CN115170895B - Ocean region classification method and device based on image processing - Google Patents
Ocean region classification method and device based on image processing Download PDFInfo
- Publication number
- CN115170895B CN115170895B CN202211098518.4A CN202211098518A CN115170895B CN 115170895 B CN115170895 B CN 115170895B CN 202211098518 A CN202211098518 A CN 202211098518A CN 115170895 B CN115170895 B CN 115170895B
- Authority
- CN
- China
- Prior art keywords
- ocean
- remote sensing
- sensing image
- training data
- satellite
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Remote Sensing (AREA)
- Astronomy & Astrophysics (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a device for classifying ocean areas based on image processing, and relates to the field of ocean area classification. The ocean region classification method based on image processing comprises the following steps: determining environmental information corresponding to marine remote sensing images acquired one day before the satellite; obtaining a classification result of the ocean regions in the previous day; determining a training data set corresponding to the environmental information from pre-stored training data sets corresponding to different environmental information; training data sets corresponding to different environmental information respectively comprise marine remote sensing image samples and marine area classification results corresponding to the marine remote sensing image samples; updating a training data set according to ocean remote sensing images acquired by a satellite on the previous day and the classification result of ocean areas on the previous day; updating the classification model based on the updated training data set; and determining the current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the updated classification model. The ocean region classification method can improve the accuracy of ocean region classification.
Description
Technical Field
The present invention relates to sea area classification, and more particularly, to a method and apparatus for sea area classification based on image processing.
Background
In the prior art, the remote sensing image acquired by a marine satellite can be used for classifying marine areas and dividing a glacier area and a non-glacier area of the sea.
When classifying the ocean regions, the conventional method needs to use external data (such as a priori data) as input, and then establish a classification model, which can be used for classifying the ocean regions. The method has higher requirement on the accuracy of external data, and the timeliness and the efficiency are reduced by inputting the external data; and the classification model does not consider seasonal changes corresponding to the remote sensing images, so that the precision of the classification model in different seasons and different months is inconsistent, such as: if the classification model is constructed based on data in winter, the accuracy is poor when the classification model is used for ocean area classification of remote sensing images in summer.
Therefore, the existing ocean area classification method has poor accuracy.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art that is already known to a person skilled in the art.
Disclosure of Invention
The invention aims to provide a method and a device for classifying ocean regions based on image processing, which can improve the accuracy of ocean region classification.
In order to achieve the above object, an embodiment of the present invention provides an image processing-based sea region classification method, including: determining environmental information corresponding to marine remote sensing images acquired one day before the satellite; obtaining a classification result of the ocean regions in the previous day; the previous ocean area classification result is a classification result obtained based on ocean remote sensing images and classification models acquired by a satellite on the previous day, and the previous ocean area classification result is used for dividing an iced area and a non-iced area of the ocean on the previous day; determining a training data set corresponding to different environment information from pre-stored training data sets corresponding to the environment information; training data sets corresponding to different environmental information respectively comprise marine remote sensing image samples and marine area classification results corresponding to the marine remote sensing image samples; updating the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the classification result of the marine regions on the previous day; updating the classification model based on the updated training data set; determining a current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the updated classification model; and the current sea area classification result is used for dividing the glacier area and the non-glacier area of the sea in the current day.
In one or more embodiments of the present invention, the environment information includes: at least one of temperature information, season information, climate information, and accident information.
In one or more embodiments of the present invention, the sea area classifying method further includes: acquiring an initial training data set; the initial training data set comprises: the method comprises the steps that a plurality of ocean remote sensing image samples, environment information corresponding to the ocean remote sensing image samples and ocean area classification results corresponding to the ocean remote sensing image samples are obtained; classifying the ocean remote sensing images and ocean area classification results corresponding to the ocean remote sensing image samples according to different environmental information to obtain a classified training data set; judging whether different environment information corresponding to the classified training data set meets preset environment information conditions or not; and if so, storing the classified training data set to obtain the pre-stored training data sets corresponding to different environmental information.
In one or more embodiments of the present invention, the preset environmental information condition includes: the method for classifying the ocean regions comprises the following steps of (1) classifying the ocean regions according to the type and the number of the environmental information and the number of different environmental information corresponding to each environmental information, wherein the method for classifying the ocean regions further comprises the following steps: if not, determining the types of the environment information to be supplemented of the different environment information according to the type quantity of the environment information; determining the to-be-supplemented environmental information of each environmental information type according to the different environmental information quantities corresponding to each environmental information; acquiring a first ocean remote sensing image corresponding to the type of the environmental information to be supplemented and an ocean region classification result corresponding to the first ocean remote sensing image; acquiring a second ocean remote sensing image corresponding to the environmental information to be supplemented and an ocean area classification result corresponding to the second ocean remote sensing image; updating the classified training data set according to the type of the environmental information to be supplemented, the first ocean remote sensing image and the ocean area classification result corresponding to the first ocean remote sensing image, and the second ocean remote sensing image and the ocean area classification result corresponding to the second ocean remote sensing image; and storing the updated classified training data set to obtain the pre-stored training data sets corresponding to different environmental information.
In one or more embodiments of the present invention, the updating the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the previous day marine area classification result includes: determining the similarity between the marine remote sensing image acquired on the previous day of the satellite and a marine remote sensing image sample in a training data set corresponding to the environment information; judging whether the similarity meets a preset similarity condition or not; and if so, adding the marine remote sensing image acquired by the satellite on the previous day and the classification result of the marine area on the previous day into the training data set.
