Remote sensing monitoring method and device for duckweed-type water bodies based on deep learning

Abstract

The present invention provides a method and device for remote sensing monitoring of duckweed-type water bodies based on deep learning. The method includes: obtaining remote sensing tiles of duckweed-type water bodies as RGB three-channel samples; and performing calculations on the RGB three-channel samples based on chromatic angle, slope sum, and green channel index to obtain chromatic angle bands, slope sum bands, and green channel index bands, and performing band synthesis to obtain ASGI remote sensing image tiles; setting attribute thresholds based on the duckweed-type water bodies in the ASGI remote sensing image tiles, and constructing a sample data set of duckweed-type water bodies; training an initial deep learning model with the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies. This application solves the technical problem of the prior art in which, when using remote sensing means to identify duckweed-type water bodies, the recognition accuracy is reduced due to the defect that some duckweed-type water bodies are small in area and easily confused with other ground objects.

Description

Technical Field

[0001] The present invention relates to the field of water environment remote sensing technology, and in particular to a duckweed-type water body remote sensing monitoring method and device based on deep learning. Background Art

[0002] In addition to the typical black and smelly black water bodies in cities, new manifestations of black and smelly water bodies in rural areas have emerged, including duckweed and algal blooms. The Lemna family is the world's smallest monocotyledonous flowering plant, growing rapidly and enduringly. It comprises four genera and approximately ten species. Numerous studies have shown that various duckweed species can remove nitrogen and phosphorus nutrients from water, making them widely used in water purification. However, when large numbers of duckweed accumulate or bloom, covering all or part of a water surface, they can cause hypoxia and eutrophication, hindering water reoxygenation and the availability of light to aquatic organisms. This can lead to the death of aquatic organisms due to lack of oxygen, and a gradual decline in water quality. However, little research has examined the adverse effects of duckweed becoming a dominant species on water quality. Furthermore, harmful algal blooms have long been a common ecological and environmental problem facing freshwater lakes and ponds worldwide.

[0003] There are three possible scenarios for water bodies covered by duckweed. First, the water quality is normal, with duckweed floating on the surface and not yet forming a dominant population. The duckweed can be harvested and used as food for livestock and poultry. Second, the water quality is poor, and duckweed is artificially released for water purification. Third, the water body is black and smelly, where duckweed completely or partially covers the water surface, forming a dominant population, hindering the growth of other photosynthetic plants and animals in the water, causing the death of these organisms and turning the water black and smelly. Therefore, effective identification of duckweed-infected water bodies is crucial. However, since duckweed-infected water bodies are mostly found in rural areas, their distribution is large and scattered, making them difficult to identify through ground-based inspections. Remote sensing, on the other hand, primarily relies on distinguishing duckweed-infected water bodies from other landforms based on their physical characteristics. While this method is fast and covers a wide range, it also requires higher resolution remote sensing images. Furthermore, some duckweed-infected water bodies are small in size and easily confused with other landforms, resulting in reduced identification accuracy. Summary of the Invention

[0004] The present invention provides a method and device for remote sensing monitoring of duckweed-type water bodies based on deep learning, which is used to solve the technical problem in the existing technology of using remote sensing means to identify duckweed-type water bodies, because some duckweed-type water bodies have a small area and are easily confused with other ground objects, resulting in reduced recognition accuracy.

[0005] The present invention provides a method for remote sensing monitoring of duckweed-type water bodies based on deep learning, comprising the following steps:

[0006] Obtain remote sensing tiles of duckweed-type water bodies as RGB three-channel samples;

[0007] Performing a calculation based on the chromaticity angle on the RGB three-channel samples to obtain a chromaticity angle band, performing a calculation based on the slope sum on the RGB three-channel samples to obtain a slope sum band, and performing a calculation based on the green channel index on the RGB three-channel samples to obtain a green channel index band;

[0008] Performing band synthesis on the chromatic angle band, the slope sum band, and the green channel index band to obtain an ASGI remote sensing image tile;

[0009] Setting attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles, and constructing a sample dataset of duckweed-type water bodies;

[0010] An initial deep learning model is trained using the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

[0011] In some embodiments, performing the calculation based on the chromaticity angle on the RGB three-channel samples to obtain the chromaticity angle band includes:

[0012] Get the band linear summation coefficient of map remote sensing;

[0013] Calculate the stimulus values ​​of the R, G, and B values ​​of the RGB three-channel samples according to the linear summation coefficients of the bands to obtain the stimulus values ​​of the XYZ three primary colors;

[0014] Determining a chromaticity abscissa and a chromaticity ordinate based on the XYZ three primary color stimulus values;

[0015] A chromaticity angle calculation is performed on the chromaticity horizontal coordinate and the chromaticity vertical coordinate to obtain a chromaticity angle band.

