An automated inspection system for aquaculture

The automated inspection system enables automatic lifting and visual recognition of the feeding basket, solving the problem of difficult manual operation and improving the automation and feeding accuracy of aquaculture.

CN119814980BActive Publication Date: 2025-10-31HUILAI COUNTY GONGDAOREN AGRI DEV CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510013185.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-01-06
Publication Date
2025-10-31
Estimated Expiration
2045-01-06

AI Technical Summary

Technical Problem

In current aquaculture, lifting and observing the feeding basket mainly relies on manual labor and visual inspection, which is cumbersome and labor-intensive, especially in the middle of the pond or in deep water, resulting in uneven feeding.

Method used

An automated inspection system is adopted, including a control center, a feeding device, and a camera recognition device, to realize the automatic lifting and visual recognition of the feeding basket, and to accurately deliver feed and trace elements or medicines in combination with the feeding device.

Benefits of technology

It enables automated lifting and observation of the feeding basket, reduces manual operation, improves the accuracy of feeding and the degree of automation in aquaculture, and reduces the need for manpower.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119814980B_ABST
    Figure CN119814980B_ABST
Patent Text Reader

Abstract

This invention relates to the field of aquaculture technology, and in particular to an automated aquaculture inspection system, comprising a control center, a lifting device for lifting and returning a feeding basket, and a camera recognition device for visually recognizing the feeding basket lifted by the lifting device. This automated aquaculture inspection system can replace manual lifting and returning of the feeding basket in the water, and can replace manual observation, thus achieving automated inspection and being better suited for aquaculture.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of aquaculture technology, and in particular to an automated inspection system for aquaculture. Background Technology

[0002] With population growth and improved living standards, people's demand for seafood such as fish and shrimp is increasing. In order to meet people's demand for fish and shrimp while reducing the burden of marine resource development and better protecting the aquatic ecological environment, more and more people are starting to farm fish and shrimp.

[0003] Because the feeding habits of fish and shrimp vary depending on their growth stage and seasonal changes, both overfeeding and underfeeding can negatively impact their health and growth. Therefore, most current practices involve lifting the feeding basket and observing the remaining feed to adjust the amount of feed for the next feeding, thus optimizing feeding and preventing overfeeding or underfeeding. Currently, most of these methods involve manually lifting the feeding basket in the water and observing it visually. This is not only cumbersome and labor-intensive, but also difficult to do when the feeding basket is located in the middle of the pond or in a deep area. Summary of the Invention

[0004] The purpose of this invention is to provide an automated inspection system for aquaculture that addresses the shortcomings of existing technologies. This system can replace manual methods for lifting and placing feeding baskets in the water, and can replace manual visual inspection, thus achieving automated inspection and being better suited for aquaculture.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is: an automated inspection system for aquaculture, including a control center, a lifting device for lifting and returning a feeding basket, and a camera recognition device for visually recognizing the feeding basket lifted by the lifting device.

[0006] A further improvement to the above solution is that the automated inspection system for aquaculture of the present invention also includes a feeding device for adding feed and / or trace elements and / or medicines to the feeding basket.

[0007] A further improvement to the above solution is that the lifting device includes a guide rail, a base that slides along the guide rail, a traveling power mechanism for driving the base to slide, a lifting platform that is slidably connected to the bottom of the base in a vertical direction, a lifting power mechanism for driving the lifting platform to slide, and lifting claws disposed on the lifting platform.

[0008] A further improvement to the above solution is that the lifting claw includes a connecting part for connecting with the lifting platform, and the front and rear ends of the connecting part are respectively formed with bent parts extending downwards, and the bottom of the two bent parts are respectively formed with lifting parts extending horizontally in the direction close to the middle position of the connecting part.

[0009] The lifting power mechanism includes a lifting rope for connecting to the lifting platform and a rope power mechanism for winding or unwinding the lifting rope, the rope power mechanism being mounted on the machine base.