In one or more embodiments of the present invention, the sea area classifying method further includes: if not, feeding back ocean remote sensing images acquired by the satellite on the previous day and the classification result of the ocean areas on the previous day to the user; receiving a corrected ocean region classification result fed back by a user; and adding the marine remote sensing images acquired on the previous day of the satellite and the correction classification result into the training data set.
In one or more embodiments of the present invention, the determining the similarity between the marine remote sensing images acquired on the previous day of the satellite and the marine remote sensing image samples in the training data set corresponding to the environment information includes: determining a first similarity between a marine remote sensing image acquired on the previous day of the satellite and a first marine remote sensing image sample; the ocean area classification result corresponding to the first ocean remote sensing image sample is the same as the ocean area classification result in the previous day; determining a second similarity between a marine remote sensing image acquired on the previous day of the satellite and a second marine remote sensing image sample; the ocean area classification result corresponding to the second ocean remote sensing image sample is different from the ocean area classification result in the previous day; determining final similarity according to the first similarity, the second similarity, a first weight value corresponding to the first similarity and a second weight value corresponding to the second similarity; the sum of the first weight value and the second weight value is 1.
In one or more embodiments of the present invention, the sea area classifying method further includes: determining environment information corresponding to the marine remote sensing image acquired by the satellite on the same day; judging whether the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the same day is the same as the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the previous day; if not, executing the step of determining the environmental information corresponding to the ocean remote sensing image acquired by the satellite in the previous day; and if so, determining the current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the classification model.
In one or more embodiments of the invention, the marine remote sensing image acquired by the satellite on the day before and the marine remote sensing image acquired by the satellite on the day are both preprocessed images; the pretreatment comprises the following steps: at least one processing mode of denoising processing, invalid data processing, enhancement processing and registration processing.
The embodiment of the invention provides an ocean region classification device based on image processing, which comprises: the functional modules are used for realizing the ocean region classification method based on image processing and one or more corresponding embodiments.
Compared with the prior art, in the ocean region classification scheme adopted by the embodiment of the invention, on one hand, the classification model is not fixed, but is updated along with daily ocean remote sensing images and corresponding ocean region classification results, so that the classification model can adapt to seasonal-change remote sensing images. On the other hand, the training data set corresponding to the classification model comprises daily marine remote sensing images and corresponding marine area classification results, and also comprises training data corresponding to the environmental information corresponding to the daily marine remote sensing images, so that the data in the updated training data set corresponding to the classification model is relatively comprehensive, and the precision of the classification model is improved. Therefore, the scheme not only enables the classification model to adapt to seasonal changes, but also improves the precision of the classification model. Furthermore, the accuracy of the ocean area classification result determined based on the classification model and the ocean remote sensing image is improved.
Drawings
FIG. 1 is a flow chart of a method for ocean region classification based on image processing according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an image processing-based sea region classification device according to an embodiment of the present invention;
fig. 3 is a schematic configuration diagram of an image processing apparatus according to an embodiment of the present invention.
Description of the main reference numbers:
200-image processing based sea area classification means, 210-determination module, 220-acquisition module, 230-update module, 300-image processing device, 310-processor, 320-memory.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
The technical scheme provided by the embodiment of the invention can be applied to various application scenes needing ocean area classification, such as: marine monitoring application scenarios, marine analysis application scenarios, and the like. Based on the different application scenarios, the technical solution can be applied to different hardware environments, for example: marine monitoring systems, marine analysis systems, and the like.
In addition, the technical solution adopted by the embodiment of the present invention is based on an image processing technology, and therefore, in any hardware environment, the hardware operating environment corresponding to the technical solution should have an image processing capability, for example: a computer, a dedicated image processing device, etc.
In order to realize the classification of the ocean area, the ocean remote sensing image is processed according to the embodiment of the invention. The ocean remote sensing image is an ocean image acquired by a satellite. Ocean regions can include two major types: glaciers and non-glaciers. By classifying the ocean area, it is convenient to better analyze or monitor the ocean.
As shown in fig. 1, a flowchart of a method for classifying sea areas based on image processing according to an embodiment of the present invention is provided, where the method includes:
step 110: and determining environment information corresponding to the marine remote sensing image acquired by the satellite on the previous day.
Step 120: and obtaining the classification result of the ocean areas in the previous day. The previous ocean area classification result is a classification result obtained based on ocean remote sensing images acquired by the satellite in the previous day and a classification model, and is used for dividing an iced area and a non-iced area of the ocean in the previous day.
Step 130: a training data set corresponding to the environment information is determined from a pre-stored training data set corresponding to different environment information. The training data sets corresponding to different environmental information respectively comprise ocean remote sensing image samples and ocean area classification results corresponding to the ocean remote sensing image samples.
Step 140: and updating the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the classification result of the marine area on the previous day.
Step 150: the classification model is updated based on the updated training data set.
Step 160: and determining the current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the updated classification model. The current-day ocean region classification result is used for dividing the glacier region and the non-glacier region of the current-day ocean.
