to assist the unmanned feeding device in finding the cage and controlling the first camera to capture images of the cage.

[0082] The second camera can be located at any position on the unmanned feeding device, and there is no limit on the number of second cameras. That is to say, there can be multiple second cameras, and they are located in various directions of the unmanned feeding device. During the operation of the unmanned feeding device, the second camera can continuously capture images until the cage is detected in the image captured by the second camera, and then the frame image is used as the second image. The unmanned feeding device can identify the position of the cage through the image captured by the second camera. Since the field of view of the second camera is larger than the field of view of the first camera, the second camera can capture the cage more easily than the first camera. The unmanned feeding device can determine the position of the cage based on the second image, so that the unmanned feeding device can move to the cage position faster, so that the unmanned feeding device controls the first camera to capture the first image, reduces the time of searching for the cage, and increases the feeding rate.

[0083] Step 304: Perform target recognition on the net cages included in the multiple frames of the second image, and determine a second relative position relationship between the net cages and the unmanned feeding device according to the target recognition result.

[0084] Target recognition refers to detecting the cages contained in the second image and marking the image position of the cages in the second image to obtain the outline of the cages or a block diagram of the cages. The target result obtained by target recognition of the cages may include the position data, status information and other cage-related data of the cages floating on the water in the second image. In some embodiments, the unmanned feeding device may perform target recognition on the cages contained in multiple frames of the second image respectively, thereby identifying the position data and status information of the cages floating on the water, and determining the second relative position relationship between the cages and the unmanned feeding device based on the position information and status information. The second relative position relationship may include information such as the relative distance and relative direction between the cages and the unmanned feeding device.

[0085] Step 306: According to the second relative position relationship, the unmanned feeding equipment and / or the feeding device are controlled to move so that the feeding device is aligned with the cage.

[0086] In some embodiments, the unmanned feeding device can determine the movement information corresponding to the unmanned feeding device and / or the feeding device according to the second relative position relationship, and control the unmanned feeding device and / or the feeding device to move. The movement information may include movement angle, movement distance, movement speed, etc.

[0087] In an embodiment of the present application, the unmanned feeding device is provided with a second camera, and the field of view of the second camera is larger than the field angle of the first camera. Therefore, the second camera can more easily obtain the second relative position relationship between the cage and the unmanned feeding device, so that the unmanned feeding device can move according to the second relative position relationship, so that the feeding device can be aimed at the cage faster, and the first camera on the feeding device can also capture the image of the cage to obtain a first image, so that by analyzing the first image, the feeding device can be aimed at the center of the cage faster to release bait, thereby improving the feeding rate.

[0088] like Figure 4 As shown, in another embodiment, a control method for an unmanned feeding device is provided, which can be applied to the above-mentioned unmanned feeding device. The method may include the following steps:

[0089] Step 402, calculating the relative distance between the discharge port of the moved feeding device and the center of the mesh box.

[0090] In some embodiments, the feeding device releases the bait through the corresponding discharge port. Therefore, to align the feeding device with the center of the cage, it is necessary to align the discharge port of the feeding device with the center of the cage. That is to say, in the first image captured by the first camera, the discharge port of the feeding device must be aligned with the center of the cage so that the bait released by the feeding device can fall in the center of the cage.

[0091] In some embodiments, the relative length between the discharge port of the feeding device and the first camera is a fixed value, or the unmanned feeding device can control the length change of the discharge port to determine the relative length between the discharge port and the first camera, so the unmanned feeding device can calculate the relative distance between the discharge port of the feeding device after the movement and the center of the cage through the first relative position relationship and the relative length between the discharge port and the first camera. The relative distance may include relative height, horizontal distance, and straight-line distance between the discharge port and the center of the cage.

[0092] In some embodiments, the unmanned feeding device can calculate the relative distance between the outlet of the moved feeding device and the center of the cage by determining the longitude and latitude corresponding to the center of the cage and the outlet of the moved feeding device respectively.

[0093] In some specific embodiments, the unmanned feeding device can determine the coordinates of the center of the cage and the discharge port in the world coordinate system respectively through the first relative position relationship; and calculate the relative distance between the center of the cage and the discharge port in the world coordinate system according to the corresponding coordinates. Specifically, the unmanned feeding device can use algorithms such as vector inner product or Pythagorean theorem to calculate the relative distance between two points in the world coordinate system.

