A method for monitoring an environment of a marine ranch

Abstract

The present application relates to the field of marine ranching environment monitoring, and more particularly to a marine ranching environment monitoring method. Collecting a marine ranching water surface image, obtaining original pixel values of each band; performing image preprocessing on the marine ranching water surface image to obtain normalized pixel values of each band; introducing an algae spectral index calculation algorithm based on the normalized pixel values of each band to generate an algae spectral index; based on the algae spectral index, marking algae regions and non-algae regions and generating a binary image; based on the binary image, calculating the algae coverage; based on the normalized pixel values of each band and the binary image, calculating the water turbidity of the non-algae region through a turbidity spectral index algorithm; comparing the algae coverage and the water turbidity with the algae coverage threshold and the water turbidity threshold respectively to generate an early warning signal. The problems of being difficult to accurately segment the algae region and being difficult to accurately reflect the water quality change of the non-algae region are solved.

Description

Technical Field

[0001] This invention relates to the field of marine ranching environmental monitoring, and more particularly to a method for marine ranching environmental monitoring. Background Technology

[0002] Marine ranching, through artificial intervention, constructs ecosystems within specific sea areas to cultivate and enhance marine biological resources, while simultaneously achieving the sustainable use of fishery resources and the protection of the marine ecological environment. However, the operation and management of marine ranches heavily rely on precise monitoring of the marine environment. Traditional marine ranching environmental monitoring primarily depends on manual sampling and laboratory analysis, supplemented by simple physical measurement equipment. Manual sampling is time-consuming and labor-intensive, with a low monitoring frequency, making real-time monitoring difficult. Furthermore, traditional equipment has limited coverage and cannot comprehensively reflect the environmental dynamics of marine ranches. In addition, its data processing and analysis cycle is relatively long, making it difficult to meet the needs of rapid decision-making. In recent years, with the expansion of marine ranching scale and the increasing demands for ecological management, traditional marine ranching environmental monitoring methods have gradually become unable to meet the needs of efficient management in modern marine ranches. Especially in the face of sudden environmental events (such as red tides and oil spills), the lag in traditional marine ranching environmental monitoring methods may lead to missed opportunities for optimal response. Therefore, developing efficient, real-time, and automated marine ranching environmental monitoring technologies has become a research hotspot.

[0003] Marine ranching environmental monitoring technology is a crucial link in ensuring the sustainable development of marine ranches. From traditional sampling and analysis to modern intelligent monitoring, technological advancements have significantly improved monitoring efficiency and accuracy. In the future, with the further integration of technologies such as sensors, the Internet of Things, and artificial intelligence, marine ranching environmental monitoring will achieve a higher level of automation and intelligence, providing stronger support for marine ecological protection and fishery resource management. Summary of the Invention

[0004] This invention provides a method for monitoring the marine ranch environment to address the problem that in marine ranch water surface images, the slight difference in spectral characteristics between algae and water makes it difficult to accurately segment algae areas; and the problem that traditional turbidity measurement methods have low sensitivity to scattering differences in low turbidity areas, making it difficult to accurately reflect water quality changes in non-algae areas.

[0005] A marine ranching environment monitoring method of the present invention includes the following:

[0006] Images of the marine ranch water surface were acquired, and the raw pixel values ​​of each band were obtained. Image preprocessing was performed on the marine ranch water surface images to obtain normalized pixel values ​​for each band. Based on the normalized pixel values ​​of each band, an algae spectral index calculation algorithm was introduced to generate the algae spectral index. Based on the algae spectral index, algae-rich and non-algae-rich regions were marked, and a binarized image was generated. Based on the binarized image, the algae coverage rate was calculated. Based on the normalized pixel values ​​of each band and the binarized image, the turbidity of the water in non-algae-rich regions was calculated using a turbidity spectral index algorithm. The algae coverage rate and water turbidity were compared with algae coverage rate thresholds and water turbidity thresholds, respectively, to generate an early warning signal.

[0007] Preferably, images of the marine ranch water surface are acquired to obtain the raw pixel values ​​of the four bands: red, green, blue, and near-infrared. The marine ranch water surface images are then preprocessed to obtain the normalized pixel values ​​of the four bands: red, green, blue, and near-infrared.

[0008] Preferably, in the implementation of the algae spectral index calculation algorithm, an exponential nonlinear term for the red and green bands is introduced, and the sum of the normalized pixel values ​​of the red band and the normalized pixel values ​​of the green band is processed by an exponential decay function.

[0009] Preferably, in the implementation of the algae spectral index calculation algorithm, the molecular part of the algae spectral index is obtained by subtracting the exponential nonlinear term of the red and green bands from the normalized pixel values ​​of the near-infrared band.