Compared with the prior art, in the ocean region classification scheme adopted by the embodiment of the invention, on one hand, the classification model is not fixed, but is updated along with daily ocean remote sensing images and corresponding ocean region classification results, so that the classification model can adapt to seasonal-change remote sensing images. On the other hand, the training data set corresponding to the classification model comprises daily marine remote sensing images and corresponding marine area classification results, and also comprises training data corresponding to the environmental information corresponding to the daily marine remote sensing images, so that the data in the updated training data set corresponding to the classification model is relatively comprehensive, and the precision of the classification model is improved. Therefore, the scheme not only enables the classification model to adapt to seasonal changes, but also improves the accuracy of the classification model. Furthermore, the accuracy of the ocean area classification result determined based on the classification model and the ocean remote sensing image is improved.
Before the sea area classification method is introduced, the inventive concept of the sea area classification method is briefly introduced. In the ocean region classification method, firstly, an initial classification model is configured, the initial classification model can be obtained based on external data (prior data) training, and ocean remote sensing images can be subjected to ocean region classification processing. On the basis of the initial classification model, ocean remote sensing images acquired by the satellite every day need to be classified in ocean areas, namely, ocean area classification results are updated every day. During the daily update of the sea area classification results, the initial classification model is also updated. That is, before determining the daily sea area classification result, the previous sea area classification result and the corresponding sea remote sensing image are used to update the classification model, and then the updated classification model is used to determine the daily sea area classification result, which is a continuous process. Moreover, the updating of the classification model not only depends on the actually acquired remote sensing image and the corresponding classification result, but also depends on a pre-stored training data set corresponding to the environmental information.
In step 110, environmental information corresponding to the ocean remote sensing images acquired one day before the satellite is determined.
It can be understood that the remote sensing images acquired by the satellite on the previous day can be images acquired at different time periods, and the corresponding environment information may be different at different time periods. Therefore, after the marine remote sensing image is acquired by the satellite, marine environment information at the acquisition time is also acquired and is used as environment information corresponding to the marine remote sensing image. The marine environment information can be acquired by other marine monitoring equipment; the determination may also be implemented by means of environment information prediction, which is not limited herein.
Furthermore, the marine remote sensing image may be a remote sensing image of a specific marine area to be monitored. When a plurality of areas need to be monitored, marine remote sensing images of the plurality of areas are respectively collected, and then area identification is carried out respectively.
In some embodiments, the environmental information includes: at least one of temperature information, season information, climate information, and accident information.
Temperature information can be understood as air temperature, for example: 10 degrees. The season information may be understood as the current season, for example: in summer. Climate information may be understood as weather conditions, for example: cloudy days, rainy days, etc. Accident information can be understood as whether an accident has occurred, for example: tornadoes, foreign object strikes, and the like.
In some embodiments, the sea region classification method may further include: determining environmental information corresponding to a marine remote sensing image acquired by a satellite on the same day; judging whether the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the same day is the same as the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the previous day; if not, go to step 110; and if so, determining the current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the classification model.
In this embodiment, the acquisition mode of the environmental information corresponding to the marine remote sensing image acquired on the current day by the satellite is the same as the acquisition mode of the environmental information corresponding to the marine remote sensing image acquired on the previous day, and the description thereof will not be repeated.
It can be understood that whether the environmental information corresponding to the marine remote sensing image acquired by the satellite on the current day is the same as the environmental information corresponding to the marine remote sensing image acquired by the satellite on the previous day is judged, if so, the classification model does not need to be updated, and the classification result of the marine area on the current day can be determined directly according to the remote sensing image acquired by the satellite on the current day and the classification model. If not, indicating that the classification model needs to be updated, step 110 and step 160 can be performed.
In the embodiment of the invention, all the remote ocean sensing images can be preprocessed images. The pretreatment mode comprises the following steps: at least one processing mode of denoising processing, invalid data processing, enhancement processing and registration processing.
Specific implementations of these processes are described in the art and will not be described herein. Besides these processing methods, other general processing methods for marine remote sensing images can be used, and are not limited herein.
In step 120, the last day sea area classification result is obtained. In combination with the introduction of the inventive concept, the previous ocean area classification result is a classification result obtained based on ocean remote sensing images and classification models acquired by satellites on the previous day, and the previous ocean area classification result is used for dividing the glacier area and the non-glacier area of the ocean on the previous day.
It is to be understood that the classification model is a preconfigured classification model if the previous day was the first monitoring day. If the previous day is not the first monitoring day, the classification model is also an updated classification model, and the corresponding updating method may refer to the description of the following embodiments, that is, the updating method is the same as the updating method of the classification model of this day.
In conjunction with the description of the aforementioned inventive concept, the preconfigured classification models can be implemented by referring to techniques that are well-established in the art, and will not be described in detail herein.
In an embodiment of the present invention, the classification model may be a random forest model, a neural network model, or the like, and is not limited herein. Correspondingly, when different classification models are adopted, a corresponding classification algorithm and a corresponding model training mode are adopted.
In addition, the ocean area classification result related in the embodiment of the invention is used for dividing the glacier area and the non-glacier area, and compared with the ocean remote sensing image, the glacier area and the non-glacier area can be directly marked on the ocean remote sensing image through identification. Or the glacier area part and the non-glacier area part can be directly split from the ocean remote sensing image. The specific presentation manner of the classification result should be combined with the setting manner of the label in the training data set of the classification model, i.e. the configuration manner of the classification result in the training data set, which is not limited herein.