[0094] Step 404, determining the bait launching force corresponding to the feeding device according to the relative distance and the discharge amount.

[0095] Among them, the output may include the type, weight, volume, etc. of bait required by the fish swarm currently cultured in the cage. The bait launching force refers to the force generated when the feeding device launches the bait; due to the influence of the gravity of the bait itself and other ocean factors, the bait launching force will determine the position where the bait falls into the water surface. Therefore, if the bait needs to fall into the center of the cage, the bait launching force corresponding to the feeding device needs to be accurately controlled.

[0096] In some embodiments, if the feeding port of the feeding device is perpendicular to the water surface and aligned with the center of the cage, the feeding device can directly feed the bait to the center of the cage through the feeding port according to the discharge amount.

[0097] In some embodiments, if the discharge port of the feeding device is not perpendicular to the water surface, the bait launched by the feeding device through the feeding port will generally fall into the water surface in a parabolic trajectory. Therefore, the bait launching force corresponding to the feeding device will affect the position where the bait falls into the water surface. The discharge amount may be proportional to the bait launching force. Under the same relative distance, the larger the discharge amount, the greater the influence of gravity on the bait in the vertical direction, and the greater the bait launching force; under the same discharge amount, the higher the relative height of the discharge port to the center of the cage, the greater the horizontal distance, and the greater the bait launching force.

[0098] In some embodiments, the unmanned feeding equipment can obtain the sea surface wind speed and sea surface wind direction collected by the wind speed and wind direction sensor; according to the sea surface wind speed, sea surface wind direction, relative distance and discharge amount, the bait launching force corresponding to the feeding device is calculated. Among them, the greater the sea surface wind speed, the greater the bait launching force; if the sea surface wind direction is along the direction of the bait at the feeding port, the bait launching force is reduced; if the sea surface wind direction is opposite to the direction of the bait at the feeding port, the bait launching force is increased. Thereby reducing the influence of the sea surface wind on the launching process.

[0099] Step 406, controlling the feeding device to launch the bait according to the bait launching force, so that the bait is dropped into the center of the net cage.

[0100] In some embodiments, during the time period of bait delivery, the environmental factors of the unmanned feeding device and the cage may change, causing the cage center of the cage to change. If the bait delivery force is fixed during this process, the position of the bait falling on the water surface will be relatively fixed, resulting in the inability to continuously deliver the bait to the center of the cage. Therefore, the unmanned feeding device can continuously adjust the bait delivery force according to the actual situation during the delivery of bait, so that the bait can be continuously delivered to the center of the cage.

[0101] In an embodiment of the present application, after the unmanned feeding device moves the unmanned feeding device and / or the feeding device according to the second relative position relationship and determines that the feeding device in the first image is aligned with the center of the cage, the relative distance between the discharge port of the feeding device and the center of the cage and the discharge amount and other aspects are comprehensively considered, and the position of the bait dropped on the water surface is controlled by adjusting the bait launching force, so that the bait launched by the feeding device can be accurately dropped in the center of the cage, so that the bait can be evenly diffused and distributed in the cage starting from the center of the cage, reducing the waste of bait and the time for fish to obtain bait, thereby improving the breeding efficiency; and, using unmanned feeding equipment to breed fish in cages can substantially reduce manual operation, which not only avoids the risk of manual travel to deep-sea cages, but also reduces errors in manual feeding, greatly improves the accuracy of feeding and reduces breeding costs.

[0102] like Figure 5 As shown, in another embodiment, a control method for an unmanned feeding device is provided, which can be applied to the above-mentioned unmanned feeding device. The method may include the following steps:

[0103] Step 502: Acquire multiple frames of first images by capturing images of the cage through a first camera.

[0104] The description of step 502 may refer to the relevant description of step 210 in the above embodiment, which will not be repeated here.

[0105] Step 504: Identify the grid cages contained in each frame of the first image, and determine the target image area corresponding to the grid cages in each frame of the first image.