[0010] Telescopic hinge assemblies are respectively provided between the front end and / or rear end and / or both sides of the lifting platform and the base.

[0011] A further improvement to the above solution is that the guide rail is a rod-shaped structure, and the traveling power mechanism includes several traveling power modules. Each traveling power module includes two tapered rollers symmetrically arranged at the front and rear ends of the guide rail, and a roller power mechanism for driving the two tapered rollers to rotate. The two tapered rollers are respectively rolled on the front and rear ends of the guide rail, and the output end of the roller power mechanism drives and connects the two tapered rollers.

[0012] A further improvement to the above scheme is that the roller power mechanism includes two power output gears that rotate coaxially with the two tapered rollers, two power input gears, and a gear power mechanism for driving the two power input gears to rotate. A transmission chain is wound around the power output gears and power input gears on the same side. The power output gears are connected to the power input gears on the same side through corresponding transmission chains. The gear power mechanism is mounted on a base, and the output end of the gear power mechanism drives and connects to the two power input gears.

[0013] A further improvement to the above scheme is that the base moves back and forth along the guide rail in a straight line, a circle, or an oblong shape.

[0014] An automated inspection method for aquaculture, applicable to an automated inspection system for aquaculture comprising a control center, a feeding device for lifting and lowering a feeding basket, a camera recognition device for visually recognizing the feeding basket lifted by the feeding device, and a feeding device for adding feed and / or trace elements and / or medicines to the feeding basket, comprising the following steps:

[0015] A. The lifting device lifts the feeding basket located in the water upwards;

[0016] B. The camera recognition device visually identifies the remaining feed in the feeding basket lifted by the feeding device and / or the growth status of the fish and shrimp in the feeding basket.

[0017] C. The lifting device lowers the feeding basket back down after visual recognition is completed;

[0018] D. The feeding device feeds the corresponding feed according to the recognition results of the camera recognition device, including accurately feeding the feed according to the remaining feed and feeding the trace elements and / or medicines according to the growth status of the fish and shrimp.

[0019] A further improvement to the above scheme is that step D, which involves precisely dispensing feed based on the remaining feed, includes the following specific steps:

[0020] D1. Calculate the remaining feed area based on the divided feed areas;

[0021] D2. Compare the remaining feed area with the preset remaining feed area threshold.

[0022] D3. Adjust the next feed amount based on the difference between the remaining feed area and the preset remaining feed area threshold. The preset feed amount is N, the next feed amount is N1, the remaining feed area is S, and the preset remaining feed area threshold is S1. The preset feed amount N, the next feed amount N1, the remaining feed area S, and the preset remaining feed area threshold S1 satisfy the following condition: N1 = N(1 - (S - S1) / S1).

[0023] A further improvement to the above scheme is that step D1, calculating the remaining feed area based on the divided feed area, specifically includes the following steps:

[0024] D11. Acquire image information and preprocess the image information to extract image features;

[0025] D12. Based on the image features, construct a model to distinguish between feed and excrement;

[0026] D13. Using the discriminative model, the target region in the image is identified. An image segmentation algorithm is used to segment the target region in the image to obtain the segmented feed region and excrement region. Based on the segmented feed region, the remaining feed area is calculated.

[0027] The preprocessing of image information in step D11 specifically includes:

[0028] The image is denoised to obtain the denoised image;

[0029] Based on the denoised image, extract color features, shape features, and texture features;

[0030] The construction of the feed-excrement differentiation model in step D12 specifically includes:

[0031] Based on the extracted image features, construct a feature set for feed samples and a feature set for excrement samples;

[0032] A discrimination model is generated by training based on the feed sample feature set and the excrement sample feature set using machine learning algorithms. The machine learning algorithms include at least one of support vector machine, random forest and convolutional neural network.

[0033] The beneficial effects of the present invention are as follows: The present invention provides an automated inspection system for aquaculture, including a control center, a lifting device for lifting and returning the feeding basket, and a camera recognition device for visually recognizing the feeding basket lifted by the lifting device.

