Sea ice and seawater classification method, system, storage medium and electronic device

Abstract

The present invention discloses a sea ice and seawater classification method, system, storage medium and electronic device. By extracting the grayscale histogram distribution and image perceptual hash of a DDM image as the features of the DDM image, and then performing similarity calculations with the DDM image stored in the system database, the method achieves the effect of distinguishing sea water and sea ice, effectively improving the speed of calculation while reducing the consumption of computing resources. The present invention can improve the accuracy of data prediction while reducing the dependence on the amount of data samples; the present invention has low requirements for computing power and can achieve the effect of identifying sea ice and seawater in DDM images without excessive computing resources; the algorithm complexity of the present invention is low and the algorithm is highly interpretable. The grayscale histogram distribution and perceptual hash value of the image can be used as binary representations on a computer to effectively represent the information of the image. By calculating the similarity with the reference DDM image stored in the system database, the method can achieve the effect of distinguishing sea ice and seawater.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more particularly to a sea ice and seawater classification method, system, storage medium and electronic device. Background Art

[0002] Changes in sea ice are closely linked to climate change in the atmospheric environment. Sea ice formation is a phenomenon unique to high-latitude oceans and the polar regions and is a crucial component of atmospheric exchange. The formation, growth, and melting of sea ice reflect the heat exchange process between the ocean and the atmosphere. The presence of sea ice affects the hydrological characteristics, ocean current patterns, atmospheric circulation, and heat exchange in high-latitude regions. It is also one of the most prominent marine hazards. Therefore, research on sea ice detection is of great significance. Global Navigation Satellite System Reflectometry (GNSS-R) is a new microwave remote sensing technology. As a combination of active and passive remote sensing, it offers the advantages of high temporal and spatial resolution and low cost. Delay-Doppler Map (DDM) data is generated by cross-correlating received GNSS-R signals with local replicas. DDM data is affected by the roughness of the reflecting surface. Sea ice surfaces are generally smoother than seawater surfaces, providing a basis for distinguishing sea ice from seawater using DDM data.

[0003] Current methods for classifying sea ice and seawater based on spaceborne GNSS-R Doppler time-lapse imagery primarily rely on deep learning methods, such as convolutional neural networks and deep neural networks. Deep learning is widely considered an excellent technology and a new trend in remote sensing image processing. It can completely learn feature representations from data and has strong learning capabilities.

[0004] However, compared with traditional methods, deep learning has the following disadvantages: First, deep learning methods are highly dependent on data, and when the training data samples are unbalanced, it will lead to deviations in data prediction; second, it requires a large amount of computing resources, and deep learning has high requirements for computing power and high hardware configuration costs; third, due to the complexity and black box nature of deep learning algorithms, its interpretability is relatively weak, and it is difficult to understand how the algorithm works internally. Summary of the Invention

[0005] In order to solve at least one of the problems described in the above background technology, the present invention provides a sea ice and seawater classification method, system, storage medium and electronic device.

[0006] According to one aspect of the present invention, a method for classifying sea ice and seawater is provided, comprising:

[0007] Acquire target delay-Doppler images of the L1 level product of the GNSS reflectometry technology;

[0008] Determine the grayscale histogram distribution vector diagram of the target delay Doppler image, and calculate the image perception hash value of the target delay Doppler image;

[0009] Calculating the similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in the preset database, and calculating the Hamming distance between the image-aware hash value of the target delay Doppler image and the image-aware hash value of the reference delay Doppler image stored in the preset database;

[0010] Determining, from a preset database based on the similarity calculation result, a plurality of reference Delay Doppler images having a similarity with the target Delay Doppler image greater than a first threshold, and determining, from the preset database based on the Hamming distance calculation result, a plurality of reference Delay Doppler images having a Hamming distance with the target Delay Doppler image greater than a second threshold, to obtain a reference Delay Doppler image set corresponding to the target Delay Doppler image;

[0011] The reference delay Doppler images in the reference delay Doppler image set are classified, and the target type corresponding to the reference delay Doppler image with the largest number is determined according to the classification result, and the target type is determined as the classification result of the target delay Doppler image for distinguishing sea ice and seawater.