[0016] In some embodiments, performing the calculation based on the slope sum on the RGB three-channel samples to obtain the slope and band includes:

[0017] Determine a first ratio of the B value to the G value in the RGB three-channel sample;

[0018] Determine a second ratio of the R value to the G value in the RGB three-channel sample;

[0019] The sum of the first ratio and the second ratio is used as the slope and band.

[0020] In some embodiments, the green channel index band is obtained by performing calculation based on the green channel index on the RGB three-channel samples, including:

[0021] Determine the sum of the channel values ​​of the B value and the R value in the RGB three-channel sample;

[0022] The ratio of the square of the G value in the RGB three-channel sample to the sum of the channel values ​​is used as the green channel index band.

[0023] In some embodiments, setting an attribute threshold based on the duckweed-type water body in the ASGI remote sensing image tile and constructing a sample dataset of the duckweed-type water body includes:

[0024] Determine the surface vector of the duckweed-shaped water body in the ASGI remote sensing image tile;

[0025] When the duckweed-type water body is a real duckweed-type water body, setting a first attribute threshold for the surface vector of the duckweed-type water body;

[0026] When the duckweed-type water body is a non-duckweed-type water body, setting a second attribute threshold for the surface vector of the duckweed-type water body;

[0027] The data labels corresponding to the ASGI remote sensing image tiles are constructed according to the first attribute threshold and the second attribute threshold, and a sample data set of duckweed-type water bodies is obtained.

[0028] In some embodiments, the training of the initial deep learning model using the sample data set to obtain a duckweed-type water body recognition model includes:

[0029] Inputting the training samples of the sample data set into the initial deep learning model for forward propagation to obtain the predicted attribute values ​​of the training samples for duckweed-type water bodies;

[0030] Constructing a training loss value based on the difference between the predicted attribute value and the data label of the training sample, and backpropagating the training loss value in the initial deep learning model to update the model parameters of the initial deep learning model;

[0031] When the training loss value begins to converge or reaches a preset training round, the training of the initial deep learning model is terminated to obtain a duckweed-type water body recognition model.

[0032] The present invention also provides a duckweed-type water body remote sensing monitoring device based on deep learning, which includes the following modules:

[0033] The acquisition module is used to obtain remote sensing tiles of duckweed-type water bodies as RGB three-channel samples;

[0034] A determination module is configured to perform a calculation based on the chromaticity angle on the RGB three-channel samples to obtain a chromaticity angle band, perform a calculation based on the slope sum on the RGB three-channel samples to obtain a slope sum band, and perform a calculation based on the green channel index on the RGB three-channel samples to obtain a green channel index band;

[0035] A synthesis module, configured to synthesize the chroma angle band, the slope sum band, and the green channel index band to obtain an ASGI remote sensing image tile;

[0036] A construction module is used to set attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles and to construct a training data set of duckweed-type water bodies;

[0037] The recognition module is used to train the initial deep learning model through the training data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

[0038] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, it implements any of the deep learning-based remote sensing monitoring methods for duckweed-type water bodies described above.

[0039] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements any of the deep learning-based remote sensing monitoring methods for duckweed-type water bodies described above.

[0040] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described methods for remote sensing monitoring of duckweed-type water bodies based on deep learning.

[0041] The present invention provides a method and device for remote sensing monitoring of duckweed-type water bodies based on deep learning. Based on remote sensing tiles of duckweed-type water bodies, three physical features are constructed for identifying duckweed-type water bodies: the chromatic angle band, the slope sum band, and the green channel index band. The three physical features are combined to obtain an ASGI remote sensing image tile. A training data set for duckweed-type water bodies is constructed by setting attribute thresholds to train an initial deep learning model to obtain a duckweed-type water body recognition model for identifying duckweed-type water bodies. Thus, duckweed-type water bodies are simultaneously identified using the three physical features of the chromatic angle band, the slope sum band, and the green channel index band. This can specifically eliminate the confusing effects of other landforms, improve recognition accuracy, and overcome the defect in the prior art that the resolution of remote sensing images is low, resulting in reduced recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 It is a flow chart of the duckweed-type water body remote sensing monitoring method based on deep learning provided by the present invention.