[0010] Preferably, in the implementation of the algae spectral index calculation algorithm, the molecular part of the algae spectral index is normalized by the normalized pixel values ​​in the near-infrared band and the normalized pixel values ​​in the blue band.

[0011] Preferably, a binary image is generated by comparing the algal spectral index with a preset algal spectral index threshold, and the algal region is marked as 1 and the non-algal region is marked as 0.

[0012] Preferably, in the implementation of the turbidity spectral index algorithm, the logarithmic ratio of the normalized pixel value of the blue band to the normalized pixel value of the green band is calculated, and the average turbidity index is calculated based on the logarithmic ratio.

[0013] Preferably, the water turbidity is obtained by multiplying the average turbidity index by a calibration factor.

[0014] Preferably, the warning signals include: normal, algae warning, turbidity warning and dual warning.

[0015] The beneficial effects of the technical solution of the present invention are:

[0016] 1. By acquiring surface images of marine ranches using multispectral cameras mounted on drones or fixed platforms, and combining image preprocessing with algae spectral index calculation algorithms, precise quantification of algae coverage in marine ranches was achieved. Utilizing the difference between the high reflectivity of the near-infrared band and the absorption characteristics of the red and green bands, an algae spectral index was constructed through exponential nonlinear transformation. This significantly amplified the spectral differences between algae and the water body, enabling the differentiation between algae-rich and non-algae-rich areas. Binary images were generated, and algae coverage was calculated, providing a high-precision, automated solution for monitoring the ecological environment of marine ranches, reducing the cost and error of manual monitoring.

[0017] 2. Based on binarized images and normalized pixel values ​​of each band, a turbidity spectral index algorithm is proposed. By combining the logarithmic ratio of the blue and green bands with nonlinear transformation, the sensitivity to low turbidity regions is enhanced, enabling accurate calculation of turbidity in non-algal areas. The algorithm fully utilizes the differences in scattering characteristics of suspended matter of different particle sizes by the blue and green bands. Combined with calibration coefficients, the reliability and consistency of turbidity measurement are ensured, providing a scientific basis for the assessment of water quality in marine ranches.

[0018] 3. By comparing algae coverage and water turbidity with preset algae coverage and water turbidity thresholds, the system can automatically perform logical judgments, generate intuitive early warning signals, and display them through a visual interface. This significantly improves the response speed of marine ranch managers to ecological and water quality anomalies and effectively reduces environmental risks.

[0019] 4. By comprehensively monitoring algae coverage and water turbidity, potential problems in the marine ranch ecosystem can be identified in a timely manner, providing data support for scientific management and ecological protection, which helps maintain the ecological balance of the marine ranch and ensures the sustainable development of aquaculture. Attached Figure Description

[0020] Figure 1 This is a flowchart of a marine ranch environmental monitoring method according to the present invention. Detailed Implementation

[0021] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] The following description, in conjunction with the accompanying drawings, details a specific scheme for a marine ranch environmental monitoring method provided by the present invention.

[0024] See attached document Figure 1 The diagram illustrates a flowchart of a marine ranching environmental monitoring method according to an embodiment of the present invention, which includes the following steps:

[0025] S1. Acquire images of the marine ranch water surface and obtain the raw pixel values ​​of each band; perform image preprocessing on the marine ranch water surface images to obtain normalized pixel values ​​of each band; based on the normalized pixel values ​​of each band, introduce an algae spectral index calculation algorithm to generate an algae spectral index; based on the algae spectral index, mark algae regions and non-algae regions and generate a binarized image; based on the binarized image, calculate the algae coverage.

[0026] Images of the marine ranch surface are acquired using multispectral cameras mounted on drones or fixed platforms, obtaining raw pixel values ​​in four bands: red (R, wavelength 600-700nm), green (G, wavelength 500-600nm), blue (B, wavelength 400-500nm), and near-infrared (NIR, wavelength 700-900nm).

[0027] To improve the image quality of marine ranch surface images, image preprocessing was performed to obtain normalized pixel values ​​for each band. Image preprocessing included denoising and normalization; denoising employed Gaussian filtering to smooth noise in the marine ranch surface images while preserving spectral features; normalization mapped the original pixel values ​​of each band to their corresponding values ​​using a min-max normalization method. Interval.

[0028] Based on the normalized pixel values ​​of each band, an algal spectral index calculation algorithm is introduced. Based on multispectral imaging technology, the algal spectral index is constructed by using the high reflectivity of the near-infrared band and the difference in absorption characteristics between the red and green bands through exponential nonlinear transformation.