[0012] Optionally, acquiring a target delay Doppler image of an L1 product of a global navigation satellite system reflection technology includes:

[0013] Collect initial time-delay Doppler images of the L1 level products of the GNSS reflectometry technology over a period of time;

[0014] Performing quality control on the initial delay Doppler image based on the quality flag provided by the global navigation satellite system reflection technology, removing abnormal images from the initial delay Doppler image, and performing noise reduction processing on the initial delay Doppler image;

[0015] According to the user's selection operation, a corresponding delay Doppler image is selected from the initial delay Doppler images after noise reduction processing as the target delay Doppler image.

[0016] Optionally, calculating the image-aware hash value of the target delay-Doppler image includes:

[0017] Calculate the difference hash value of the target delay-Doppler image;

[0018] Calculate the preliminary perceptual hash value of the target delay-Doppler image;

[0019] An image-perceptual hash value of the target delay-Doppler image is determined based on the difference hash value and the preliminary perceptual hash value.

[0020] Optionally, calculating the similarity between the grayscale histogram distribution vector of the target Delay Doppler image and a reference Delay Doppler image stored in a preset database includes:

[0021] The similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in the preset database is calculated using the following formula:

[0022]

[0023] Where, P a Represents the grayscale histogram distribution vector of the target delay-Doppler image; represents the average value of the grayscale histogram distribution vector of the target delay Doppler image; P i Represents the grayscale histogram distribution vector of the i-th reference delay-Doppler image stored in the preset database; represents the average value of the grayscale histogram distribution vector of the i-th reference delay-Doppler image stored in the preset database.

[0024] Optionally, the method further includes creating a reference delay Doppler image in a preset database by:

[0025] Collect historical time-delay Doppler images of the L1 level products of the Global Navigation Satellite System reflectometry technology and OSISAF sea ice data, including sea ice concentration and sea ice type data;

[0026] Based on the quality flags provided by GNSS reflectometry, quality control of historical time-delay Doppler images and OSISAF sea ice data was performed to remove outliers.

[0027] Perform spatiotemporal matching of historical time-delay Doppler images with OSISAF sea ice data, and select the time-delay Doppler image of seawater, the time-delay Doppler image of one-year ice, and the time-delay Doppler image of multi-year ice as the matching result;

[0028] The time delay Doppler image of seawater is stored in a preset database, and the time delay Doppler image of one-year ice and the time delay Doppler image of multi-year ice are stored in the preset database according to a preset ratio to obtain a reference time delay Doppler image in the preset database.

[0029] According to another aspect of the present invention, a sea ice and seawater classification system is provided, comprising:

[0030] Image acquisition module, used to obtain target delay Doppler images of L1 level products of global navigation satellite system reflection technology;

[0031] A first calculation module is used to determine a grayscale histogram of a target delay-Doppler image and calculate an image-perceptual hash value of the target delay-Doppler image;

[0032] A second calculation module is used to calculate the similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in the preset database, and calculate the Hamming distance between the image perception hash value of the target delay Doppler image and the image perception hash value of the reference delay Doppler image stored in the preset database;

[0033] a first determining module, configured to determine, from a preset database based on a similarity calculation result, a plurality of reference Delay Doppler images having a similarity with the target Delay Doppler image greater than a first threshold, and, based on a Hamming distance calculation result, determine, from the preset database, a plurality of reference Delay Doppler images having a Hamming distance with the target Delay Doppler image greater than a second threshold, to obtain a reference Delay Doppler image set corresponding to the target Delay Doppler image;

[0034] The second determination module is used to classify the reference delay Doppler images in the reference delay Doppler image set, determine the target type corresponding to the reference delay Doppler image with the largest number according to the classification results, and determine the target type as the classification result of the target delay Doppler image for distinguishing sea ice and seawater.

[0035] Optionally, the image acquisition module is specifically configured to:

[0036] Collect initial time-delay Doppler images of the L1 level products of the GNSS reflectometry technology over a period of time;

[0037] Performing quality control on the initial delay Doppler image based on the quality flag provided by the global navigation satellite system reflection technology, removing abnormal images from the initial delay Doppler image, and performing noise reduction processing on the initial delay Doppler image;

[0038] According to the user's selection operation, a corresponding delay Doppler image is selected from the initial delay Doppler images after noise reduction processing as the target delay Doppler image.