[0106] The target image area refers to the area in the first image that contains the cages; the target image area can be the largest circumscribed rectangular area in the first image that contains the cages. In some embodiments, the unmanned feeding equipment can perform target recognition on the cages contained in the first image through various target recognition algorithms to obtain the target image area corresponding to the cages in each frame of the first image. The target recognition algorithm may include but is not limited to machine learning-based methods (artificial feature + classifier method), deep learning-based methods (including Faster-RCNN (Faster Regions with CNN features) framework, SSD (Single Shot MultiBox Detector) framework, FPN (Feature Pyramid Network) framework, YOLO (you only look once) framework, etc.).

[0107] In some specific embodiments, the YOLOV7 network architecture can be used to identify the network box included in the first image pair, and obtain the target image area corresponding to the network box in each frame of the first image. The YOLOV7 network architecture is a typical target detection method based on deep learning, and belongs to the seventh generation network architecture of the YOLO series evolution. The basic components of the YOLOV7 network architecture include an input module, a backbone module, and a head module. Among them, the input module is used to obtain the first image, mainly through the backbone module to obtain the feature map corresponding to the first image, and then the head module is used for training and reasoning to obtain the corresponding target image area in the first image.

[0108] In some embodiments, after determining the target image area corresponding to the cage in each frame of the first image, the unmanned feeding device can use a foreground segmentation method to perform foreground segmentation on the target image area. The foreground segmentation method that can be used includes but is not limited to a threshold-based segmentation method (gray threshold method, region growing algorithm, etc.), an edge detection-based segmentation method (gradient algorithm processing, Roberts (cross differential) algorithm, Laplacian (Laplace) algorithm, etc.), a segmentation algorithm based on an active contour model (active contours), a segmentation algorithm based on a large SAM segmentation model, etc. The target image area is segmented from the first image to facilitate subsequent feature extraction and avoid feature extraction errors.

[0109] Step 506 , identifying and extracting features of the floating body components contained in each target image region, and obtaining features of the floating body components corresponding to each target image region.

[0110] Since the cage floats on the water surface through multiple floating body components, the unmanned feeding equipment can determine the corresponding position of the cage on the water surface by identifying and analyzing the floating body components, thereby determining the center of the cage. The floating body component refers to an object with floating ability, which is a special floating body for cage aquaculture facilities; the floating body component may include but is not limited to a buoy, a floating barrel, a float, etc. The floating body component features may include the position of each floating body component in the target image area, the size of the image area occupied by each floating body component in the target image area, the shape of each floating body component, etc.

[0111] Figure 1b is one frame of the first image 150 captured by the first camera 130, such as Figure 1b As shown, the first image 150 is an image content including the net cage 140. In the process of the unmanned feeding device 110 feeding the fish school, the unmanned feeding device 110 can control the first camera 130 to continuously collect images, and the unmanned feeding device 110 obtains multiple frames of first images 150 collected by the first camera 130; the unmanned feeding device 110 identifies the net cage 140 included in each frame of the first image 150, and determines the target image area 160 corresponding to the net cage 140 in each frame of the first image; identifies and extracts features of the floating body components included in the target image area 160, and obtains the floating body component features corresponding to the target image area 160; and calculates the net cage center ( Figure 1a The first relative position relationship between the net cage 140 (black dot in the figure) and the feeding device 120; the unmanned feeding device 110 can control the feeding device 120 to move according to the first relative position relationship, so that the feeding device 120 is aimed at the center of the net cage to feed the bait, thereby completing the accurate feeding of the fish school by the unmanned feeding device 110.

[0112] In some embodiments, the floating body components contained in the first target image area are identified and features are extracted to obtain multiple target detection frames contained in the first target image area; the first target image area is the target image area in any frame of the first image; each target detection frame corresponds to at least one floating body component in the cage; based on the feature information corresponding to the multiple target detection frames, the features of the floating body components corresponding to the first target image area are determined. The target detection frame is the maximum circumscribed rectangular frame corresponding to at least one floating body component in the first target image area; the feature information includes the image position of each target detection frame, the detection frame size of each target detection frame, and the change of multiple target detection frames.