In step 130, a training data set corresponding to the environment information is determined from among training data sets corresponding to different environment information stored in advance. The training data sets corresponding to different environmental information respectively comprise ocean remote sensing image samples and ocean area classification results corresponding to the ocean remote sensing image samples.
In some embodiments, the configuration process of the training data sets corresponding to different environmental information includes: acquiring an initial training data set; the initial training data set includes: the method comprises the steps that a plurality of ocean remote sensing image samples, environment information corresponding to the ocean remote sensing image samples and ocean area classification results corresponding to the ocean remote sensing image samples are obtained; classifying the ocean remote sensing images and ocean area classification results corresponding to the ocean remote sensing image samples according to different environmental information to obtain a classified training data set; judging whether different environment information corresponding to the classified training data set meets preset environment information conditions or not; if yes, storing the classified training data set to obtain a pre-stored training data set corresponding to different environmental information.
The plurality of ocean remote sensing image samples may be samples obtained from an existing database, or may be historically acquired ocean remote sensing images, which is not limited herein. Correspondingly, the classification result of the ocean region corresponding to the ocean remote sensing images can be a classification result determined by other methods, for example: the classification result determined manually is not limited herein.
When a plurality of ocean remote sensing image samples are classified according to environmental information, several kinds of environmental information are determined, and then the ocean remote sensing image samples corresponding to each environmental information are classified into the same category. For the same kind of environment information, different environment information is also included, for example: although both are temperature environment information, the temperatures are different. Therefore, for the same type of environmental information, the marine remote sensing image samples corresponding to the same environmental information value or content can be classified into the same type according to the specific environmental information value or content.
In some embodiments, the preset environmental information conditions include: the number of types of environment information and the number of different environment information corresponding to each environment information.
And judging whether the type quantity of the environmental information meets the condition or not according to the classified training data set, if so, judging the quantity of different environmental information corresponding to each environmental information, and if so, determining that the environmental information meets the condition. If either or both of the former and the latter are not met, determining that the condition is not met.
In some embodiments, if different pieces of environment information corresponding to the classified training data set do not meet preset environment information conditions, determining types of environment information to be supplemented of the different pieces of environment information according to the type number of the environment information; determining the to-be-supplemented environmental information of each environmental information type according to the different environmental information quantities corresponding to each environmental information; obtaining a first ocean remote sensing image corresponding to the type of the environmental information to be supplemented and an ocean area classification result corresponding to the first ocean remote sensing image; acquiring a second ocean remote sensing image corresponding to environmental information to be supplemented and an ocean area classification result corresponding to the second ocean remote sensing image; updating the classified training data set according to the type of the environment information to be supplemented, the first ocean remote sensing image and the ocean region classification result corresponding to the first ocean remote sensing image, and the second ocean remote sensing image and the ocean region classification result corresponding to the second ocean remote sensing image; and storing the updated classified training data set to obtain a pre-stored training data set corresponding to different environment information.
In the preset environment information condition, the number of kinds of environment information and the number of different environment information in each environment information are defined. Then, when the condition is not met, the environmental information may be supplemented.
For example: if the number of the types of the limited environment information is 4 and the number of the types of the actual environment information is only 3, the type of the environment information to be supplemented is at least one type of environment information. Similarly, if the number of different pieces of environmental information in each environmental information category defined in the environmental information condition is 10 and actually is only 8, the pieces of environmental information to be supplemented are at least 2 pieces of environmental information different from the existing 8 pieces of environmental information.
After determining the type of the environmental information to be supplemented and the environmental information to be supplemented, obtaining a first ocean remote sensing image and an ocean area classification result corresponding to the first remote sensing image. It can be understood that the first ocean remote sensing image is an ocean remote sensing image acquired under the environment information corresponding to the type of the environment information to be supplemented.
And obtaining the second ocean remote sensing image and an ocean area classification result corresponding to the second ocean remote sensing image. It can be understood that the second marine remote sensing image is a marine remote sensing image acquired under the environmental information to be supplemented.
Then, based on the to-be-supplemented environmental information and the to-be-supplemented environmental information type, the ocean region classification results corresponding to the first ocean remote sensing image and the first remote sensing image and the ocean region classification results corresponding to the second ocean remote sensing image and the second ocean remote sensing image are classified and updated into the classified training data set, and then the updated classified training data set can be obtained, so that the pre-stored training data sets corresponding to different environmental information can be obtained.
And determining a training data set matched with the current environmental information based on pre-stored training data sets corresponding to different environmental information. Thus, in step 140, the training data set is updated according to the ocean remote sensing images acquired by the satellite one day before and the ocean area classification result of the satellite one day before.
In some embodiments, step 140 comprises: determining the similarity between the marine remote sensing image acquired on the previous day of the satellite and a marine remote sensing image sample in a training data set corresponding to the environment information; judging whether the similarity meets a preset similarity condition or not; and if so, adding the ocean remote sensing images acquired by the satellite on the previous day and the classification result of the ocean area on the previous day into the training data set.