[0039] Optionally, the first calculation module is specifically configured to:

[0040] Calculate the difference hash value of the target delay-Doppler image;

[0041] Calculate the preliminary perceptual hash value of the target delay-Doppler image;

[0042] An image-perceptual hash value of the target delay-Doppler image is determined based on the difference hash value and the preliminary perceptual hash value.

[0043] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and the computer program is used to execute the method according to any one of the above aspects of the present invention.

[0044] According to another aspect of the present invention, an electronic device is provided, comprising: a processor; a memory for storing instructions executable by the processor; and the processor for reading the executable instructions from the memory and executing the instructions to implement the method described in any one of the above aspects of the present invention.

[0045] The present invention extracts input DDM image features based on an image grayscale histogram and an image perceptual hashing method, calculates the grayscale distribution similarity between the input DDM image and each reference DDM image stored in a system database, and simultaneously calculates the perceptual hash value of the input DDM image, and calculates the Hamming distance with the perceptual hash of each reference DDM image stored in the system database. The reference DDM images with similar grayscale histograms and the reference DDM images with the closest Hamming distance are output, the number of reference DDM images of each type is counted, and the type with the largest number is output as the classification result for DDM images to distinguish sea ice from seawater. The present invention extracts the grayscale histogram and image perceptual hash of the DDM image as features of the DDM image, and then performs similarity calculations with the DDM images stored in the system database to achieve the purpose of distinguishing seawater from sea ice, effectively improving the calculation speed while reducing the consumption of computing resources. The present invention can improve the accuracy of data prediction while reducing dependence on the amount of data samples; the present invention has low requirements for computing power and can achieve the effect of identifying sea ice and sea water in DDM images without excessive computing resources; the algorithm complexity of the present invention is low and the algorithm interpretability is strong. The grayscale histogram distribution and perceptual hash value of the image can be used as binary representations on the computer to effectively represent the information of the image. By calculating the similarity with the reference DDM image stored in the system database, the effect of distinguishing sea ice and sea water can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] A more complete understanding of exemplary embodiments of the present invention may be obtained by referring to the following drawings:

[0047] Figure 1 1 is a flow chart of a method for classifying sea ice and seawater provided by an exemplary embodiment of the present invention;

[0048] Figure 2 This is an overall flow chart of sea ice and seawater classification provided by an exemplary embodiment of the present invention;

[0049] Figure 3 is a flow chart of making a reference DDM image in a preset database provided by an exemplary embodiment of the present invention;

[0050] Figure 4 It is a schematic structural diagram of a sea ice and seawater classification system provided by an exemplary embodiment of the present invention;

[0051] Figure 5 This is a structure of an electronic device provided by an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0052] Below, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0053] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0054] Figure 1 The flow chart of the sea ice and seawater classification method provided by the present invention is shown in FIG. Figure 1 As shown, the sea ice and seawater classification methods include:

[0055] Step S101: Acquire a target delay-Doppler image of an L1-level product of a global navigation satellite system reflection technology;

[0056] Optionally, obtaining the target time-delay Doppler image of the L1-level product of the global navigation satellite system reflection technology includes: collecting the initial time-delay Doppler image of the L1-level product of the global navigation satellite system reflection technology within a period of time; performing quality control on the initial time-delay Doppler image according to the quality flag provided by the global navigation satellite system reflection technology, removing abnormal images of the initial time-delay Doppler image, and performing noise reduction processing on the initial time-delay Doppler image; and selecting a corresponding time-delay Doppler image from the initial time-delay Doppler image after the noise reduction processing as the target time-delay Doppler image according to the user's selection operation.

[0057] In the embodiment of the present invention, Figure 2 As shown, GNSS-R L1 data is collected over a period of time. Quality control is then performed based on the quality flag provided by the GNSS-R L1 data, and data with Nan or negative values ​​are removed. Next, noise reduction is performed on the DDM data. Finally, the user selects a DDM image as the target delay-Doppler image input to the discrimination system.

[0058] Step S102: determining a grayscale histogram of the target Delay Doppler image, and calculating an image-perceptual hash value of the target Delay Doppler image;

[0059] Optionally, calculating the image-perceptual hash value of the target Delay Doppler image includes: calculating a differential hash value of the target Delay Doppler image; calculating a preliminary perceptual hash value of the target Delay Doppler image; and determining the image-perceptual hash value of the target Delay Doppler image based on the differential hash value and the preliminary perceptual hash value.