[0034] The control center controls the lifting device to lift the feeding basket located in the water. A camera recognition device visually identifies the remaining feed in the basket and / or the growth status of the fish and shrimp in the basket. The lifting device then lowers the basket back down after visual identification. This automated aquaculture inspection system can replace manual lifting and lowering of the feeding basket located in the water, and can replace manual observation, thus achieving automated inspection and being better suited for aquaculture. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of an automated inspection system for aquaculture according to the present invention.

[0036] Figure 2 This is a schematic diagram of the material lifting device of the present invention.

[0037] Explanation of reference numerals in the attached drawings: 1. Material lifting device; 11. Guide rail; 12. Machine base; 13. Traveling power mechanism; 14. Lifting platform; 151. Lifting rope; 152. Rope power mechanism; 153. Telescopic hinge assembly; 16. Lifting claw; 161. Connecting part; 162. Bending part; 163. Lifting part; 17. Traveling power module; 171. Tapered roller; 172. Roller power mechanism; 172. Power output gear; 1721. Power input gear; 1722. Gear power mechanism; 1723. Transmission chain; 1724. Camera recognition device; 2. Control center; 3. Feeding device; 4. Detailed Implementation

[0038] The invention will now be further described with reference to the accompanying drawings, such as... Figure 1-2As shown, the automated aquaculture inspection system of the present invention includes a control center 3, a feeding device 1 for lifting and lowering a feeding basket, and a camera recognition device 2 for visually recognizing the feeding basket lifted by the feeding device 1. The control center 3 controls the feeding device 1 to lift the feeding basket located in the water upwards, and the camera recognition device 2 visually recognizes the remaining feed in the feeding basket and / or the growth status of the fish and shrimp in the feeding basket. The feeding device 1 then lowers the feeding basket back down after visual recognition. The automated aquaculture inspection system of the present invention can replace manual lifting and lowering of the feeding basket located in the water, and can replace manual observation, thus achieving automated inspection and being better suited for aquaculture.

[0039] The automated inspection system for aquaculture of the present invention also includes a feeding device 4 for adding feed and / or trace elements and / or medicines to the feeding basket; the system automatically feeds the feed according to the visual recognition result of the camera recognition device 2, which can further improve the overall automation and intelligence of aquaculture, thereby further reducing the workload of staff.

[0040] The material lifting device 1 includes a guide rail 11, a base 12 that slides along the guide rail 11, a traveling power mechanism 13 for driving the base 12 to slide, a lifting platform 14 that is vertically connected to the lower part of the base 12, a lifting power mechanism for driving the lifting platform 14 to slide, and a lifting claw 16 disposed on the lifting platform 14. First, the lifting power mechanism drives the lifting claw 16 to move downward and lower the height of the lifting claw 16 below the top of the feeding basket. Then, the traveling power mechanism 13 drives the base 12 to slide and moves the lifting claw 16 to a position below the top of the feeding basket. Finally, the lifting power mechanism drives the lifting claw 16 to move upward and lift the top of the feeding basket upward. The feeding basket is lifted out of the water. After observing the remaining feed in the basket and / or the growth of the fish and shrimp in the basket, the lifting mechanism drives the lifting claw 16 to move downward and return the feeding basket to its original position. The lifting mechanism continues to drive the lifting claw 16 downward until its height is lower than the top of the feeding basket. Then, the traveling mechanism 13 drives the base 12 to slide and move the lifting claw 16 away from the feeding basket, completing the entire process of automatically lifting and returning the feeding basket. The feeding device 1 of the present invention can replace manual lifting and returning of the feeding basket, which can reduce the workload of operators.

[0041] The lifting claw 16 includes a connecting portion 161 for connecting to the lifting platform 14. The front and rear ends of the connecting portion 161 are respectively formed with bent portions 162 extending downward. The bottom of the two bent portions 162 are respectively formed with lifting portions 163 extending horizontally in a direction close to the middle position of the connecting portion 161. The distance between the two lifting portions 163 of the present invention is less than the width of the top of the feeding basket, and the distance between the two bent portions 162 is greater than the width of the top of the feeding basket, so as to realize the upward lifting and downward lowering of the top of the feeding basket.