The method for determining the similarity between the marine remote sensing image acquired one day before the satellite and the marine remote sensing image sample in the training data set corresponding to the environmental information comprises the following steps: determining a first similarity between a marine remote sensing image acquired one day before the satellite and a first marine remote sensing image sample; the ocean area classification result corresponding to the first ocean remote sensing image sample is the same as the ocean area classification result in the previous day; determining a second similarity between the marine remote sensing image acquired one day before the satellite and a second marine remote sensing image sample; the ocean area classification result corresponding to the second ocean remote sensing image sample is different from the ocean area classification result in the previous day; determining final similarity according to the first similarity, the second similarity, a first weight value corresponding to the first similarity and a second weight value corresponding to the second similarity; the sum of the first weight value and the second weight value is 1.
The first similarity and the second similarity can be understood as the similarity between the remote sensing images, and various implementation modes can be adopted, and the calculation modes of the first similarity and the second similarity are only required to be the same.
The first weight value and the second weight value may be set in combination with the influence of different ocean region classification results on the classification result of the ocean remote sensing image acquired in the previous day, which is not limited herein.
The preset similarity condition can be a similarity threshold, and when the similarity is smaller than the similarity threshold, the similarity is considered to meet the preset similarity condition; and when the similarity is greater than or equal to the similarity threshold, the similarity is considered to be not in accordance with a preset similarity condition.
Therefore, if the preset similarity condition is met, the difference between the ocean remote sensing images acquired by the satellite on the previous day and the ocean area classification result on the previous day and the data in the training data set is shown, and the difference can be directly added to the training data set.
In some embodiments, if the preset similarity condition is not met, feeding back ocean remote sensing images acquired in the previous day of the satellite and the classification result of the ocean areas in the previous day to a user; receiving a corrected ocean region classification result fed back by a user; and adding the marine remote sensing images acquired on the previous day of the satellite and the correction classification result into a training data set.
In this embodiment, if the preset similarity condition is not met, it is indicated that the marine remote sensing images acquired on the previous day of the satellite and the classification results of the marine regions on the previous day are not different from the data in the training data set, and at this time, the results can be fed back to the user and corrected by the user.
The corrected ocean region classification result can be the same as the uncorrected ocean region classification result, namely, the user considers that the result does not need to be corrected and can be directly added into the training data set; or may be different from the uncorrected sea area classification result, i.e. the user corrects the result so that it can be added to the training data set.
After updating the training data set, the classification model is updated based on the updated training data set in step 150. That is, the updated training data set is used as the training data of the classification model, and the classification model is trained to obtain a trained model, i.e., an updated classification model.
In step 160, the current-day ocean area classification result is determined according to the ocean remote sensing image acquired by the satellite on the current day and the updated classification model.
In the step, the ocean remote sensing image acquired by the satellite on the day is input into the updated classification model, and the classification result of the ocean area on the day can be output by the updated classification model.
As can be seen from the description of the foregoing embodiment, in the ocean region classification scheme adopted in the embodiment of the present invention, on one hand, the classification model is not fixed, but is updated along with the daily ocean remote sensing image and the corresponding ocean region classification result, so that the classification model can adapt to the seasonal remote sensing image. On the other hand, the training data set corresponding to the classification model comprises daily marine remote sensing images and corresponding marine area classification results, and also comprises training data corresponding to the environmental information corresponding to the daily marine remote sensing images, so that the data in the updated training data set corresponding to the classification model is relatively comprehensive, and the precision of the classification model is improved. Therefore, the scheme not only enables the classification model to adapt to seasonal changes, but also improves the accuracy of the classification model. Furthermore, the accuracy of the ocean region classification result determined based on the classification model and the ocean remote sensing image is improved.
As shown in fig. 2, an embodiment of the present invention further provides an image processing-based sea area classification apparatus 200, including: a determination module 210, an acquisition module 220, and an update module 230.
The determining module 210 is configured to determine environment information corresponding to a marine remote sensing image acquired by a satellite on the previous day; an obtaining module 220, configured to obtain a previous ocean area classification result; the previous ocean area classification result is a classification result obtained based on ocean remote sensing images and classification models acquired by a satellite on the previous day, and the previous ocean area classification result is used for dividing an iced area and a non-iced area of the ocean on the previous day; an updating module 230, configured to determine, from pre-stored training data sets corresponding to different pieces of environment information, a training data set corresponding to the environment information; training data sets corresponding to different environmental information respectively comprise marine remote sensing image samples and marine area classification results corresponding to the marine remote sensing image samples; updating the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the classification result of the marine regions on the previous day; updating the classification model based on the updated training data set; the determining module 210 is further configured to: determining a current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the updated classification model; and the current sea area classification result is used for dividing the glacier area and the non-glacier area of the sea in the current day.
In some embodiments, the obtaining module 220 is further configured to: acquiring an initial training data set; the initial training data set comprises: the method comprises the steps that a plurality of ocean remote sensing image samples, environment information corresponding to the ocean remote sensing image samples and ocean area classification results corresponding to the ocean remote sensing image samples are obtained; classifying the ocean remote sensing images and the ocean area classification results corresponding to the ocean remote sensing image samples according to different environmental information to obtain a classified training data set; judging whether different environment information corresponding to the classified training data set meets preset environment information conditions or not; and if so, storing the classified training data set to obtain the pre-stored training data set corresponding to different environmental information.