[0060] In the embodiments of the present invention, research has found that the image histogram is one of the key features for describing image texture information. It represents the brightness distribution of an image, providing a histogram of the number of pixels at a certain brightness or brightness range within the image. This is essentially counting the number of pixels at a certain brightness within an image. Image histograms are widely used in various image processing fields, including image classification, due to their low computational cost and numerous advantages, such as invariance to translation, rotation, and scaling.

[0061] Image-aware hashing technology is currently widely used in multimedia information security fields such as image retrieval. Therefore, it can also be used to describe image information and map the image's feature data to a short string code, such as a binary string. It is a compact representation based on image content, which enables visually identical images to be mapped to the same or similar hash values, and different images to be mapped to different hash values. It has the advantages of strong robustness and strong distinguishability.

[0062] Therefore, the present invention adopts traditional image understanding and recognition methods. Affected by the reflective surface medium, the DDM images of sea ice and sea water have obvious differences. Therefore, image histogram distribution vector, image perception hash and other information can be extracted as detailed representation information to describe the image.

[0063] In an embodiment of the present invention, the grayscale histogram distribution is one of the important tools for describing the texture characteristics of an image. The grayscale histogram of the input DDM image is calculated as follows: the number of grayscale values ​​in [0, 255] is counted to obtain a 255*1 vector, where each value represents the number of grayscale values ​​corresponding to the image.

[0064] In an embodiment of the present invention, the specific steps of calculating the binary sequence of image hash values ​​of the input target DDM image include:

[0065] 1) Calculate the image difference hash. First, resize the image to 9 columns by 8 rows. Second, grayscale the image. Third, iterate through the 9 columns by 8 rows of pixels and compare the current pixel with the pixels to its left and right. If the pixel to the left of the current pixel is greater than the pixel to its right, the current pixel's position is marked as 1; otherwise, it is marked as 0. Each row has 9 pixels, and 8 values ​​can be compared. Therefore, 8 rows of pixels can generate 8*8 comparison values. The difference hash value at this point is a 64-bit binary sequence, denoted as hash1.

[0066] 2) Calculate the image perceptual hash; first, resize the image to 32 columns * 32 rows; second, perform grayscale processing; third, perform discrete cosine transform (DCT) on the image and retain the 8 columns * 8 rows in the upper left corner of the transformed matrix, denoted as D 8*8 Calculate D 8*8 The average value of mean ; Fourth, traverse D 8*8 Matrix, which will combine each pixel value with d mean Compare. Greater than or equal to d mean , recorded as 1, less than the average value, recorded as 0, so we can get an 8*8 comparison value. The perceived hash value at this time is a 64-bit binary sequence, recorded as hash2.

[0067] 3) Concatenate the calculated hash1 and hash2 together as the image hash value binary sequence of the target DDM image.

[0068] Step S103: calculating the similarity between the grayscale histogram distribution vector of the target Delay Doppler image and the reference Delay Doppler image stored in the preset database, and calculating the Hamming distance between the image perceptual hash value of the target Delay Doppler image and the image perceptual hash value of the reference Delay Doppler image stored in the preset database;

[0069] Optionally, calculating the similarity between the grayscale histogram distribution vector of the target Delay Doppler image and the reference Delay Doppler image stored in a preset database includes calculating the similarity between the grayscale histogram distribution vector of the target Delay Doppler image and the reference Delay Doppler image stored in the preset database using the following formula:

[0070]

[0071] Where, P a Represents the grayscale histogram distribution vector of the target delay-Doppler image; represents the average value of the grayscale histogram distribution vector of the target delay Doppler image; P i Represents the grayscale histogram distribution vector of the i-th reference delay-Doppler image stored in the preset database; represents the average value of the grayscale histogram distribution vector of the i-th reference delay-Doppler image stored in the preset database.