[0042] The lifting power mechanism includes a lifting rope 151 for connecting to the lifting platform 14 and a rope power mechanism 152 for winding or unwinding the lifting rope 151. The rope power mechanism 152 is mounted on the base 12. When the lifting rope 151 is unwound by the rope power mechanism 152, the lifting platform 14 moves downward accordingly. When the lifting rope 151 is wound by the rope power mechanism 152, the lifting platform 14 moves upward accordingly. By controlling the unwinding or winding of the lifting rope 151 by the rope power mechanism 152, the lifting platform 14 can be raised or lowered accordingly, thereby controlling the raising or lowering of the lifting claw 16. Compared with the lifting platform 14 driven by a cylinder, motor and lead screw, the present invention drives the lifting platform 14 to rise or fall through the cooperation of the lifting rope 151 and the rope power mechanism 152, resulting in a simpler, lighter and more reasonable overall structure.

[0043] Telescopic hinge assemblies 153 are respectively provided between the front end and / or rear end and / or both sides of the lifting platform 14 and the base 12. The telescopic hinge assemblies 153 can automatically extend and retract synchronously while the lifting platform 14 is lifting. The setting of the telescopic hinge assemblies 153 can improve the structural stability between the lifting platform 14 and the base 12, thereby better ensuring that the lifting platform 14 can lift and lower smoothly, that is, better ensuring that the lifting claw 16 can accurately lift and put back the top of the feeding basket. In this embodiment, the lifting platform 14 and the base 12 are provided with telescopic hinge assemblies 153 at the front end and both sides, but no telescopic hinge assembly 153 is provided at the rear end. This can improve the structural stability between the lifting platform 14 and the base 12 and facilitate the operation of the camera module and other components set on the lifting platform 14.

[0044] The guide rail 11 has a rod-shaped structure. The traveling power mechanism 13 includes several traveling power modules 17. Each traveling power module 17 includes two tapered rollers 171 symmetrically arranged at the front and rear ends of the guide rail 11 and a roller power mechanism 172 for driving the two tapered rollers 171 to rotate. The two tapered rollers 171 roll on the front and rear ends of the guide rail 11 respectively. The output end of the roller power mechanism 172 drives and connects the two tapered rollers 171. Compared with the sliding of the base 12 by using a cylinder, motor and lead screw, the slide can only perform linear reciprocating motion. The traveling power mechanism 13 of this utility model realizes the sliding of the base 12 by using tapered rollers 171. It not only has a smooth overall transmission, but is also suitable for reciprocating movement of shapes such as circles or waists. Therefore, it can be better suited for lifting and putting back the feeding baskets in multiple aquaculture.

[0045] The roller power mechanism 172 includes two power output gears 1721 that rotate coaxially with the two tapered rollers 171, two power input gears 1722, and a gear power mechanism 1723 for driving the two power input gears 1722 to rotate. The power output gears 1721 and the power input gears 1722 on the same side are connected by a transmission chain 1724. The power output gears 1721 are connected to the power input gears 1722 on the same side through the corresponding transmission chain 1724. The gear power mechanism 1723 is mounted on the base 12. The output end of the gear power mechanism 1723 drives the two power input gears 1722 to rotate. The two power input gears 1722 drive the two power output gears 1721 to rotate through the corresponding transmission chain 1724, thereby driving the two tapered rollers 171 to rotate and moving the base 12 along the guide rail 11.

[0046] In this embodiment, the number of teeth of the power input gear 1722 is greater than the number of teeth of the power output gear 1721, which can increase the rotational speed of the tapered roller 171, thereby increasing the moving speed of the base 12.