In some embodiments, the obtaining module 220 is further configured to: if not, determining the types of the environment information to be supplemented of the different environment information according to the type quantity of the environment information; determining environmental information to be supplemented of each environmental information type according to different environmental information quantities corresponding to each environmental information; acquiring a first ocean remote sensing image corresponding to the type of the environmental information to be supplemented and an ocean region classification result corresponding to the first ocean remote sensing image; acquiring a second ocean remote sensing image corresponding to the environmental information to be supplemented and an ocean area classification result corresponding to the second ocean remote sensing image; updating the classified training data set according to the type of the environmental information to be supplemented, the first ocean remote sensing image and the ocean area classification result corresponding to the first ocean remote sensing image, and the second ocean remote sensing image and the ocean area classification result corresponding to the second ocean remote sensing image; and storing the updated classified training data set to obtain the pre-stored training data sets corresponding to different environmental information.
In some embodiments, the update module 230 is further configured to: determining the similarity between the marine remote sensing image acquired on the previous day of the satellite and a marine remote sensing image sample in a training data set corresponding to the environment information; judging whether the similarity meets a preset similarity condition or not; and if so, adding the marine remote sensing image acquired by the satellite on the previous day and the classification result of the marine area on the previous day into the training data set.
In some embodiments, the update module 230 is further configured to: if not, feeding back ocean remote sensing images acquired by the satellite on the previous day and the classification result of the ocean areas on the previous day to the user; receiving a corrected ocean region classification result fed back by a user; and adding the marine remote sensing images acquired on the previous day of the satellite and the correction classification result into the training data set.
In some embodiments, the determining module 210 is further configured to: determining a first similarity between a marine remote sensing image acquired on the previous day of the satellite and a first marine remote sensing image sample; the ocean area classification result corresponding to the first ocean remote sensing image sample is the same as the ocean area classification result in the previous day; determining a second similarity between a marine remote sensing image acquired on the previous day of the satellite and a second marine remote sensing image sample; the marine area classification result corresponding to the second marine remote sensing image sample is different from the marine area classification result in the previous day; determining final similarity according to the first similarity, the second similarity, a first weight value corresponding to the first similarity and a second weight value corresponding to the second similarity; the sum of the first weight value and the second weight value is 1.
In some embodiments, the determining module 210 is further configured to: determining environment information corresponding to the ocean remote sensing image acquired by the satellite on the same day; judging whether the environmental information corresponding to the marine remote sensing image acquired by the satellite on the same day is the same as the environmental information corresponding to the marine remote sensing image acquired by the satellite on the previous day; if not, executing the step of determining the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the previous day; and if so, determining the current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the classification model.
As shown in fig. 3, an embodiment of the present invention further provides an image processing apparatus 300, which includes a processor 310 and a memory 320, where the processor 310 and the memory 320 are communicatively connected, and the image processing apparatus 300 is used as an execution subject of the foregoing sea area classification method.
The processor 310 and the memory 320 are electrically connected directly or indirectly to realize data transmission or interaction. For example, electrical connections between these components may be made through one or more communication or signal buses. The aforementioned sea area classification methods respectively include at least one software function module that can be stored in the memory 320 in the form of software or firmware (firmware).
The processor 310 may be an integrated circuit chip having signal processing capabilities. The Processor 310 may be a general-purpose Processor including a CPU (Central Processing Unit), an NP (Network Processor), and the like; but may also be a digital signal processor, an application specific integrated circuit, an off-the-shelf programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components. Which may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 320 may store various software programs and modules, such as program instructions/modules corresponding to the image processing method and apparatus provided by the embodiment of the present invention. The processor 310 executes various functional applications and data processing, i.e., implements the method in the embodiment of the present invention, by executing software programs and modules stored in the memory 320.
The Memory 320 may include, but is not limited to, a RAM (Random Access Memory), a ROM (Read Only Memory), a PROM (Programmable Read-Only Memory), an EPROM (Erasable Read-Only Memory), an EEPROM (electrically Erasable Read-Only Memory), and the like.
It is to be understood that the configuration shown in fig. 3 is merely illustrative, and that the image processing apparatus 300 may include more or fewer components than shown in fig. 3, or have a different configuration than shown in fig. 3.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a computer, the method for classifying an ocean region according to any one of the above embodiments is executed.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of specific exemplary embodiments of the invention has been presented for the purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (7)
1. A method for classifying ocean regions based on image processing is characterized by comprising the following steps:
determining environmental information corresponding to marine remote sensing images acquired one day before the satellite;
obtaining a classification result of the ocean regions in the previous day; the previous ocean area classification result is a classification result obtained based on ocean remote sensing images and classification models acquired by the satellite on the previous day, and the previous ocean area classification result is used for dividing an glacier area and a non-glacier area of the ocean on the previous day;
determining a training data set corresponding to different environment information from pre-stored training data sets corresponding to the environment information; training data sets corresponding to different environmental information respectively comprise marine remote sensing image samples and marine area classification results corresponding to the marine remote sensing image samples;
updating the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the classification result of the marine area on the previous day;
updating the classification model based on the updated training data set; and
determining a current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite in the current day and the updated classification model, wherein the current-day ocean area classification result is used for dividing a glacier area and a non-glacier area of the ocean in the current day;
the updating of the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the marine regional classification results on the previous day comprises the following steps: determining the similarity between the marine remote sensing image acquired on the previous day of the satellite and a marine remote sensing image sample in a training data set corresponding to the environment information; judging whether the similarity meets a preset similarity condition or not; if yes, adding the ocean remote sensing image acquired by the satellite on the previous day and the classification result of the ocean area on the previous day into the training data set; if not, feeding back ocean remote sensing images acquired by the satellite on the previous day and the classification result of the ocean areas on the previous day to the user; receiving a corrected ocean region classification result fed back by a user; adding the ocean remote sensing images acquired on the previous day of the satellite and the corrected ocean area classification results into the training data set;
the determining the similarity between the marine remote sensing image acquired on the previous day of the satellite and the marine remote sensing image sample in the training data set corresponding to the environment information comprises the following steps: determining a first similarity between a marine remote sensing image acquired on the previous day of the satellite and a first marine remote sensing image sample; the ocean area classification result corresponding to the first ocean remote sensing image sample is the same as the ocean area classification result in the previous day; determining a second similarity between a marine remote sensing image acquired on the previous day of the satellite and a second marine remote sensing image sample; the ocean area classification result corresponding to the second ocean remote sensing image sample is different from the ocean area classification result in the previous day; determining final similarity according to the first similarity, the second similarity, a first weight value corresponding to the first similarity and a second weight value corresponding to the second similarity; the sum of the first weight value and the second weight value is 1.