[0072] In an embodiment of the present invention, the input DDM image is compared with the reference DDM images stored in the system database. First, the similarity of the grayscale histogram distribution of the input DDM image and the grayscale histogram distribution of all reference DDM images is calculated respectively, and multiple (for example, but not limited to the first 10) reference DDM images that are most similar to the DDM are output. The formula for calculating the grayscale histogram similarity is as follows:

[0073]

[0074] Among them, P a Represents the grayscale histogram distribution vector of the target delay Doppler image, with a size of 256*1; represents the average value of the grayscale histogram distribution vector of the target delay Doppler image; P i Represents the grayscale histogram distribution vector of the i-th reference delay-Doppler image stored in the preset database; represents the average value of the grayscale histogram distribution vector of the i-th reference delay-Doppler image stored in the preset database;

[0075] Secondly, the Hamming distance between the image hash value of the input DDM image and the image perceptual hash values ​​of all reference DDM images is calculated. If the Hamming distance is smaller, it proves that the two images are more similar, and thus multiple (for example but not limited to 10) reference DDM images with the closest Hamming distance to the DDM are output.

[0076] Table 1 below shows the average Hamming distance of perceptual hashing of seawater / sea ice DDM images:

[0077] Table 1

[0078] Seawater DDM Sea Ice DDM Seawater DDM 12.03305785123967 24.9 Sea Ice DDM 24.9 15.82

[0079] Optionally, the method also includes preparing a reference delay Doppler image in a preset database in the following manner: collecting historical delay Doppler images of L1-level products of global navigation satellite system reflection technology and OSISAF sea ice data, wherein the sea ice data includes data on sea ice density and sea ice type; performing quality control on the historical delay Doppler images and OSISAF sea ice data according to the quality flag provided by the global navigation satellite system reflection technology, and removing outliers; performing spatiotemporal matching on the historical delay Doppler images and the OSISAF sea ice data, selecting a delay Doppler image whose matching result is seawater, selecting a delay Doppler image whose matching result is one-year ice, and selecting a delay Doppler image whose matching result is multi-year ice; storing the delay Doppler image of seawater in a preset database, storing the delay Doppler image of one-year ice and the delay Doppler image of multi-year ice in the preset database according to a preset ratio, and obtaining a reference delay Doppler image in the preset database.

[0080] In the embodiment of the present invention, the specific steps of preparing the reference delay Doppler image in the preset database are as follows: Figure 3 Shown, including:

[0081] 1) Collection of GNSS-R L1 level DDM images and OSISAF sea ice data, including sea ice density and sea ice type data.

[0082] 2) Data preprocessing: This step involves quality control of the GNSS-R L1 data and spatiotemporal matching with the reference sea ice data. Quality control uses the L1 quality flags to filter out observations with large spacecraft attitude errors or anomalies, high uncertainty in transmitter power or antenna gain, invalid or anomaly DDM data, instrument data transmission and calibration issues, and data with Nan or negative values. The principle of spatiotemporal matching is to match the latitude, longitude, and time of the GNSS-R mirror reflection point with the reference sea ice data using bilinear interpolation.

[0083] 3) Filter out data that matches OSISAF data as seawater and store the corresponding GNSS-R L1 DDM images in the sea ice and seawater system identification database. Filter out data that matches OSISAF data as one-year ice and multi-year ice, and store the corresponding GNSS-R L1 DDM images in a 1:1 ratio in the sea ice and seawater system identification database.

[0084] Step S104: determining, from a preset database based on the similarity calculation result, a plurality of reference Delay Doppler images having a similarity with the target Delay Doppler image greater than a first threshold, and determining, from the preset database based on the Hamming distance calculation result, a plurality of reference Delay Doppler images having a Hamming distance with the target Delay Doppler image greater than a second threshold, to obtain a reference Delay Doppler image set corresponding to the target Delay Doppler image;

[0085] Step S105: Classify the reference Delay Doppler images in the reference Delay Doppler image set, determine the target type corresponding to the reference Delay Doppler image with the largest number according to the classification result, and determine the target type as the classification result of the target Delay Doppler image for distinguishing sea ice and seawater.

[0086] The present invention extracts input DDM image features based on image grayscale histogram and image-aware hashing. By mapping the image feature data to a short string code as a compact representation of the image content instead of taking the entire DDM image as input, the system's computing resource consumption can be effectively reduced.