[0047] The base 12 moves back and forth along the guide rail 11 in a straight line, a circle, or an oblong shape. In this embodiment, the base 12 moves back and forth along the guide rail 11 in an oblong shape, which can better lift and put back multiple feeding baskets in aquaculture, making it more practical.

[0048] Compared to setting up camera recognition devices 2 in multiple locations, the camera recognition device 2 in this embodiment is set on the lifting platform 14 and can move in real time with the feeding device 1, thereby better realizing visual recognition of the feeding baskets at different locations; the camera recognition device 2 can take pictures and visually recognize the remaining feed in the feeding basket and / or the growth status of the fish and shrimp in the feeding basket, thereby replacing manual observation and further improving the overall automation. The visual recognition results of the camera recognition device 2 can be used to remind or automatically realize the feeding, addition of trace elements or medicines, etc.

[0049] An automated inspection method for aquaculture, applicable to an automated inspection system for aquaculture comprising a control center 3, a feeding device 1 for lifting and returning a feeding basket, a camera recognition device 2 for visually recognizing the feeding basket lifted by the feeding device 1, and a feeding device 4 for adding feed and / or trace elements and / or medicines to the feeding basket, comprising the following steps:

[0050] A. The lifting device 1 lifts the feeding basket located in the water upwards;

[0051] B. The camera recognition device 2 performs visual recognition on the remaining feed in the feeding basket lifted by the feeding device 1 and / or the growth status of the fish and shrimp in the feeding basket.

[0052] C. The material lifting device 1 lowers the feeding basket after visual recognition is completed.

[0053] D. Feeding device 4 feeds the corresponding feed according to the recognition result of camera recognition device 2, including accurately feeding feed according to the remaining feed and feeding trace elements and / or medicines according to the growth status of fish and shrimp.

[0054] The present invention provides an automated inspection method for aquaculture, which enables automated and intelligent aquaculture.

[0055] Specifically, step D, which involves precisely dispensing feed based on the remaining feed, includes the following steps:

[0056] D1. Calculate the remaining feed area based on the divided feed areas;

[0057] D2. Compare the remaining feed area with the preset remaining feed area threshold.

[0058] D3. Adjust the next feed amount based on the difference between the remaining feed area and the preset remaining feed area threshold. The preset feed amount is N, the next feed amount is N1, the remaining feed area is S, and the preset remaining feed area threshold is S1. The preset feed amount N, the next feed amount N1, the remaining feed area S, and the preset remaining feed area threshold S1 satisfy the following condition: N1 = N1 - (S - S1) / S1. For example, if the remaining feed area S is 80% of the preset remaining feed area threshold S1, then the next feed amount N1 is... 1.2N; For example, if the remaining feed area S is 150% of the preset remaining feed area threshold S1, then the next feed amount N1 is 0.5N; Compared to simply judging whether the remaining feed area is greater than or less than the preset remaining feed area and then simply increasing or decreasing the next feed amount, satisfying the following condition between the preset feed amount N, the next feed amount N1, the remaining feed area S, and the preset remaining feed area threshold S1: N1=N1-(S-S1) / S1 can achieve more precise feeding and better feeding in aquaculture.

[0059] Step D1, calculating the remaining feed area based on the divided feed areas, specifically includes the following steps:

[0060] D11. Acquire image information and preprocess the image information to extract image features;

[0061] D12. Based on the image features, construct a model to distinguish between feed and excrement;

[0062] D13. Using the discriminative model, the target region in the image is identified. An image segmentation algorithm is used to segment the target region in the image to obtain the segmented feed region and excrement region. Based on the segmented feed region, the remaining feed area is calculated.

[0063] The preprocessing of image information in step D11 specifically includes:

[0064] The image is denoised to obtain the denoised image;

[0065] Based on the denoised image, extract color features, shape features, and texture features;

[0066] The construction of the feed-excrement differentiation model in step D12 specifically includes:

[0067] Based on the extracted image features, construct a feature set for feed samples and a feature set for excrement samples;

[0068] A discrimination model is trained and generated based on the feed sample feature set and the excrement sample feature set using a machine learning algorithm. The machine learning algorithm includes at least one of support vector machine, random forest and convolutional neural network.