2. The sea area classification method according to claim 1, wherein the environmental information includes:
at least one of temperature information, season information, climate information, and accident information.
3. The sea area classification method according to claim 1 or 2, characterized in that the sea area classification method further comprises:
acquiring an initial training data set; the initial training data set comprises: the method comprises the steps that a plurality of ocean remote sensing image samples, environment information corresponding to the ocean remote sensing image samples and ocean area classification results corresponding to the ocean remote sensing image samples are obtained;
classifying the ocean remote sensing images and ocean area classification results corresponding to the ocean remote sensing image samples according to different environmental information to obtain a classified training data set;
judging whether different environment information corresponding to the classified training data set meets preset environment information conditions or not;
and if so, storing the classified training data set to obtain the pre-stored training data sets corresponding to different environmental information.
4. The sea area classification method according to claim 3, wherein the preset environmental information conditions include: the method for classifying the ocean regions comprises the following steps of (1) classifying the ocean regions, wherein the type number of the environment information and the number of different environment information corresponding to each environment information are as follows:
if not, determining the types of the environment information to be supplemented of the different environment information according to the type quantity of the environment information;
determining environmental information to be supplemented of each environmental information type according to different environmental information quantities corresponding to each environmental information;
acquiring a first ocean remote sensing image corresponding to the type of the environmental information to be supplemented and an ocean area classification result corresponding to the first ocean remote sensing image;
acquiring a second ocean remote sensing image corresponding to the environmental information to be supplemented and an ocean area classification result corresponding to the second ocean remote sensing image;
updating the classified training data set according to the type of the environmental information to be supplemented, the first ocean remote sensing image and the ocean area classification result corresponding to the first ocean remote sensing image, and the second ocean remote sensing image and the ocean area classification result corresponding to the second ocean remote sensing image;
and storing the updated classified training data set to obtain the pre-stored training data sets corresponding to different environmental information.
5. The sea area classifying method according to claim 1, further comprising:
determining environment information corresponding to the marine remote sensing image acquired by the satellite on the same day;
judging whether the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the same day is the same as the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the previous day;
if not, executing the step of determining the environmental information corresponding to the ocean remote sensing image acquired by the satellite on the previous day;
and if so, determining the current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the classification model.
6. The sea area classification method according to claim 1, wherein the sea remote sensing images acquired on the previous day by the satellite and the sea remote sensing images acquired on the current day by the satellite are both preprocessed images; the pretreatment comprises the following steps: at least one processing mode of denoising processing, invalid data processing, enhancement processing and registration processing.
7. An ocean region classification device based on image processing is characterized by comprising:
the determining module is used for determining environment information corresponding to the ocean remote sensing image acquired by the satellite in the previous day;
the acquisition module is used for acquiring the classification result of the ocean area in the previous day; the previous ocean area classification result is a classification result obtained based on ocean remote sensing images and classification models acquired by satellites in the previous day, and the previous ocean area classification result is used for dividing glacier areas and non-glacier areas of oceans in the previous day;
the updating module is used for determining a training data set corresponding to the environmental information from pre-stored training data sets corresponding to different environmental information; training data sets corresponding to different environmental information respectively comprise marine remote sensing image samples and marine area classification results corresponding to the marine remote sensing image samples; updating the training data set according to the marine remote sensing images acquired by the satellite on the previous day and the classification result of the marine area on the previous day; updating the classification model based on the updated training data set;
the determination module is further configured to: determining a current-day ocean area classification result according to the ocean remote sensing image acquired by the satellite on the current day and the updated classification model; the current sea area classification result is used for dividing a glacier area and a non-glacier area of the sea in the current day;
the update module is further to: determining the similarity between the marine remote sensing image acquired on the previous day of the satellite and a marine remote sensing image sample in a training data set corresponding to the environment information; judging whether the similarity meets a preset similarity condition or not; if yes, adding the ocean remote sensing image acquired by the satellite on the previous day and the classification result of the ocean area on the previous day into the training data set; if not, feeding back ocean remote sensing images acquired by the satellite on the previous day and the classification result of the ocean areas on the previous day to the user; receiving a corrected ocean region classification result fed back by a user; adding the ocean remote sensing images acquired on the previous day of the satellite and the corrected ocean area classification results into the training data set;