[0113] The specific method of extracting features of the floating body components contained in the target image area in step 506 can refer to the method of extracting features of the cage contained in the first image in step 220, which will not be repeated here.

[0114] Step 508: Determine the image coordinates of the cage center corresponding to the cage in each frame of the first image according to the floating body component features corresponding to the target image area in each frame of the first image.

[0115] In some embodiments, the unmanned feeding device can determine the deformation information of the cage in the target image area contained in each frame of the first image through the features of the attached components corresponding to the target image area in each frame of the first image; and calculate the image coordinates of the cage center corresponding to the cage in each frame of the first image through the deformation information. The deformation information may include the arc of the cage in the target image area, the side length of the cage in the target image area, etc.

[0116] In some specific embodiments, the unmanned feeding equipment can determine the image coordinates of the cage center corresponding to the cage in each frame of the first image by analyzing the changing relationship between the curvature of the cage in the target image area in each frame of the first image and the original curvature of the cage; the original curvature of the cage refers to the curvature of the cage when it is first placed in the water.

[0117] Step 510, calculating the first relative position relationship between the center of the cage and the feeding device according to the image coordinates corresponding to the center of the cage in each frame of the first image.

[0118] In some embodiments, the unmanned feeding equipment can convert the image coordinates corresponding to the center of the cage in each frame of the first image into the second coordinates of the center of the cage in the world coordinate system; based on the coordinates corresponding to the feeding device in the world coordinate system and the second coordinates, determine the first relative position relationship between the center of the cage and the feeding device.

[0119] In some embodiments, the unmanned feeding device can detect the offset coordinates of the image coordinates corresponding to the center of the cage in each frame of the first image, remove the offset coordinates that are not within the target range, and cluster the image coordinates within the target range to determine multiple cluster centers corresponding to multiple image coordinates; the mean of the multiple cluster centers is used as the cage center, and the first relative position relationship between the cage center and the feeding device is calculated. The clustering methods that can be used include but are not limited to partition-based clustering methods, density-based clustering methods, hierarchical clustering methods, etc.

[0120] In some specific embodiments, the unmanned feeding device can use the k-means clustering method to cluster the image coordinates within the target range to obtain M cluster centers corresponding to the multiple image coordinates; the mean of the multiple cluster centers is used as the cage center, and the first relative position relationship between the cage center and the feeding device is calculated, where M is a positive integer. The k-means clustering method belongs to the partitioning clustering method, which requires the number of clusters or cluster centers to be specified in advance, and iterates repeatedly until the goal of "points within the cluster are close enough, and points between clusters are far enough" is finally achieved.

[0121] In some embodiments, the unmanned feeding device can determine the first relative position relationship between the center of the cage and the feeding device according to the relative position relationship between the image coordinates corresponding to the center of the cage in each frame of the first image and the feeding device; according to the average of each relative position relationship. For example, the relative position relationship may include information such as the relative distance and relative direction between the center of the cage and the feeding device, wherein the relative direction between the center of the cage and the feeding device is defined as 0 degrees and 360 degrees due east, 90 degrees due north, 180 degrees due west, and 270 degrees due south. Assume that the unmanned feeding device acquires three frames of first images, and the relative position relationship between the image coordinates corresponding to the center of the cage in each frame of the first image and the feeding device is (relative direction 66 degrees, relative distance 300 meters), (relative direction 24 degrees, relative distance 360 ​​meters) and (relative direction 81 degrees, relative distance 210 meters); then the first relative position relationship between the center of the cage and the feeding device is (relative direction 57 degrees, relative distance 290 meters).

[0122] Step 512, controlling the feeding device to move according to the first relative position relationship, so that the feeding device is aligned with the center of the cage to feed the bait.

[0123] The description of step 512 may refer to the relevant description of step 240 in the above embodiment, which will not be repeated here.