[0087] The present invention has the advantage of being able to quickly identify DDM images of sea ice and seawater when faced with massive amounts of satellite remote sensing data, especially spaceborne GNSS-R data. Image-aware hashing technology maps image feature data to a short string code, a compact representation based on image content. It maps visually identical images to the same or similar hashes, while different images are mapped to different hashes. In practical applications, using image hashes to represent the image itself and mapping the image itself to a short sequence of numbers can effectively reduce image storage costs and computational complexity, enabling efficient data processing when faced with massive amounts of DDM data. A DDM image is input, its grayscale histogram and image-aware hash are calculated, and then the grayscale histogram similarity and the Hamming distance of the image-aware hash are calculated with each reference sea ice and seawater DDM image stored in the discrimination system database. Multiple images with the highest similarity and the closest Hamming distance are output, respectively. The output reference DDM images are statistically analyzed, and the types of each type of DDM image are statistically analyzed. The result with the most statistical types is output, thereby achieving discrimination of sea ice and seawater DDM images.

[0088] In summary, the present invention extracts input DDM image features based on the image grayscale histogram and the image perceptual hashing method, calculates the grayscale distribution similarity between the input DDM image and each reference DDM image stored in the system database, and at the same time, calculates the perceptual hash value of the input DDM image, and calculates the Hamming distance with the perceptual hash of each reference DDM image stored in the system database. The reference DDM image with similar grayscale histogram distribution and the reference DDM image with the closest Hamming distance are output, and the number of reference DDM images of each type is counted, and the type with the largest number is output as the classification result of the DDM image for distinguishing sea ice and seawater. The present invention extracts the grayscale histogram and image perceptual hash of the DDM image as the features of the DDM image, and then performs similarity calculations with the DDM images stored in the system database to achieve the purpose of distinguishing seawater and sea ice, effectively improving the calculation speed while reducing the consumption of computing resources. The present invention can improve the accuracy of data prediction while reducing dependence on the amount of data samples; the present invention has low requirements for computing power and can achieve the effect of identifying sea ice and sea water in DDM images without excessive computing resources; the algorithm complexity of the present invention is low and the algorithm interpretability is strong. The grayscale histogram distribution and perceptual hash value of the image can be used as binary representations on the computer to effectively represent the information of the image. By calculating the similarity with the reference DDM image stored in the system database, the effect of distinguishing sea ice and sea water can be achieved.

[0089] Exemplary Systems

[0090] Figure 4 FIG. 1 is a schematic diagram of the structure of a sea ice and seawater classification system provided by an exemplary embodiment of the present invention. Figure 4 As shown, the system 400 includes:

[0091] The image acquisition module 410 is used to acquire a target delay Doppler image of the L1 level product of the global navigation satellite system reflection technology;

[0092] A first calculation module 420 is configured to determine a grayscale histogram of a target delay-Doppler image and calculate an image-perceptual hash value of the target delay-Doppler image;

[0093] A second calculation module 430 is configured to calculate the similarity between the grayscale histogram distribution vector of the target Delay Doppler image and the reference Delay Doppler image stored in a preset database, and calculate the Hamming distance between the image-perceptual hash value of the target Delay Doppler image and the image-perceptual hash value of the reference Delay Doppler image stored in the preset database;

[0094] A first determining module 440 is configured to determine, from a preset database, a plurality of reference Delay Doppler images having a similarity greater than a first threshold to the target Delay Doppler image based on the similarity calculation result, and to determine, from the preset database, a plurality of reference Delay Doppler images having a Hamming distance greater than a second threshold to the target Delay Doppler image based on the Hamming distance calculation result, to obtain a reference Delay Doppler image set corresponding to the target Delay Doppler image;

[0095] The second determination module 450 is used to classify the reference delay Doppler images in the reference delay Doppler image set, determine the target type corresponding to the reference delay Doppler image with the largest number according to the classification results, and determine the target type as the classification result of the target delay Doppler image for distinguishing sea ice and seawater.

[0096] Optionally, the image acquisition module 410 is specifically configured to:

[0097] Collect initial time-delay Doppler images of the L1 level products of the GNSS reflectometry technology over a period of time;

[0098] Performing quality control on the initial delay Doppler image based on the quality flag provided by the global navigation satellite system reflection technology, removing abnormal images from the initial delay Doppler image, and performing noise reduction processing on the initial delay Doppler image;

[0099] According to the user's selection operation, a corresponding delay Doppler image is selected from the initial delay Doppler images after noise reduction processing as the target delay Doppler image.

[0100] Optionally, the first calculation module 420 is specifically configured to:

[0101] Calculate the difference hash value of the target delay-Doppler image;

[0102] Calculate the preliminary perceptual hash value of the target delay-Doppler image;

[0103] An image-perceptual hash value of the target delay-Doppler image is determined based on the difference hash value and the preliminary perceptual hash value.