[0044] Figure 2 This is a schematic diagram of the framework of the deep learning-based remote sensing monitoring method for duckweed-type water bodies provided by the present invention.

[0045] Figure 3 This is a schematic diagram of distinguishing easily confused ground objects after ASGI band combination provided by the present invention.

[0046] Figure 4 It is a structural schematic diagram of the duckweed-type water body remote sensing monitoring device based on deep learning provided by the present invention.

[0047] Figure 5 It is a schematic diagram of the physical structure of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0049] The following combination Figure 1 Describe the present invention's method for remote sensing monitoring of duckweed-type water bodies based on deep learning,

[0050] Figure 1 This is a flow chart of the method for remote sensing monitoring of duckweed-type water bodies based on deep learning provided by the present invention, such as Figure 1 As shown, the method includes the following steps 101 to 105.

[0051] Step 101: Obtain remote sensing tiles of duckweed-type water bodies as RGB three-channel samples.

[0052] First, we need to obtain remote sensing tiles of duckweed-like water bodies. Specifically, these tiles are Google Earth remote sensing image tiles with high spatial resolution. After preprocessing, they are converted into RGB (Red, Green, and Blue) three-channel samples. During preprocessing, remote sensing images with varying colors and high levels of cloudiness are removed, while those with uniform colors and low cloudiness are retained.

[0053] Step 102: Perform calculation based on the chromaticity angle on the RGB three-channel samples to obtain the chromaticity angle band, perform calculation based on the slope sum on the RGB three-channel samples to obtain the slope sum band, and perform calculation based on the green channel index on the RGB three-channel samples to obtain the green channel index band.

[0054] See here Figure 2 , Figure 2 This is a schematic diagram of the framework of the deep learning-based remote sensing monitoring method for duckweed-type water bodies provided by the present invention. After acquiring Google Earth remote sensing image tiles as RGB three-channel samples, the ASGI band calculation and synthesis module in this embodiment of the present invention calculates three physical characteristics of the RGB three-channel samples: the chromatic angle band (a), the slope sum band (S), and the green channel index band (GI) to distinguish duckweed-type water bodies. ASGI is an abbreviation for the combination of the three bands: chromatic angle (A), slope sum (S), and green channel index (GI). The order in which these three bands are calculated is not specific; however, this order is used for illustrative purposes only. This should not be a limitation in actual implementation.

[0055] For the chromatic angle band a, the RGB three-channel samples are calculated based on the chromatic angle to obtain the chromatic angle band. First, the band linear summation coefficients of the map remote sensing are obtained. These coefficients are fixed parameters in the chromatic angle algorithm. Then, based on the band linear summation coefficients, the R value, G value, and B value of the RGB three-channel samples are calculated for the stimulus value to obtain the XYZ three primary color stimulus values, which are expressed as the following formulas (1) to (3):

[0056] (1)

[0057] (2)

[0058] (3)

[0059] In the above formulas (1) to (3), R, G, and B represent the R value, G value, and B value of the RGB three-channel samples, respectively, and the other fixed parameters are the band linear summation coefficients of map remote sensing.

[0060] Next, based on the XYZ primary color stimulus values, the chromaticity coordinates are calculated, that is, the chromaticity horizontal coordinate x and the chromaticity vertical coordinate y are determined. The calculation formulas are expressed as follows: Formula (4) to Formula (5):

[0061] (4)

[0062] (5)

[0063] In the above formulas (4) and (5), X, Y, and Z represent the stimulus values ​​of the three primary colors, respectively.

[0064] Finally, the chromaticity angle is calculated for the chromaticity horizontal coordinate x and the chromaticity vertical coordinate y to obtain the chromaticity angle band a. The chromaticity angle calculation method is as follows:

[0065] (6)

[0066] In the above formula (6), 、 They represent the calibration values ​​of the chromaticity horizontal coordinate x and the chromaticity vertical coordinate y respectively.