determining a first similarity between a marine remote sensing image acquired one day before the satellite and a first marine remote sensing image sample; the ocean area classification result corresponding to the first ocean remote sensing image sample is the same as the ocean area classification result in the previous day; determining a second similarity between a marine remote sensing image acquired on the previous day of the satellite and a second marine remote sensing image sample; the ocean area classification result corresponding to the second ocean remote sensing image sample is different from the ocean area classification result in the previous day; determining final similarity according to the first similarity, the second similarity, a first weight value corresponding to the first similarity and a second weight value corresponding to the second similarity; the sum of the first weight value and the second weight value is 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211098518.4A CN115170895B (en) | 2022-09-09 | 2022-09-09 | Ocean region classification method and device based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211098518.4A CN115170895B (en) | 2022-09-09 | 2022-09-09 | Ocean region classification method and device based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115170895A CN115170895A (en) | 2022-10-11 |
| CN115170895B true CN115170895B (en) | 2022-11-22 |
Family
ID=83482381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211098518.4A Active CN115170895B (en) | 2022-09-09 | 2022-09-09 | Ocean region classification method and device based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115170895B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036474A (en) * | 2014-06-12 | 2014-09-10 | 厦门美图之家科技有限公司 | Automatic adjustment method for image brightness and contrast |
| CN111652289A (en) * | 2020-05-15 | 2020-09-11 | 中国科学院空天信息创新研究院 | Sea ice and seawater segmentation method for synthetic aperture radar image |
| CN112052744A (en) * | 2020-08-12 | 2020-12-08 | 成都佳华物链云科技有限公司 | Environment detection model training method, environment detection method and device |
| CN112966656A (en) * | 2021-03-29 | 2021-06-15 | 国家卫星海洋应用中心 | Data processing method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110929592A (en) * | 2019-11-06 | 2020-03-27 | 北京恒达时讯科技股份有限公司 | Extraction method and system for outer boundary of mariculture area |
| US11113580B2 (en) * | 2019-12-30 | 2021-09-07 | Industrial Technology Research Institute | Image classification system and method |
-
2022
- 2022-09-09 CN CN202211098518.4A patent/CN115170895B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036474A (en) * | 2014-06-12 | 2014-09-10 | 厦门美图之家科技有限公司 | Automatic adjustment method for image brightness and contrast |
| CN111652289A (en) * | 2020-05-15 | 2020-09-11 | 中国科学院空天信息创新研究院 | Sea ice and seawater segmentation method for synthetic aperture radar image |
| CN112052744A (en) * | 2020-08-12 | 2020-12-08 | 成都佳华物链云科技有限公司 | Environment detection model training method, environment detection method and device |
| CN112966656A (en) * | 2021-03-29 | 2021-06-15 | 国家卫星海洋应用中心 | Data processing method and device |
Non-Patent Citations (2)
| Title |
|---|
| A comparison of classification techniques for glacier change detection using multispectral images;Rahul Nijhawan et.al;《Perspectives in Science》;20160930;第8卷;第377-380页 * |
| 基于遥感和GIS的我国季风海洋型冰川区冰碛物覆盖型冰川边界的自动识别;宋波等;《冰川冻土》;20070615(第03期);第128-134页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115170895A (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113884961B (en) | SOC calibration method, modeling device, computer equipment and medium | |
| CN109063116B (en) | Data identification method and device, electronic equipment and computer readable storage medium | |
| CN111814835A (en) | Training method and device of computer vision model, electronic equipment and storage medium | |
| CN113255792B (en) | Data anomaly point detection method, device, system and storage medium | |
| CN110705718A (en) | Model interpretation method and device based on cooperative game and electronic equipment | |
| CN110705573A (en) | Automatic modeling method and device of target detection model | |
| CN114414935A (en) | Automatic positioning method and system for feeder fault area of power distribution network based on big data | |
| CN114280276A (en) | Agricultural monitoring system and method | |
| CN110866682B (en) | Underground cable early warning method and device based on historical data | |
| CN115170895B (en) | Ocean region classification method and device based on image processing | |
| CN113723467A (en) | Sample collection method, device and equipment for defect detection | |
| CN112326882B (en) | Air quality sensor processing method and device | |
| CN116821729B (en) | Data processing system for driving measurement calibration | |
| CN112633074A (en) | Pedestrian information detection method and device, storage medium and electronic equipment | |
| CN117083621A (en) | Detector training method, device and storage medium | |
| CN115222145A (en) | Driving range prediction method and system based on new energy automobile operation big data | |
| CN114896134A (en) | Metamorphic test method, device and equipment for target detection model | |
| CN110942179A (en) | Automatic driving route planning method and device and vehicle | |
| CN111507495A (en) | Method and device for predicting missing wind measurement data | |
| CN113505642B (en) | Method, device, equipment and storage medium for improving target re-identification generalization | |
| CN116958838B (en) | Forest resource monitoring method and system based on unmanned aerial vehicle aerial survey technology | |
| CN116704008B (en) | Method and device for judging object based on picture area calculation and application of method and device | |
| CN111369054B (en) | WRF-based high-altitude meteorological element forecasting method and device | |
| US20220092385A1 (en) | Information processing device and information processing system | |
| CN117493957A (en) | Crop identification method and system for agricultural irrigation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |