Deep sea polymetallic nodule resource evaluation system based on AUV platform and operation method

Abstract

The invention relates to a deep sea polymetallic nodule resource evaluation system based on an AUV platform and an operation method, and belongs to the technical field of deep sea mineral resource exploration. The low-power-consumption processor comprises an image preprocessing module, a self-adaptive threshold segmentation module, a morphological processing module and a physical size estimation module; the low-power-consumption processor performs related processing on the input picture to evaluate the size and coverage rate of the tuberculosis; according to the system, the algorithm is optimized, and the performance requirement on an online processor is reduced. A common low-power-consumption processor can be adopted, the normal working power consumption of the processor is within 300mW, meanwhile, equipment such as the high-definition camera, the flash lamp and the acoustic communication module is in a power-off zero-power-consumption state in most of time, and is turned on to work for a short time only when needed, so that the working power consumption of the whole system is greatly reduced, and the working efficiency is improved. And the working time is effectively prolonged.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of deep-sea mineral resource exploration, and particularly relates to a deep-sea polymetallic nodule resource evaluation system based on an AUV platform and an operation method. BACKGROUND

[0002] In the context of the gradual shortage of land resources, deep-sea polymetallic nodules become a new focus of global resource competition because they are rich in key metals such as manganese, nickel and cobalt. Carrying out polymetallic nodule resource investigation can not only fill the blank of deep-sea resource cognition and ensure national resource security, but also lay the foundation for future development and is of great significance to the exploration of deep-sea geological and ecological evolution.

[0003] Among many investigation methods, using an autonomous underwater vehicle (AUV) platform to carry a camera / camera device to carry out optical investigation of deep-sea polymetallic nodule mining areas, collecting optical data of the deep-sea seabed, and extracting information such as deep-sea seabed nodule coverage, abundance and bottom type is an important technical means for current deep-sea polymetallic nodule resource evaluation.

[0004] Currently, optical investigation generally adopts a self-contained storage mode, and the generated video or photo is usually stored in a local device, and the data is read and analyzed after the AUV is recovered. The existing optical system technology method of the AUV platform cannot realize interaction with a ship or a shore base due to the large amount of optical data, has poor real-time performance and low efficiency; the system lacks a low-power consumption strategy, has high overall power consumption and short working time; and because the quality of images cannot be obtained in time, the exposure and focusing may be wrong due to changes in the seabed environment, resulting in low-quality images, so the failure risk is high. SUMMARY

[0005] The application proposes a deep-sea polymetallic nodule resource optical evaluation system based on an AUV platform to solve the above technical problems. The system includes a low-power consumption processor, a high-definition camera, a flash, an altimeter and a power supply control unit, and the parts of the system work cooperatively to realize real-time estimation and return of the coverage of polymetallic compounds on the basis of maximum low-power consumption operation.

[0006] The application is implemented by the following technical solutions:

[0007] A deep-sea polymetallic nodule resource evaluation system based on an AUV platform, the system including a low-power consumption processor; the low-power consumption processor including an image preprocessing module, an adaptive threshold segmentation module, a morphological processing module and a physical size estimation module; the low-power consumption processor performing relevant processing on input photos to evaluate nodule size and coverage;

[0008] The image preprocessing module operates as follows: the shooting direction is automatically corrected by parsing EXIF ​​metadata of the acquired image, and the original RGB image is converted into a single-channel grayscale image by weighted grayscale conversion;

[0009] The adaptive threshold segmentation module uses the Otsu optimization algorithm to determine the optimal segmentation threshold.

[0010]

[0011] Where ω0,ω1 are the pixel percentages of each class, and μ0,μ1 are the class mean values. The variance is between classes.

[0012] The morphological processing module uses elliptical structural elements for morphological operations, which are performed in two steps:

[0013] The first round of erosion calculations eliminates noise points:

[0014] The second round of expansion restores the target to its original size: I clean =I open ·B 5×5 ;

[0015] Among them, I thresh The binary image after thresholding, I open ,I clean Image after opening and closing operations

[0016] •: Morphological opening and closing operator, B 3×3 B 5×5 Elliptic operation structuring element;

[0017] The physical size estimation module derives the size formula based on the pinhole camera model.

[0018]

[0019] Among them, W sensor =36mm is the sensor width, u is the object distance, and f is the focal length.

[0020] Furthermore, the low-power processor also includes a result output module, which outputs the results after the image data processing is completed.