[0067] In this embodiment of the present invention, chromatic angle is used as a physical feature to distinguish duckweed-type water bodies from other landforms. This is because chromatic angle is effective in distinguishing duckweed-type water bodies from light green vegetation, red paint, bright blue paint, and bright green paint. Because the chromatic angle of duckweed-type water bodies is primarily distributed between 181.6° and 182.0°, the chromatic angles of light green vegetation (181.1° to 181.8°) and red paint (179.7° to 180.2°) are smaller than those of duckweed-type water bodies, while the chromatic angles of bright blue paint (181.9° to 182.9°) and bright green paint (182.7° to 183.2°) are larger than those of duckweed-type water bodies. Therefore, chromatic angle can effectively distinguish easily confused landforms such as red paint and bright green paint.

[0068] For the slope and band S, the slope and band are calculated based on the slope sum of the RGB three-channel samples. The slope and band are calculated based on the R value, G value, and B value in the RGB three-channel samples. First, the first ratio of the B value to the G value in the RGB three-channel samples is determined, and then the second ratio of the R value to the G value in the RGB three-channel samples is determined. Finally, the sum of the first ratio and the second ratio is taken as the slope and band S, which is expressed as the following formula (7):

[0069] (7)

[0070] In the above formula (7), R, G, and B represent the R value, G value, and B value of the RGB three-channel sample, respectively.

[0071] In the embodiments of the present invention, the sum of slopes is used as a physical feature to distinguish chromatic angle duckweed-like water bodies from other landforms. This is because the sum of slopes can effectively distinguish duckweed-like water bodies from dark green paint. The sum of slopes of duckweed-like water bodies is primarily in the range of 1.3-1.6, whereas the sum of slopes of dark green paint is mostly less than 1.3 or greater than 1.6. Therefore, the sum of slopes can distinguish duckweed-like water bodies from dark green paint to a certain extent.

[0072] For the green channel index GI, the green channel index is calculated based on the RGB three-channel samples to obtain the green channel index band. Similarly, the calculation of the green channel index band is also based on the calculation of the R value, G value, and B value in the RGB three-channel samples. First, the channel value sum of the B value and the R value in the RGB three-channel samples is determined. Then, the ratio of the square of the G value in the RGB three-channel samples to the channel value sum is used as the green channel index band, which is expressed as the following formula (8):

[0073] (8)

[0074] In the above formula (8), R, G, and B represent the R value, G value, and B value of the RGB three-channel sample, respectively.

[0075] In this embodiment of the present invention, the Green Channel Index is used as a physical feature to distinguish chromatic angle duckweed-type water bodies from other landforms. This is because the Green Channel Index can effectively distinguish duckweed-type water bodies from sparse grassland, green water bodies, and artificial grassland. The Green Channel Index of duckweed-type water bodies mostly ranges from 105.1 to 136.9, while the Green Channel Index of sparse grassland is less than 105. The Green Channel Index of green water bodies and artificial grassland is mostly less than 105. Therefore, the Green Channel Index is effective in distinguishing sparse grassland, green water bodies, and artificial grassland.

[0076] Step 103: synthesize the chromaticity angle band, the slope sum band, and the green channel index band to obtain an ASGI remote sensing image tile.

[0077] After calculating the chromaticity angle band (a), slope sum band (S), and green channel index band (GI) of the RGB three-channel samples respectively through step 102, the chromaticity angle band, slope sum band, and green channel index band can be band synthesized to obtain an ASGI remote sensing image tile.

[0078] Step 104: Set attribute thresholds based on the duckweed-type water bodies in the ASGI remote sensing image tiles, and construct a training dataset of duckweed-type water bodies.

[0079] like Figure 2 As shown, after ASGI remote sensing image tiles are constructed through the ASGI band calculation and synthesis module, the next stage is model training. This embodiment of the present invention uses a deep learning-based recognition model to accurately identify duckweed-type water bodies. First, a duckweed-type water body recognition model must be trained to learn the physical characteristics of the three bands mentioned above. The initial deep learning model can use models such as MobileNet-Unet, MobileNet, SqueezeNet, and ShuffleNet. The MobileNet-Unet model is used as an example for illustration.

[0080] Here, we set attribute thresholds based on duckweed-like water bodies in ASGI remote sensing image tiles and construct a sample dataset of duckweed-like water bodies. First, we need to label duckweed-like water bodies in the training dataset and determine the surface vectors of duckweed-like water bodies in the ASGI remote sensing image tiles. Then, we construct the duckweed-like water body sample dataset. To better train the initial deep learning model and ensure convincing recognition results, we need to construct a sample dataset of at least 1,000.

[0081] A planar shp file (also known as Shapefile file) is constructed for each ASGI remote sensing image tile, and then the duckweed-type water body is manually outlined in detail in this shp file. Each duckweed-type water body is outlined in the form of a planar vector. Subsequently, in the ArcGIS software, the surface vector is edited and attribute values ​​are set, that is, the duckweed-type water body and the background features are distinguished by marking the attribute values. Specifically, when the duckweed-type water body is a real duckweed-type water body, a first attribute threshold is set for the planar vector of the duckweed-type water body, for example, 255. When the duckweed-type water body is a non-duckweed-type water body (such as a background feature), a second attribute threshold is set for the planar vector of the duckweed-type water body, for example, 0.

[0082] Finally, the data labels corresponding to the ASGI remote sensing image tiles are constructed based on the first attribute threshold of 255 and the second attribute threshold of 0, and a sample data set of duckweed-type water bodies is obtained. Here, the marked first attribute threshold of 255 and the second attribute threshold of 0 can be directly used as the data labels corresponding to the ASGI remote sensing image tiles. After marking the attribute values ​​of each shp file, these shp files are converted into tif files, and the naming method is consistent with the name of the ASGI remote sensing image tile. In this way, the data labels corresponding to the ASGI remote sensing image tiles can be matched according to the file name. Finally, all ASGI remote sensing image tiles and corresponding data labels are cut according to the same cutting rules, for example, they are all uniformly cut into 512*512 pixel blocks as the sample data set of duckweed-type water bodies.

[0083] Step 105: Train the initial deep learning model using the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

[0084] like Figure 2As shown, after the duckweed water body sample set is constructed, it is necessary to divide the duckweed water body sample set, that is, to divide the constructed sample data set. In the embodiment of the present invention, the sample data set is divided into a training set, a validation set, and a test set according to a ratio of 8:1:1 or 7:2:1. The training set is used to train the initial deep learning model Mobilenet-Uet, the validation set is used to verify the effect of the trained Mobilenet-Uet, and the test set is used to test and evaluate the recognition effect of Mobilenet-Uet. At the same time, it is necessary to ensure that the ratio of positive samples to negative samples in each type of divided data set is close to 1:1.

[0085] The following describes the process of training an initial deep learning model, Mobilenet-Uet, using a sample dataset to obtain a model for identifying duckweed-type water bodies. Training samples from the sample dataset (training set) are fed into the initial deep learning model for forward propagation, resulting in predicted attribute values ​​for duckweed-type water bodies. During the forward propagation process, the initial deep learning model learns three physical characteristics of duckweed-type water bodies from the training samples: the chromatic angle band (a), the slope sum band (S), and the green channel index band (GI). Based on these three physical characteristics, the model predicts the predicted attribute values ​​for duckweed-type water bodies. A training loss is then constructed based on the difference between the predicted attribute values ​​and the data labels (i.e., the labeled attribute values) of the training samples. This training loss is then used to perform backpropagation through the initial deep learning model. During the backpropagation process, the model gradients are calculated and updated using a gradient optimization algorithm to update the model parameters of the initial deep learning model. When the training loss converges or reaches the preset number of training rounds, the initial deep learning model has learned the optimal set of model parameters. Training of the initial deep learning model ends, and the initial deep learning model with the optimal model parameters can be used as the duckweed-type water body identification model.

[0086] like Figure 2 As shown in the figure, after obtaining the optimal model parameters through training with the training set and subsequently developing the duckweed-type water body recognition model, the model prediction process enters the accuracy verification phase. Here, the duckweed-type water body sample set is divided into validation and test sets to verify the model's accuracy (i.e., prediction performance). The duckweed-type water body recognition model can identify duckweed-type water bodies in both the validation and test sets, completing the model prediction process and further verifying model accuracy.