[0124] In the embodiment of the present application, the unmanned feeding device can more accurately analyze and obtain the image coordinates corresponding to the center of the cage corresponding to the cage in each frame of the first image according to the characteristics of the floating body components corresponding to each target image area, thereby obtaining a first relative position relationship with a higher accuracy between the center of the cage and the feeding device; the unmanned feeding device can move the feeding device so that the feeding device can more accurately deliver bait to the center of the cage, realize accurate control of bait delivery, and enable the bait to accurately fall into the center of the cage, thereby realizing the accurate feeding operation of the unmanned feeding device, improving the accuracy of the unmanned feeding device in delivering bait, and the bait delivered to the center of the cage can be evenly distributed in the cage, reducing the waste of bait and the time for fish to obtain bait, thereby improving the breeding efficiency. In addition, the use of unmanned feeding equipment to breed fish in the cage can fully reduce manual operation, avoid the risk of manual travel to the deep-sea cage, and reduce the errors of manual feeding, greatly improve the accuracy of feeding and reduce breeding costs.

[0125] like Figure 6 As shown, in one embodiment, a control device 600 for an unmanned feeding device is provided, which can be applied to the unmanned feeding device described above. The control device 600 for the unmanned feeding device can include an image acquisition module 610, a feature extraction module 620, a central calculation module 630, and a mobile feeding module 640.

[0126] The image acquisition module 610 is used to acquire multiple frames of first images by collecting images of the cage through a first camera.

[0127] The feature extraction module 620 is used to extract features of the grid boxes contained in each frame of the first image to obtain grid box features corresponding to each frame of the first image.

[0128] The center calculation module 630 is used to calculate the first relative position relationship between the center of the cage corresponding to the cage and the feeding device according to the cage features corresponding to the multiple frames of first images.

[0129] The rotating feeding module 640 is used to control the movement of the feeding device according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait.

[0130] Optionally, the unmanned feeding device also includes a second camera, and the field of view of the second camera is greater than the field of view of the first camera.

[0131] As an optional implementation, the control device 600 of the unmanned feeding equipment also includes a cage identification module and a force calculation module.

[0132] The net cage recognition module is used to obtain multiple frames of second images obtained by capturing images of the net cage through a second camera; to identify the net cage contained in the multiple frames of the second image, and to determine the second relative position relationship between the net cage and the unmanned feeding device according to the target recognition result; and to control the unmanned feeding device and / or the feeding device to move according to the second relative position relationship so that the feeding device is aligned with the net cage.

[0133] The force calculation module is used to calculate the relative distance between the discharge port of the feeding device after movement and the center of the cage; determine the bait launching force corresponding to the feeding device according to the relative distance and the discharge amount; control the feeding device to launch the bait according to the bait launching force so that the bait is placed in the center of the cage.

[0134] Optionally, the unmanned feeding equipment further includes a wind speed and direction sensor, and the wind speed and direction sensor is located on the feeding device.

[0135] The force calculation module is also used to obtain the sea surface wind speed and sea surface wind direction collected by the wind speed and wind direction sensor; according to the sea surface wind speed, sea surface wind direction, relative distance and discharge amount, the bait launching force corresponding to the feeding device is calculated.

[0136] Optionally, the cage comprises a plurality of buoyancy assemblies.

[0137] In some embodiments, the feature extraction module 620 is also used to identify the cage contained in the first image of each frame, and determine the target image area corresponding to the cage in the first image of each frame; identify and extract features of the floating body components contained in each target image area, and obtain the floating body component features corresponding to each target image area.

[0138] Optionally, the feature extraction module 620 is also used to identify and extract features of the floating components contained in the first target image area, and obtain multiple target detection frames contained in the first target image area; the first target image area is the target image area in any frame of the first image; based on the feature information corresponding to the multiple target detection frames, the features of the floating components corresponding to the first target image area are determined; each target detection frame corresponds to at least one floating component in the cage.

[0139] In some embodiments, the center calculation module 630 is also used to determine the image coordinates of the mesh box center corresponding to the mesh box in each frame of the first image based on the characteristics of the floating body component corresponding to the target image area in each frame of the first image; and calculate the first relative position relationship between the mesh box center and the feeding device based on the image coordinates of the mesh box center in each frame of the first image.

[0140] Optionally, the rotating feeding module 640 is also used to determine the corresponding moving direction and moving angle of the feeding device according to the first relative position relationship; according to the moving direction and moving angle, the feeding device is controlled to move so that the feeding device continues to align with the center of the cage to feed bait.