[0021] Furthermore, the evaluation system also includes a power supply control unit, which can control the power supply to or off of various system components.

[0022] Furthermore, the evaluation system also includes a high-definition camera, a flash, and an altimeter; the altimeter and the low-power processor communicate bidirectionally via a UART interface, the flash and the low-power processor interact via input and output interfaces, and the high-definition camera and the low-power processor are connected via a USB interface.

[0023] This invention also provides a method for operating the evaluation system. A high-definition camera completes the shooting with the help of a flash. After the shooting is completed, a low-power processor acquires the image from the high-definition camera via a USB interface. The low-power processor establishes a pixel-physical mapping model by fusing the camera's optical parameters, performs a referenceless estimation of the size of the polymetallic nodules, and then obtains the number and size of all nodules in the image, thus obtaining the assessment result of the conjugate coverage at the corresponding location. While the low-power processor is processing the image, the power supply control unit controls the other units in the evaluation system to power off. When the low-power processor finishes processing and outputs the result, the power supply control unit wakes up the other units in the evaluation system and sends the processing result of the low-power processor to the mother ship. After obtaining the result data, the mother ship determines whether it is necessary to issue a parameter adjustment command by judging the image quality and the reasonableness of the result. After receiving the command, the low-power processor saves the shooting parameters and then powers off and goes into hibernation to wait for the next round of work tasks.

[0024] Furthermore, the altimeter is used to obtain the actual distance between the camera and the seabed when the AUV is navigating close to the bottom, assisting the high-definition camera in focusing.

[0025] The advantages of this invention compared to existing optical evaluation system survey methods:

[0026] 1. The system of this invention can acquire the size and coverage information of multimetallic compounds at the current location in real time during AUV navigation, which greatly improves the efficiency of seabed exploration and investigation.

[0027] 2. The system of this invention realizes real-time closed-loop control of relevant parameters during the exploration and survey process. Through feedback, the shooting parameters can be adjusted in real time, which improves image clarity and exploration accuracy, and greatly reduces the risk of AUV optical operation failure.

[0028] 3. This system has optimized its algorithm, reducing the performance requirements of the online processor. Therefore, a common low-power processor can be used, with a normal operating power consumption of less than 300mW. At the same time, devices such as high-definition cameras and flashlights are in a zero-power state most of the time, only being turned on for short periods when needed. This greatly reduces the overall system power consumption and effectively extends the operating time. Attached Figure Description

[0029] Figure 1This is a schematic diagram of the system structure of the present invention;

[0030] Figure 2 This is an algorithm diagram of the low-power processor of the present invention. Detailed Implementation

[0031] The technical solution of the present invention will be further explained below through embodiments, but the scope of protection of the present invention is not limited in any way by the embodiments.

[0032] Example 1

[0033] A deep-sea polymetallic nodule resource assessment system based on an AUV platform, such as Figure 1 As shown, the system includes a low-power processor; the low-power processor includes an image preprocessing module, an adaptive threshold segmentation module, a morphological processing module, and a physical size estimation module; the low-power processor performs correlation processing on the input image to evaluate the nodule size and coverage. The low-power processor is responsible for the task execution of the entire system and the coordination of the various components. The high-definition camera, combined with a flash, is used to acquire high-definition images of the polymetallic nodules on the seabed; the altimeter is used for assisting focusing during the shooting process; and the power supply control unit is used to coordinate the operation and sleep control of various components in the system, achieving maximum low power consumption while completing necessary tasks.

[0034] As a preferred embodiment, such as Figure 2 As shown, the image preprocessing module operates as follows: the shooting direction is automatically corrected by parsing EXIF ​​metadata of the acquired image; the original RGB image is converted into a single-channel grayscale image by weighted grayscale conversion to reduce the complexity of subsequent calculations.

[0035] Non-linear grayscale conversion: G = 0.299R 2.2 +0.587G 2.2 +0.114B 2.2 ,

[0036] Gaussian filtering:

[0037] The adaptive threshold segmentation module uses the Otsu optimization algorithm to determine the optimal segmentation threshold.

[0038]

[0039] Where ω0,ω1 are the pixel percentages of each class, and μ0,μ1 are the class mean values. The variance is between classes.