[0104] The sea ice and seawater classification system of the embodiment of the present invention corresponds to the sea ice and seawater classification method of another embodiment of the present invention, which will not be described in detail here.

[0105] Exemplary electronic devices

[0106] Figure 5 This is the structure of an electronic device provided by an exemplary embodiment of the present invention. Figure 5 As shown, the electronic device 50 includes one or more processors 51 and a memory 52 .

[0107] The processor 51 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0108] The memory 52 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may, for example, include read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 51 may run the program instructions to implement the method for information mining of historical change records and / or other desired functions of the software program of each embodiment of the present invention described above. In one example, the electronic device may further include: an input device 53 and an output device 54, which are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0109] In addition, the input device 53 may also include, for example, a keyboard, a mouse, and the like.

[0110] The output device 54 can output various information to the outside. The output device 54 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto.

[0111] Of course, to simplify, Figure 5Only some of the components related to the present invention in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0112] Exemplary computer program products and computer-readable storage media

[0113] In addition to the above-mentioned methods and devices, an embodiment of the present invention may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to perform the steps of the method according to various embodiments of the present invention described in the above "Exemplary Method" section of this specification.

[0114] The computer program product may be written in any combination of one or more programming languages ​​to implement the operations of embodiments of the present invention, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0115] In addition, an embodiment of the present invention may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, enables the processor to execute the steps of the method for information mining of historical change records according to various embodiments of the present invention described in the above "Exemplary Method" section of this specification.

[0116] The computer-readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, include but is not limited to a system, system or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0117] The basic principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in the present invention are merely illustrative and non-limiting, and should not be construed as necessarily possessed by each embodiment of the present invention. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, and are not intended to be limiting. These details do not necessarily limit the present invention to being implemented using these specific details.

[0118] Each embodiment in this specification is described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. References to the same or similar parts between the various embodiments are sufficient. For system embodiments, since they largely correspond to method embodiments, their description is relatively simple. For relevant parts, references to the description of the method embodiments are sufficient.

[0119] The block diagrams of the devices, systems, equipment, and systems involved in the present invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, systems, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "including," "comprising," "having," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0120] The method and system of the present invention may be implemented in many ways. For example, the method and system of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above sequence of steps for the method is for illustration only, and the steps of the method of the present invention are not limited to the sequence specifically described above, unless otherwise specified. In addition, in some embodiments, the present invention may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Thus, the present invention also covers recording media that store programs for executing the method according to the present invention.

[0121] It should also be noted that, in the system, device and method of the present invention, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. The above description of the disclosed aspects is provided to enable any technician in this field to make or use the present invention. Various modifications to these aspects will be very obvious to those skilled in the art, and the general principles defined here can be applied to other aspects without departing from the scope of the present invention. Therefore, the present invention is not intended to be limited to the aspects shown here, but according to the widest scope consistent with the principles disclosed here and novel features.

[0122] The above description has been provided for the purpose of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present invention to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for classifying sea ice and seawater, characterized in that: include: Acquire target delay-Doppler images of the L1 level product of the GNSS reflectometry technology; Determine the grayscale histogram distribution vector diagram of the target delay Doppler image, and calculate the image perception hash value of the target delay Doppler image; Calculating the similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in the preset database, and calculating the Hamming distance between the image-aware hash value of the target delay Doppler image and the image-aware hash value of the reference delay Doppler image stored in the preset database; Determining, from a preset database based on the similarity calculation result, a plurality of reference Delay Doppler images having a similarity with the target Delay Doppler image greater than a first threshold, and determining, from the preset database based on the Hamming distance calculation result, a plurality of reference Delay Doppler images having a Hamming distance with the target Delay Doppler image greater than a second threshold, to obtain a reference Delay Doppler image set corresponding to the target Delay Doppler image; Classifying the reference delay Doppler images in the reference delay Doppler image set, determining the target type corresponding to the reference delay Doppler image with the largest number according to the classification results, and determining the target type as the classification result of the target delay Doppler image for distinguishing sea ice and seawater; The step of calculating the image-aware hash value of the target delay-Doppler image includes: Calculate the difference hash value of the target delay-Doppler image; Calculate the preliminary perceptual hash value of the target delay-Doppler image; An image-perceptual hash value of the target delay-Doppler image is determined based on the difference hash value and the preliminary perceptual hash value.