[0029] Because algal cell structures have high reflectivity in the near-infrared band, while the reflectivity of water is close to zero, the normalized pixel value in the near-infrared band reflects the high reflectivity of algae.

[0030] Since the normalized pixel values ​​in the red and green bands reflect the absorption characteristics of algal pigments, the higher the absorption intensity, the lower the normalized pixel value. Therefore, the exponential nonlinear term in the red and green bands is processed by an exponential decay function to handle the sum of the normalized pixel values ​​in the red and green bands. The exponential decay function maps the sum of the normalized pixel values ​​in the red and green bands to... The range is such that when the sum of the normalized pixel values ​​of the red band and the green band is larger, the exponential nonlinear term of the red and green bands tends to 0, which means that it is more likely to be an algal region. When the sum of the normalized pixel values ​​of the red band and the green band is smaller, the exponential nonlinear term of the red and green bands tends to 1, which means that the spectral differences between algae and water are amplified.

[0031] Subtracting the exponential nonlinear term of the red and green bands from the normalized pixel values ​​in the near-infrared band reflects the combined characteristics of high reflectance and high absorption in algal regions; the larger the value, the more likely it is to be algae.

[0032] Normalized pixel values ​​in the near-infrared band and the blue band are used to normalize the numerator of the algal spectral index, and a minimal constant is introduced to prevent the denominator from being zero.

[0033] The formula for calculating the algal spectral index is:

[0034] ,

[0035] in, Represents pixels The algae spectral index at a location; the higher the value, the greater the likelihood of algae. Indicates the near-infrared band at the pixel level Normalized pixel value at the location; Indicates the red band at the pixel. Normalized pixel value at the location; Indicates the green band at the pixel. Normalized pixel value at the location; The exponential nonlinear term representing the red and green bands is used to simulate the absorption saturation effect of algal pigments through an exponential decay function, amplifying the absorption difference between the red and green bands. The lower the absorption, the closer it is to 1, and the higher the absorption, the closer it is to 0. Indicates the blue band at the pixel level Normalized pixel value at the location; Let represent a very small constant, used to prevent the denominator from being zero, denoted as . .

[0036] The algae spectral index of each pixel is compared with the preset algae spectral index threshold. If the algae spectral index of a pixel is greater than the algae spectral index threshold, the pixel is marked as an algae region; otherwise, it is marked as a non-algae region, i.e., water. The algae spectral index threshold is determined based on the statistical distribution of the spectral index, such as selecting the 90th quantile (0.35).

[0037] A binarized image is generated after comparing it with a preset algae spectral index threshold, where algae regions are marked as 1 and non-algae regions are marked as 0.

[0038] The algae coverage rate is obtained by counting the number of pixels marked as algae out of all pixels and dividing that number by the total number of pixels. The formula is expressed as follows:

[0039] ,

[0040] in, This indicates algae coverage, ranging from 0 to 1, and is used to represent the proportion of area covered by algae. Used to count the number of pixels that meet the conditions; Indicates an indicator function, when The value is 1 when the time is right and 0 otherwise. It is used to classify pixels into algal regions or non-algal regions to achieve binarized segmentation. This represents the algal spectral index threshold, used to distinguish between algal and non-algal regions, with a value range of [value range missing]. ; This represents the total number of pixels.

[0041] S2. Based on the normalized pixel values ​​and binarized images of each band, the turbidity of the water body in the non-algae area is calculated using the turbidity spectral index algorithm; the algae coverage rate and water turbidity are compared with the algae coverage rate threshold and the water turbidity threshold, respectively, to generate an early warning signal.

[0042] Based on the normalized pixel values ​​and binarized images of each band, the turbidity of the water in non-algal areas is calculated using the turbidity spectral index algorithm.

[0043] Because the blue band is more sensitive to scattering of smaller suspended particles, while the green band is more sensitive to scattering of larger suspended particles, the higher the turbidity of the water, the more small particles there are, resulting in blue light scattering intensity exceeding that of green light, leading to an increase in the logarithmic ratio. Conversely, when the turbidity is lower, the scattering intensities of blue and green light are closer, causing the logarithmic ratio to approach zero. The nonlinear nature of the logarithmic transform amplifies the scattering differences, enhancing sensitivity to low-turbidity regions. For each pixel in non-algae regions, the turbidity spectral index algorithm calculates the logarithmic ratio of the normalized pixel value in the blue band to the normalized pixel value in the green band, while adding a very small constant to avoid a zero denominator or an invalid logarithm.

[0044] The logarithmic ratio is further multiplied by the normalized pixel value of the blue band to preserve the dominant contribution of the blue band to the turbidity of the water. The average turbidity index of the non-algal region is obtained by summing all pixels in the non-algal region and taking the average value. The average turbidity index is then multiplied by the calibration coefficient to obtain the turbidity of the water.