[0141] In an embodiment of the present application, the unmanned feeding equipment can accurately obtain the first relative position relationship between the center of the cage and the feeding device based on the cage features corresponding to the multiple frames of first images collected by the first camera, and can control the unmanned feeding equipment to move the feeding device, so that the feeding device can accurately deliver bait to the center of the cage, thereby achieving precise control of bait delivery, so that the bait can fall into the center of the cage, thereby realizing the precise feeding operation of the unmanned feeding equipment, improving the accuracy of bait delivery by the unmanned feeding equipment, and the bait delivered to the center of the cage can be evenly distributed in the cage, reducing the waste of bait and the time it takes for fish to obtain bait, thereby improving the breeding efficiency.

[0142] Figure 7 FIG. 1 is a structural block diagram of an unmanned feeding device in one embodiment. Figure 7 As shown, the unmanned feeding device 700 may include one or more of the following components: a processor 710, a memory 720 coupled to the processor 710, and a feeding device 730. The memory 720 may store one or more computer programs, and the one or more computer programs may be configured to implement the methods described in the above embodiments when executed by one or more processors 710; the feeding device 730 may include a feeding port for feeding bait.

[0143] The processor 710 may include one or more processing cores. The processor 710 uses various interfaces and lines to connect the various parts of the entire unmanned feeding device 700, and executes various functions and processes data of the unmanned feeding device 700 by running or executing instructions, programs, code sets or instruction sets stored in the memory 720, and calling data stored in the memory 720. Optionally, the processor 710 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 710 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 710, but may be implemented separately through a communication chip.

[0144] The memory 720 may include a random access memory (RAM) or a read-only memory (ROM). The memory 720 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 720 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. The data storage area may also store data created by the unmanned feeding device 700 during use, etc.

[0145] It can be understood that the unmanned feeding device 700 may include more or fewer structural elements than those in the above structural block diagram, for example, including a power supply, input buttons, cameras, speakers, screens, RF (Radio Frequency) circuits, Wi-Fi (Wireless Fidelity) modules, Bluetooth modules, sensors, etc., and are not limited here.

[0146] An embodiment of the present application discloses a storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method described in the above embodiments.

[0147] The embodiments of the present application disclose a computer program product, including a computer program, and the computer program can be executed by a processor to implement the methods described in the above embodiments.

[0148] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM), etc.

[0149] As used herein, any reference to memory, storage, database, or other medium may include nonvolatile and / or volatile memory. Suitable nonvolatile memory may include ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), or Flash memory. Volatile memory may include random access memory (RAM), which is used as an external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), memory bus direct RAM (RDRAM), and direct memory bus dynamic RAM (DRDRAM).

[0150] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. Those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required for the present application.

[0151] In the various embodiments of the present application, it should be understood that the size of the serial numbers of the above-mentioned processes does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0152] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0155] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several requests for a computer device (which can be a personal computer, a server or a network device, etc., specifically a processor in a computer device) to perform some or all of the steps of the above-mentioned methods of various embodiments of the present application.

[0156] The above is a detailed introduction to the control method, device, equipment and storage medium of an unmanned feeding device disclosed in the embodiment of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those of ordinary skill in the art, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A control method for unmanned feeding equipment, characterized in that: Applied to unmanned feeding equipment, the unmanned feeding equipment includes a feeding device and a first camera, the first camera is located on the feeding device, the unmanned feeding equipment also includes a second camera, the field of view of the second camera is greater than the field of view of the first camera, the second camera is used to assist the unmanned feeding equipment to find the cage and control the first camera to collect images of the cage; the method includes: Acquire multiple frames of second images obtained by capturing images of the cage through the second camera; Performing target recognition on the net cages included in the plurality of frames of second images, and determining a second relative position relationship between the net cages and the unmanned feeding device according to the target recognition result; According to the second relative position relationship, controlling the unmanned feeding equipment and / or the feeding device to move so that the feeding device is aligned with the net cage; Acquire multiple frames of first images obtained by capturing images of the cage through the first camera; Extracting features of the mesh boxes contained in each frame of the first image to obtain mesh box features corresponding to each frame of the first image; Calculating a first relative position relationship between a net cage center corresponding to the net cage and the feeding device according to the net cage features respectively corresponding to the multiple frames of first images; The feeding device is controlled to move according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait.