[0087] Specifically, the model's accuracy is verified using a test set or new ASGI remote sensing image tiles independent of the training set. The predicted attribute values ​​are used as the predictions by the duckweed-type water body recognition model, while the data labels (i.e., manually outlined and annotated attribute values) are used as the true values. The accuracy of the duckweed-type water body recognition model is evaluated using precision, overall accuracy (OA), recall, false alarm rate (FPR), miss rate (FNR), the harmonic mean of precision and recall (F1-score), and intersection of union (IoU).

[0088] Among them, the above indicators are calculated by the true positive (TP), true negative (TN), false negative (FP), and false positive (FN) in the confusion matrix, which are introduced one by one below.

[0089] Precision is expressed as the following formula (9):

[0090] (9)

[0091] Overall Accuracy (OA) is expressed as the following formula (10):

[0092] (10)

[0093] Recall is divided into positive sample recall And the negative sample recall rate , respectively expressed as the following formula (11) and formula (12):

[0094] (11)

[0095] (12)

[0096] The false alarm rate (FPR) is expressed as follows (13):

[0097] (13)

[0098] The missed detection rate (FNR) is expressed as the following formula (14):

[0099] (14)

[0100] The F1-score is expressed as the following formula (15):

[0101] (15)

[0102] The intersection over union (IoU) is expressed as the following formula (16):

[0103] (16)

[0104] The above indicators provide a comprehensive assessment of the performance of the duckweed-type water body recognition model. A higher indicator indicates better performance and higher accuracy in identifying duckweed-type water bodies. In practical applications, the duckweed-type water body recognition model can directly distinguish duckweed-type water bodies in ASGI remote sensing image tiles.

[0105] See also Figure 3 , Figure 3 This is a schematic diagram of distinguishing easily confused features after ASGI band combination provided by the present invention. After forming an ASGI remote sensing image tile by combining three ASGI bands, namely the chromatic angle band (a), the slope sum band (S), and the green channel index band, the duckweed-type water body recognition model is used to identify the duckweed-type water body in the ASGI remote sensing image tile. Figure 3 As shown in (a), the duckweed water body recognition model accurately identifies the duckweed water body that is manually outlined in the ASGI remote sensing image tile and displays the corresponding local magnified image. Figure 3 As shown in (b), the duckweed-type water body recognition model also accurately distinguishes light green vegetation, bright blue paint, bright green paint, and red paint, and displays the corresponding local enlarged images. Figure 3 As shown in (c), the duckweed-type water body recognition model can also accurately identify the dark green paint that is most easily confused, and displays the corresponding local enlarged image.

[0106] After the accuracy of the duckweed-type water body recognition model is verified, it can be put into practical application environments. By collecting remote sensing tiles of unknown duckweed-type water bodies, the duckweed-type water body recognition model is called to effectively identify various unknown duckweed-type water bodies in the remote sensing tiles.

[0107] In summary, the embodiment of the present invention, based on the remote sensing tiles of duckweed-type water bodies, constructs three physical features for identifying duckweed-type water bodies, namely, chromatic angle band, slope sum band, and green channel index band, and combines the three physical features to obtain ASGI remote sensing image tiles, and constructs a training data set for duckweed-type water bodies by setting attribute thresholds to train the initial deep learning model, obtain a duckweed-type water body recognition model to identify duckweed-type water bodies, and verify the accuracy of the duckweed-type water body recognition model through various indicators during the training process, thereby ensuring the effectiveness of the duckweed-type water body recognition model. Thus, duckweed-type water bodies are simultaneously identified through the three physical features of chromatic angle band, slope sum band, and green channel index band, which can specifically eliminate the confusing effects of other landforms, improve the recognition accuracy, and overcome the defect in the prior art that the resolution of remote sensing images is not high, resulting in reduced recognition accuracy.

[0108] The duckweed-type water body remote sensing monitoring device based on deep learning provided by the present invention is described below. The duckweed-type water body remote sensing monitoring device based on deep learning described below and the duckweed-type water body remote sensing monitoring method based on deep learning described above can be referenced to each other.

[0109] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of the duckweed-type water body remote sensing monitoring device based on deep learning provided by the present invention. Figure 4 As shown, the duckweed-type water body remote sensing monitoring device based on deep learning includes an acquisition module 401, a determination module 402, a synthesis module 403, a construction module 404, and an identification module 405. Among them, the acquisition module 401 is used to obtain the remote sensing tile of the duckweed type water body as an RGB three-channel sample; the determination module 402 is used to perform a calculation based on the chromatic angle on the RGB three-channel sample to obtain the chromatic angle band, perform a calculation based on the slope sum on the RGB three-channel sample to obtain the slope sum band, and perform a calculation based on the green channel index on the RGB three-channel sample to obtain the green channel index band; the synthesis module 403 is used to perform band synthesis on the chromatic angle band, the slope sum band and the green channel index band to obtain an ASGI remote sensing image tile; the construction module 404 is used to set the attribute threshold according to the duckweed type water body in the ASGI remote sensing image tile, and construct a training data set for the duckweed type water body; the identification module 405 is used to train the initial deep learning model through the training data set to obtain a duckweed type water body recognition model to identify the duckweed type water body.

[0110] It can be seen from this that the beneficial effects produced by the duckweed-type water body remote sensing monitoring device based on deep learning provided by the present invention should be the same as the duckweed-type water body remote sensing monitoring method based on deep learning described above. Therefore, the beneficial effects produced by the duckweed-type water body remote sensing monitoring device based on deep learning will not be repeated here.

[0111] Figure 5 This is a schematic diagram of the physical structure of an electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 840, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. Processing 510 can call the logic instructions in the memory 530 to execute a deep learning-based remote sensing monitoring method for duckweed-type water bodies, which includes: obtaining remote sensing tiles of duckweed-type water bodies as RGB three-channel samples; performing calculations based on chromatic angles on the RGB three-channel samples to obtain chromatic angle bands, performing calculations based on slope sums on the RGB three-channel samples to obtain slope sum bands, and performing calculations based on green channel indexes on the RGB three-channel samples to obtain green channel index bands; performing band synthesis on the chromatic angle bands, the slope sum bands, and the green channel index bands to obtain ASGI remote sensing image tiles; setting attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles, and constructing a sample data set of duckweed-type water bodies; training an initial deep learning model through the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

[0112] Furthermore, the logic instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0113] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the deep learning-based remote sensing monitoring method for duckweed-type water bodies provided by the above methods, the method including: obtaining remote sensing tiles of duckweed-type water bodies as RGB three-channel samples; performing calculations based on chromatic angles on the RGB three-channel samples to obtain chromatic angle bands, performing calculations based on slope sums on the RGB three-channel samples to obtain slope sum bands, and performing calculations based on green channel indexes on the RGB three-channel samples to obtain green channel index bands; performing band synthesis on the chromatic angle bands, the slope sum bands, and the green channel index bands to obtain ASGI remote sensing image tiles; setting attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles, and constructing a sample data set of duckweed-type water bodies; training an initial deep learning model through the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

[0114] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the deep learning-based remote sensing monitoring method for duckweed-type water bodies provided by the above-mentioned methods, the method comprising: obtaining remote sensing tiles of duckweed-type water bodies as RGB three-channel samples; performing calculations based on chromatic angles on the RGB three-channel samples to obtain chromatic angle bands, performing calculations based on slope sums on the RGB three-channel samples to obtain slope sum bands, and performing calculations based on green channel indexes on the RGB three-channel samples to obtain green channel index bands; performing band synthesis on the chromatic angle bands, the slope sum bands, and the green channel index bands to obtain ASGI remote sensing image tiles; setting attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles, and constructing a sample data set of duckweed-type water bodies; training an initial deep learning model through the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

[0115] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0116] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for remote sensing monitoring of duckweed-type water bodies based on deep learning, characterized in that: The method comprises: Obtain remote sensing tiles of duckweed-type water bodies as RGB three-channel samples; Performing a calculation based on the chromaticity angle on the RGB three-channel samples to obtain a chromaticity angle band, performing a calculation based on the slope sum on the RGB three-channel samples to obtain a slope sum band, and performing a calculation based on the green channel index on the RGB three-channel samples to obtain a green channel index band; Performing band synthesis on the chromatic angle band, the slope sum band, and the green channel index band to obtain an ASGI remote sensing image tile; Setting attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles, and constructing a sample dataset of duckweed-type water bodies; An initial deep learning model is trained using the sample data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