[0040] The morphological processing module mentioned above:

[0041] Morphological operations are performed using elliptical structural elements, and the operation consists of two steps:

[0042] The first round of erosion operation eliminates noise points (opening operation eliminates noise):

[0043] The second round of expansion restores the target to its original size (closing operation fills the voids): I clean =I open ·B 5×5 ;

[0044] Among them, I thresh Binary image after thresholding;

[0045] I open ,I clean Image after opening and closing operations (final result after noise reduction and hole filling);

[0046] •: Morphological opening and closing operators (erosion followed by dilation or dilation followed by erosion);

[0047] B 3×3 B 5×5 Elliptic operation struct element.

[0048] The physical size estimation model is based on a pinhole camera model, from which the size formula is derived.

[0049]

[0050] Among them W sensor =36mm is the sensor width, u is the object distance, and f is the focal length.

[0051] As one specific implementation, the low-power processor also includes a result output module, which outputs the result after the image data processing is completed.

[0052] In one specific implementation, the evaluation system also includes a power supply control unit, which can control the power supply to or off of various system components. In system standby mode, all modules are in a power-off, zero-power state, with only the power supply control unit in a sleep state with a power consumption of μA. When the task arrives, the power supply control unit first wakes up the low-power processor. After the low-power processor operates normally, it enables the altimeter, obtains the shooting distance information from the altimeter, and then wakes up the high-definition camera and flash.

[0053] As one specific implementation, the evaluation system also includes a high-definition camera, a flash, and an altimeter. The altimeter and the low-power processor communicate bidirectionally via a UART interface. The flash and the low-power processor interact via input and output interfaces. The high-definition camera and the low-power processor are connected via a USB interface.

[0054] The shooting process of the high-definition camera is entirely controlled by a low-power processor. The shooting mode is manual (M) mode. The focus distance is given based on the altimeter's measurement results. The camera exposure parameters and flash exposure settings are adjusted accordingly based on the distance measured by the altimeter.

[0055] After shooting is completed, the low-power processor reads the image data through the USB interface, and then the power supply control unit cuts off the power to the camera and flash to ensure that the system operates with maximum low power consumption.

[0056] The operation method of the evaluation system is as follows: an altimeter measures the height of the AUV above the seabed to be photographed and feeds the parameters back to the low-power processor. Based on the distance measured by the altimeter, the camera exposure parameters and flash exposure are set. The high-definition camera completes the shooting with the help of the flash. After the shooting is completed, the low-power processor acquires the image from the high-definition camera through the USB interface. The low-power processor establishes a pixel-physical mapping model by fusing the camera's optical parameters and performs a referenceless estimation of the size of the polymetallic nodules. This allows the acquisition of the number and size of all nodules in the image, resulting in an assessment result of the nodule coverage at the corresponding location. While the low-power processor is processing the image, the power supply control unit controls the other units in the evaluation system to shut down. When the low-power processor finishes processing and outputs the results, the power supply control unit wakes up the other units in the evaluation system, such as the acoustic communication module, and sends the processing results of the low-power processor to the mother ship through the acoustic communication module. After receiving the result data, the mother ship determines whether it is necessary to issue a parameter adjustment command by judging the image quality and the reasonableness of the results. After receiving the command, the low-power processor saves the shooting parameters and then shuts down and goes into hibernation to await the next round of work. The mothership is the upper level of the AUV platform, used to deploy and recover the AUV platform, and can send control commands to the AUV in real time. It can also send direct commands to the imaging system through the AUV platform.

[0057] Example 2

[0058] The research vessel, commissioned by a company, is conducting a precise resource assessment of a polymetallic nodule-rich area in the Pacific Ocean. The survey area covers approximately 500 square kilometers at a depth of 4000-4500 meters, requiring high-precision analysis of nodule coverage, nodule size, and resource quantity.

[0059] The research vessel arrived at the pre-designated survey area and precisely anchored using a dynamic positioning system. Based on multibeam sonar data, the 500 square kilometer survey area was divided into 25 sub-areas of 20 km² each. The initial survey path was designed using a zigzag scanning mode, with the AUV operating at a depth of 5-8 meters above the seabed. The preset survey speed was 1.5 knots to ensure image quality and coverage. Self-tests of all system modules were completed: high-definition camera, flash, altimeter, and low-power processor. The power supply control unit was confirmed to be functioning normally, and the acoustic communication system signal was good.