2. The method according to claim 1, characterized in that After controlling the feeding device to move according to the first relative position relationship, the method further includes: Calculating the relative distance between the discharge port of the feeding device after the movement and the center of the net cage; Determine the bait launching force corresponding to the feeding device according to the relative distance and the discharge amount; The feeding device is controlled to launch the bait according to the bait launching force so that the bait is dropped into the center of the net cage.

3. The method according to claim 2, characterized in that The unmanned feeding equipment further comprises a wind speed and direction sensor, and the wind speed and direction sensor is located on the feeding device; Determining the bait launching force corresponding to the feeding device according to the relative distance and the discharge amount includes: Obtaining the sea surface wind speed and sea surface wind direction collected by the wind speed and wind direction sensor; The bait launching force corresponding to the feeding device is calculated according to the sea surface wind speed, the sea surface wind direction, the relative distance and the discharge amount.

4. The method according to claim 1, characterized in that: The mesh cage includes a plurality of floating body components; extracting features of the mesh cage included in each frame of the first image to obtain mesh cage features corresponding to each frame of the first image includes: Identify the mesh cage contained in each frame of the first image, and determine the target image area corresponding to the mesh cage in each frame of the first image; The floating body components contained in each target image area are identified and features are extracted to obtain the floating body component features corresponding to each target image area.

5. The method according to claim 4, characterized in that The identifying and feature extracting the floating body components contained in each target image area to obtain the floating body component features corresponding to each target image area includes: Identify and extract features of the floating body components contained in the first target image area to obtain a plurality of target detection frames contained in the first target image area; the first target image area is a target image area in any frame of the first image; Determine the floating component features corresponding to the first target image area according to the feature information corresponding to the multiple target detection frames; each of the target detection frames corresponds to at least one floating component in the cage; The calculating, based on the cage features respectively corresponding to the multiple frames of first images, a first relative position relationship between the cage center corresponding to the cage and the feeding device comprises: Determine the image coordinates of the cage center corresponding to the cage in each frame of the first image according to the floating body component features corresponding to the target image area in each frame of the first image; The first relative position relationship between the center of the cage and the feeding device is calculated according to the image coordinates corresponding to the center of the cage in each frame of the first image.

6. The method according to claim 1, characterized in that The step of controlling the feeding device to move according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait comprises: According to the first relative position relationship, determining the moving direction and moving angle corresponding to the feeding device; According to the moving direction and the moving angle, the feeding device is controlled to move so that the feeding device continuously aims at the center of the cage to feed the bait.

7. A control device for unmanned feeding equipment, characterized in that: Applied to unmanned feeding equipment, the unmanned feeding equipment includes a feeding device and a first camera, the first camera is located on the feeding device, the unmanned feeding equipment also includes a second camera, the field of view of the second camera is greater than the field of view of the first camera, the second camera is used to assist the unmanned feeding equipment to find the cage and control the first camera to collect images of the cage; The control device of the unmanned feeding equipment comprises: a net cage recognition module, configured to obtain a plurality of frames of second images obtained by collecting images of the net cage through the second camera; perform target recognition on the net cage contained in the plurality of frames of second images, and determine a second relative position relationship between the net cage and the unmanned feeding device according to the target recognition result; and control the unmanned feeding device and / or the feeding device to move according to the second relative position relationship so that the feeding device is aligned with the net cage; An image acquisition module, used to acquire multiple frames of first images obtained by collecting images of the cage through the first camera; A feature extraction module, used to extract features of the cage contained in each frame of the first image, to obtain features of the cage corresponding to each frame of the first image; A center calculation module, used for calculating a first relative position relationship between a net box center corresponding to the net box and the feeding device according to the net box features respectively corresponding to the multiple frames of first images; The rotating feeding module is used to control the movement of the feeding device according to the first relative position relationship so that the feeding device is aligned with the center of the cage to feed the bait.

8. An unmanned feeding device, characterized in that: The method comprises a feeding device, a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor implements the method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images