2. The method for remote sensing monitoring of duckweed-type water bodies based on deep learning according to claim 1, characterized in that: The calculation based on the chromaticity angle is performed on the RGB three-channel samples to obtain the chromaticity angle band, including: Get the band linear summation coefficient of map remote sensing; Calculate the stimulus values ​​of the R, G, and B values ​​of the RGB three-channel samples according to the linear summation coefficients of the bands to obtain the stimulus values ​​of the XYZ three primary colors; Determining a chromaticity abscissa and a chromaticity ordinate based on the XYZ three primary color stimulus values; A chromaticity angle calculation is performed on the chromaticity horizontal coordinate and the chromaticity vertical coordinate to obtain a chromaticity angle band.

3. The method for remote sensing monitoring of duckweed-type water bodies based on deep learning according to claim 1, characterized in that: The calculation based on the slope sum is performed on the RGB three-channel samples to obtain the slope and band, including: Determine a first ratio of the B value to the G value in the RGB three-channel sample; Determine a second ratio of the R value to the G value in the RGB three-channel sample; The sum of the first ratio and the second ratio is used as the slope and band.

4. The method for remote sensing monitoring of duckweed-type water bodies based on deep learning according to claim 1, characterized in that: The green channel index band is obtained by calculating the green channel index based on the RGB three-channel samples. The calculation includes: Determine the sum of the channel values ​​of the B value and the R value in the RGB three-channel sample; The ratio of the square of the G value in the RGB three-channel sample to the sum of the channel values ​​is used as the green channel index band.

5. The method for remote sensing monitoring of duckweed-type water bodies based on deep learning according to claim 1, characterized in that: The step of setting an attribute threshold for the duckweed-type water body in the ASGI remote sensing image tile and constructing a sample dataset of the duckweed-type water body includes: Determine the surface vector of the duckweed-shaped water body in the ASGI remote sensing image tile; When the duckweed-type water body is a real duckweed-type water body, setting a first attribute threshold for the surface vector of the duckweed-type water body; When the duckweed-type water body is a non-duckweed-type water body, setting a second attribute threshold for the surface vector of the duckweed-type water body; The data labels corresponding to the ASGI remote sensing image tiles are constructed according to the first attribute threshold and the second attribute threshold, and a sample data set of duckweed-type water bodies is obtained.

6. The method for remote sensing monitoring of duckweed-type water bodies based on deep learning according to claim 1, characterized in that: The initial deep learning model is trained by the sample data set to obtain a duckweed-type water body recognition model, including: Inputting the training samples of the sample data set into the initial deep learning model for forward propagation to obtain the predicted attribute values ​​of the training samples for duckweed-type water bodies; Constructing a training loss value based on the difference between the predicted attribute value and the data label of the training sample, and backpropagating the training loss value in the initial deep learning model to update the model parameters of the initial deep learning model; When the training loss value begins to converge or reaches a preset training round, the training of the initial deep learning model is terminated to obtain a duckweed-type water body recognition model.

7. A duckweed-type water body remote sensing monitoring device based on deep learning, characterized in that: The device comprises: The acquisition module is used to obtain remote sensing tiles of duckweed-type water bodies as RGB three-channel samples; A determination module is configured to perform a calculation based on the chromaticity angle on the RGB three-channel samples to obtain a chromaticity angle band, perform a calculation based on the slope sum on the RGB three-channel samples to obtain a slope sum band, and perform a calculation based on the green channel index on the RGB three-channel samples to obtain a green channel index band; A synthesis module, configured to synthesize the chroma angle band, the slope sum band, and the green channel index band to obtain an ASGI remote sensing image tile; A construction module is used to set attribute thresholds according to the duckweed-type water bodies in the ASGI remote sensing image tiles and to construct a training data set of duckweed-type water bodies; The recognition module is used to train the initial deep learning model through the training data set to obtain a duckweed-type water body recognition model to identify duckweed-type water bodies.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the deep learning-based remote sensing monitoring method for duckweed-type water bodies is implemented as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for remote sensing monitoring of duckweed-type water bodies based on deep learning as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for remote sensing monitoring of duckweed-type water bodies based on deep learning as described in any one of claims 1 to 6 is implemented.

Images