[0045] The formula for calculating the turbidity of water in non-algae-rich areas is:

[0046] ,

[0047] in, Indicates the turbidity of water; This represents the calibration coefficient, with a value range of 20-100 NTU. NTU is a unit of turbidity, and the specific value is determined by laboratory turbidity standard samples, such as Formazin standard solution. Represents the number of pixels in the non-algae region. The reciprocal of is used to calculate the average turbidity index; Indicates non-algae areas; This represents the logarithmic ratio of the normalized pixel values ​​in the blue band to the normalized pixel values ​​in the green band. This represents the average turbidity index.

[0048] Determining algae coverage thresholds based on water quality and ecological health standards for marine ranches. and water turbidity threshold The system compares the data with both algae coverage and water turbidity, performs logical checks, and generates an early warning signal. The logical judgment is as follows:

[0049] ,

[0050] Warning signal Output in text format, containing four possible values: normal, algae warning, turbidity warning, and dual warning.

[0051] Early warning signals can be displayed through a visual interface, helping managers to promptly detect ecological and water quality anomalies and ensure the sustainable development of marine ranches.

[0052] In summary, a method for monitoring the marine ranch environment has been developed.

[0053] The order of the embodiments is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0054] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0055] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method of monitoring an environment of a marine ranching, characterized by, The method comprises the following steps: Collecting the image of the sea ranching water surface to obtain original pixel values of each wave band; performing image preprocessing on the image of the sea ranching water surface to obtain normalized pixel values of each wave band; introducing an algae spectral index calculation algorithm based on the normalized pixel values of each wave band to generate an algae spectral index, and the calculation formula is: ， wherein, represents the algal spectral index at pixel point ; represents the normalized pixel value of the near-infrared waveband at pixel point ; represents the normalized pixel value of the red waveband at pixel point ; represents the normalized pixel value of the green waveband at pixel point ; represents the normalized pixel value of the blue waveband at pixel point ; represents a constant; Based on the algae spectral index, marking the algae area and non-algae area, and generating a binary image; based on the binary image, calculating the algae coverage; based on the normalized pixel values of each wave band and the binary image, calculating the water turbidity of the non-algae area through a turbidity spectral index algorithm, and the calculation formula is: ， wherein, represents water turbidity; represents a calibration coefficient; represents the number of pixel points of a non-algal region; represents a non-algal region; represents an average turbidity index; Comparing the algae coverage and the water turbidity with the algae coverage threshold and the water turbidity threshold respectively to generate an early warning signal.

2. The method of monitoring an environment of a marine ranching according to claim 1, wherein, Collecting the image of the sea ranching water surface to obtain original pixel values of red, green, blue and near-infrared four wave bands; performing image preprocessing on the image of the sea ranching water surface to obtain normalized pixel values of red, green, blue and near-infrared four wave bands.

3. The method of monitoring an environment of a marine ranch according to claim 2, wherein, In the implementation process of the algae spectral index calculation algorithm, an exponential nonlinear term of the red-green wave band is introduced, and the sum of the normalized pixel values of the red wave band and the normalized pixel values of the green wave band is processed through an exponential decay function.

4. The method of monitoring an environment of a marine ranch according to claim 3, wherein, In the implementation process of the algae spectral index calculation algorithm, the normalized pixel value of the near-infrared wave band is subtracted from the exponential nonlinear term of the red-green wave band to obtain the numerator part of the algae spectral index.

5. The method of monitoring an environment of a marine ranch according to claim 4, wherein, In the implementation process of the algae spectral index calculation algorithm, the numerator part of the algae spectral index is normalized by the normalized pixel value of the near-infrared wave band and the normalized pixel value of the blue wave band.

6. The mariculture environment monitoring method of claim 1, wherein, After comparing the algae spectral index with the preset algae spectral index threshold, a binary image is generated, and the algae area is marked as 1 and the non-algae area is marked as 0.

7. The mariculture environment monitoring method according to claim 2, characterized in that, In the implementation process of the turbidity spectral index algorithm, the logarithmic ratio of the normalized pixel value of the blue wave band and the normalized pixel value of the green wave band is calculated, and based on the logarithmic ratio, the average turbidity index is calculated.

8. The mariculture environment monitoring method according to claim 7, characterized in that, The average turbidity index is multiplied by a calibration coefficient to obtain the water turbidity.

9. The mariculture environment monitoring method of claim 1, wherein, The early warning signal includes: normal, algae early warning, turbidity early warning and double early warning.

Images