[0069] Since feed and excrement are quite similar, visual recognition by camera recognition device 2 is prone to distortion and errors. This invention is more accurate in calculating the remaining area of ​​feed, and can better and more accurately identify and calculate the remaining area of ​​feed, thereby ensuring better and more accurate feed delivery.

[0070] Working principle:

[0071] The control center 3 controls the lifting device 1 to lift the feeding basket located in the water upwards. The camera recognition device 2 visually identifies the remaining feed in the feeding basket lifted by the lifting device 1 and / or the growth status of the fish and shrimp in the feeding basket. The lifting device 1 then lowers the feeding basket back down after the visual identification is completed. The automated inspection system for aquaculture of the present invention can replace manual methods of lifting and lowering the feeding basket located in the water, and can replace manual observation. It can realize automated inspection and is better suited for aquaculture.

[0072] Of course, the above description is only a preferred embodiment of the present invention. Therefore, all equivalent changes or modifications made to the structure, features and principles described in the claims of this patent application are included in the scope of this patent application.

Claims

1. An automated inspection system for aquaculture, characterized in that: The system includes a control center (3), a lifting device (1) for lifting and returning the feeding basket, and a camera recognition device (2) for visually recognizing the feeding basket lifted by the lifting device (1). The lifting device (1) includes a guide rail (11), a base (12) that slides along the guide rail (11), a traveling power mechanism (13) for driving the base (12) to slide, a lifting platform (14) that slides vertically below the base (12), a lifting power mechanism for driving the lifting platform (14) to slide, and a lifting claw (16) on the lifting platform (14). The lifting claw (16) includes a connecting part (161) for connecting with the lifting platform (14), and the front and rear ends of the connecting part (161) are respectively formed with bent parts (162) extending downwards. The bottom of the two bent parts (162) are respectively formed with lifting parts (163) extending horizontally in the direction close to the middle position of the connecting part (161). The lifting power mechanism includes a lifting part for connecting with the lifting platform (14). The lifting platform (14) is connected to a lifting rope (151) and a rope power mechanism (152) for winding or unwinding the lifting rope (151). The rope power mechanism (152) is mounted on the base (12). Telescopic hinge assemblies (153) are respectively provided between the front end and / or rear end and / or both sides of the lifting platform (14) and the base (12). The guide rail (11) is a rod-shaped structure. The traveling power mechanism (13) includes several traveling power modules (17). The traveling power module (17) includes two tapered rollers (171) symmetrically arranged at the front and rear ends of the guide rail (11) and a roller power mechanism (172) for driving the two tapered rollers (171) to rotate. The two tapered rollers (171) roll on the front and rear ends of the guide rail (11) respectively. The output end of the roller power mechanism (172) drives and connects the two tapered rollers (171). The base (12) moves back and forth along the guide rail (11) in a straight line, a circle or an waist shape.

2. The automated inspection system for aquaculture according to claim 1, characterized in that: It also includes a feeding device (4) for adding feed and / or trace elements and / or medicines to the feeding basket.

3. The automated inspection system for aquaculture according to claim 1, characterized in that: The roller power mechanism (172) includes two power output gears (1721) that rotate coaxially with the two tapered rollers (171), two power input gears (1722), and a gear power mechanism (1723) for driving the two power input gears (1722) to rotate. The power output gears (1721) and the power input gears (1722) located on the same side are connected by a transmission chain (1724). The power output gears (1721) are connected to the power input gears (1722) located on the same side through the corresponding transmission chain (1724). The gear power mechanism (1723) is mounted on the base (12). The output end of the gear power mechanism (1723) drives and connects the two power input gears (1722).

Citation Information

Patent Citations

  • Breeding material platform

    CN115669587A

  • Aquaculture inspection device

    CN119196494A