2. The method according to claim 1, characterized in that The step of obtaining a target delay Doppler image of an L1 product of a global navigation satellite system reflection technology includes: Collect initial time-delay Doppler images of the L1 level products of the GNSS reflectometry technology over a period of time; Performing quality control on the initial delay Doppler image based on the quality flag provided by the global navigation satellite system reflection technology, removing abnormal images from the initial delay Doppler image, and performing noise reduction processing on the initial delay Doppler image; According to the user's selection operation, a corresponding delay Doppler image is selected from the initial delay Doppler images after noise reduction processing as the target delay Doppler image.

3. The method according to claim 1, characterized in that Calculating the similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in a preset database includes: The similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in the preset database is calculated using the following formula: Where, P a Represents the grayscale histogram distribution vector of the target delay Doppler image; represents the average value of the grayscale histogram distribution vector of the target delay Doppler image; P i Represents the first i The grayscale histogram distribution vector of the reference delay Doppler image; represents the first i The average value of the grayscale histogram distribution vector of the reference delay-Doppler image.

4. The method according to claim 1, wherein It also includes the creation of reference delay-Doppler images from a preset database by: Collect historical time-delay Doppler images of the L1 level products of the Global Navigation Satellite System reflectometry technology and OSISAF sea ice data, including sea ice concentration and sea ice type data; Based on the quality flags provided by GNSS reflectometry, quality control of historical time-delay Doppler images and OSISAF sea ice data was performed to remove outliers. Perform spatiotemporal matching of historical time-delay Doppler images with OSISAF sea ice data, and select the time-delay Doppler image of seawater, the time-delay Doppler image of one-year ice, and the time-delay Doppler image of multi-year ice as the matching result; The time delay Doppler image of seawater is stored in a preset database, and the time delay Doppler image of one-year ice and the time delay Doppler image of multi-year ice are stored in the preset database according to a preset ratio to obtain a reference time delay Doppler image in the preset database.

5. A sea ice and seawater classification system, characterized in that: include: Image acquisition module, used to obtain target delay Doppler images of L1 level products of global navigation satellite system reflection technology; A first calculation module is used to determine a grayscale histogram of a target delay-Doppler image and calculate an image-perceptual hash value of the target delay-Doppler image; A second calculation module is used to calculate the similarity between the grayscale histogram distribution vector of the target delay Doppler image and the reference delay Doppler image stored in the preset database, and calculate the Hamming distance between the image perception hash value of the target delay Doppler image and the image perception hash value of the reference delay Doppler image stored in the preset database; a first determining module, configured to determine, from a preset database based on a similarity calculation result, a plurality of reference Delay Doppler images having a similarity with the target Delay Doppler image greater than a first threshold, and, based on a Hamming distance calculation result, determine, from the preset database, a plurality of reference Delay Doppler images having a Hamming distance with the target Delay Doppler image greater than a second threshold, to obtain a reference Delay Doppler image set corresponding to the target Delay Doppler image; A second determination module is configured to classify the reference delay Doppler images in the reference delay Doppler image set, determine the target type corresponding to the reference delay Doppler image with the largest number according to the classification results, and determine the target type as the classification result of the target delay Doppler image for distinguishing sea ice and seawater; The first calculation module is specifically configured to: Calculate the difference hash value of the target delay-Doppler image; Calculate the preliminary perceptual hash value of the target delay-Doppler image; An image-perceptual hash value of the target delay-Doppler image is determined based on the difference hash value and the preliminary perceptual hash value.

6. The system according to claim 5, characterized in that The image acquisition module is specifically used to: Collect initial time-delay Doppler images of the L1 level products of the GNSS reflectometry technology over a period of time; Performing quality control on the initial delay Doppler image based on the quality flag provided by the global navigation satellite system reflection technology, removing abnormal images from the initial delay Doppler image, and performing noise reduction processing on the initial delay Doppler image; According to the user's selection operation, a corresponding delay Doppler image is selected from the initial delay Doppler images after noise reduction processing as the target delay Doppler image.

7. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and the computer program is used to execute the method according to any one of claims 1 to 4.

8. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing instructions executable by the processor; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method according to any one of claims 1 to 4.

Images