[0060] After the AUV carrier inspection was completed, the diving operation began. The AUV was hoisted into the water from the stern of the research vessel and began its autonomous descent, maintaining continuous acoustic communication with the mother ship throughout the descent. The survey work commenced at 10:00 AM and returned at 5:00 PM.

[0061] Survey results: Completed survey area: 120 square kilometers (24% of the total area); Total number of images taken: 2,847; Total number of tuberculosis tubers identified: 342,156; Average coverage rate: 22.8%; High-value area (coverage rate >25%): 38 square kilometers;

[0062] This survey fully validated the technical advantages of the assessment system: the altitude adaptive adjustment function performed excellently in complex terrain; the real-time parameter optimization of the low-power processor ensured high-quality imaging throughout the process; the precise power-off control of the power supply control unit extended the AUV's endurance by 35%, and the system power consumption was controlled to less than 60% of the traditional solution; real-time decision-making: real-time data feedback from acoustic communication enabled the mother ship to dynamically adjust the survey strategy, improving survey efficiency by 60% compared to traditional offline analysis methods, achieving a referenceless estimation accuracy of ±2% for the pixel-physical mapping model, and a nodule identification accuracy of 96.7%.

Claims

1. A deep-sea polymetallic nodule resource assessment system based on an AUV platform, characterized in that, The evaluation system includes a low-power processor; the low-power processor includes an image preprocessing module, an adaptive threshold segmentation module, a morphological processing module, and a physical size estimation module; the low-power processor performs relevant processing on the input photographs to evaluate nodule size and coverage; The image preprocessing module operates as follows: the shooting direction is automatically corrected by parsing EXIF ​​metadata of the acquired image, and the original RGB image is converted into a single-channel grayscale image by weighted grayscale conversion; The adaptive threshold segmentation module uses the Otsu optimization algorithm to determine the optimal segmentation threshold. Where ω0,ω1 are the pixel percentages of each class, and μ0,μ1 are the class mean values. For inter-class variance; The morphological processing module uses elliptical structural elements for morphological operations, which are performed in two steps: the first round of erosion operation eliminates noise points. The second round of expansion restores the target to its original size: I clean =I open ·B 5×5 ; Among them, I thresh For the binary image after thresholding, I open ,I clean The image after opening and closing operations • For morphological opening and closing operators, B 3×3 B 5×5 It is a struct element for elliptic operations; The physical size estimation module derives the size formula based on the pinhole camera model. Among them, W sensor =36mm is the sensor width, u is the object distance, and f is the focal length.

2. The deep-sea polymetallic nodule resource assessment system based on an AUV platform according to claim 1, characterized in that, The low-power processor also includes a result output module, which outputs the results after the image data processing is completed.

3. The deep-sea polymetallic nodule resource assessment system based on an AUV platform according to claim 1, characterized in that, The evaluation system also includes a power supply control unit, which can control the power supply to or off of various system components.

4. The deep-sea polymetallic nodule resource assessment system based on an AUV platform according to claim 1, characterized in that, The evaluation system also includes a high-definition camera, a flash, and an altimeter. The altimeter and the low-power processor communicate bidirectionally via a UART interface. The flash and the low-power processor interact via input and output interfaces. The high-definition camera and the low-power processor are connected via a USB interface.

5. The method of operating the evaluation system according to any one of claims 1-4, characterized in that, The high-definition camera, in conjunction with a flash, takes the picture. After the picture is taken, the low-power processor acquires the image from the high-definition camera via a USB interface. The low-power processor then establishes a pixel-physical mapping model by fusing the camera's optical parameters to perform a referenceless estimation of the size of the polymetallic nodules. This allows it to obtain the number and size of all nodules in the image and derive the assessment result of the conjugate coverage at the corresponding location. While the low-power processor is processing the image, the power supply control unit controls the other units in the assessment system to shut down. When the low-power processor finishes processing and outputs the results, the power supply control unit wakes up the other units in the assessment system and sends the processing results of the low-power processor to the mother ship. After receiving the result data, the mother ship determines whether it is necessary to issue a parameter adjustment command by judging the image quality and the reasonableness of the results. After receiving the command, the low-power processor saves the shooting parameters and then shuts down and goes into hibernation to await the next round of work.

6. The method according to claim 5, characterized in that, The altimeter is used to obtain the actual distance between the high-definition camera and the seabed when the AUV is sailing close to the bottom, to assist the high-definition camera in focusing.

Images