CN115393884B - Method, device and system for extracting and processing chart thematic information - Google Patents
Method, device and system for extracting and processing chart thematic information Download PDFInfo
- Publication number
- CN115393884B CN115393884B CN202211069968.0A CN202211069968A CN115393884B CN 115393884 B CN115393884 B CN 115393884B CN 202211069968 A CN202211069968 A CN 202211069968A CN 115393884 B CN115393884 B CN 115393884B
- Authority
- CN
- China
- Prior art keywords
- chart
- preset
- data set
- result
- information
- 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
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/42—Document-oriented image-based pattern recognition based on the type of document
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
- G06V30/19147—Obtaining 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
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
- G06V30/19173—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a device and a system for extracting and processing chart thematic information. The method, the device and the system realize automation of sea map information extraction, thereby improving the accuracy and the efficiency of automatic sea map information extraction; furthermore, the extraction processing method, the device and the system for the sea chart thematic information further improve the accuracy and the efficiency of automatic extraction of the sea map information by processing each pixel in each image layer in the sea chart to be processed.
Description
Technical Field
The present invention relates to the field of extraction processing technology of chart thematic information, and in particular, to a method, an apparatus, a computer readable storage medium and a system for extracting and processing chart thematic information.
Background
The sea chart thematic information refers to certain information represented in the sea chart, such as an intertidal zone, a 2m isodepth line, a 5m isodepth line, a land area range, a sea area range, and the like. A sea chart is a kind of map, also called a sea map, which is a map in which the sea and its adjacent land are the subjects of investigation. The sea chart is a main result of ocean mapping, ocean investigation and research, is also a space model of an ocean area, a carrier and a transmission tool of ocean information, is an analysis basis of ocean geographic environment characteristics, and is more important data of ocean development and utilization. Has important use value in various fields of navigation, fishery, ocean engineering construction, ocean demarcation, historical research, ocean military and ocean science research, and ocean development and utilization. The chart is an important achievement in the field of ocean science, consumes a great deal of manpower and financial resources to be made, expresses and records a great deal of information, and is a precious financial resource left by the former people. For decades, a large number of paper sea charts have accumulated in the marine world. However, the paper chart-expressed information cannot be used directly for digital analysis. Therefore, the specific information expressed in the chart is digitally extracted, particularly vectorized, and has high practical value in the information age under the background of big data. The paper chart is digitized or vectorized, and is mostly conventional manual vectorization, which requires huge manpower and financial resources, and is complicated in operation and time-consuming. The digital information has different degrees of fine deviation due to different degrees of fine manual operations. The final quality of the information extraction is also affected by the artificial mishandling thereof. The historical sea chart has important research value, but the early paper sea chart is stored for a long time, and the original standard color matching of various ground objects is old. The basic properties of colors such as hue, saturation, brightness, etc. show deviations, and the paper chart is also distorted after being scanned into an electronic plate, and noise points are accompanied. The color distortion phenomenon and noise points can randomly appear at different positions, and the influence of the color distortion phenomenon and the noise points cannot be completely eliminated or ignored by some filtering or some algorithm, so that the problems can be generally summarized into color distortion, and the color distortion can increase the difficulty of automatic identification and extraction. In the chart drawing, the chart produced in different periods has slight deviation in color expression for the same information, which also increases the difficulty of automatic recognition and extraction.
In the prior art, filtering, color-gathering classification, binary image segmentation and enhanced conversion of color space are generally adopted to automatically extract ocean map information.
However, the prior art still has the following drawbacks: the preparation operation is complicated and difficult to understand, and the understanding difficulty of non-professional staff is increased; when thicker line elements or thinner surface elements in the sea chart are automatically extracted and identified, confusion extraction or missing extraction can occur, and the identification effect is not ideal; causing a small amount of raster information to be misclassified and scattered throughout the chart (similar to a pepper surface scattering throughout the chart).
Accordingly, there is a need for a method, apparatus, computer-readable storage medium, and system for extracting and processing chart thematic information, which overcome the above-mentioned drawbacks of the prior art.
Disclosure of Invention
The embodiment of the invention provides a method, a device, a computer-readable storage medium and a system for extracting and processing chart thematic information, so that the accuracy and the efficiency of automatic extraction of ocean map information are improved.
An embodiment of the present invention provides a method for extracting and processing chart thematic information, where the method includes: acquiring a chart to be processed, preprocessing the chart to be processed according to a preset image preprocessing method, and acquiring a characteristic data set; extracting an identification model and a preset result requirement according to the characteristic data set and preset chart information, and identifying and determining an identification result; and correcting the identification result according to a preset grid information correction method to obtain a chart extraction result.
As an improvement of the above solution, the correcting the identification result according to a preset grid information correction method to obtain a chart extraction result specifically includes: performing grid division on the identification result according to a preset unit division size to obtain a plurality of grid units; respectively carrying out weight assignment on the central grid and the non-central grid according to a preset weight assignment method so as to obtain the weight of the central grid and the weight of the non-central grid; the grid unit comprises a central grid and a non-central grid; respectively carrying out weighted proportion calculation on the intertidal zone class and the non-intertidal zone class according to the class of the central grid, the class of the non-central grid, the central grid weight and the non-central grid weight to obtain the intertidal zone class proportion and the non-intertidal zone class proportion; determining a correction type of the central grid according to the inter-tidal zone category specific gravity and the non-inter-tidal zone category specific gravity; and outputting a chart extraction result according to the correction type of each center grid.
As an improvement of the above solution, the method for extracting the identification model and the preset result requirement according to the feature data set and the preset chart information, and identifying and determining the identification result specifically includes: inputting the characteristic data set into a preset chart information extraction and identification model to obtain a first identification result file and a second identification result file; and acquiring a recognition result according to a preset result requirement, the first recognition result file and the second recognition result file.
As an improvement of the above scheme, preprocessing the chart to be processed according to a preset image preprocessing method to obtain a feature data set, specifically including: and acquiring the gray value of each pixel in each layer of the sea chart to be processed, and taking the gray value as a characteristic data set of the sea chart to be processed.
As an improvement of the above solution, the obtaining the identification result according to the preset result requirement, the first identification result file and the second identification result file specifically includes: taking the first identification result file as a first identification result; dividing the second identification result file into a category of the intertidal zone and a category of the non-intertidal zone according to a preset dividing limit, and taking the divided content as a second identification result; threshold division is carried out on the second identification result file according to a preset interval threshold, and the divided content is used as a third identification result; and selecting and determining the identification result from the first identification result, the second identification result and the third identification result according to a preset result requirement.
As an improvement of the above solution, before obtaining the chart to be processed and preprocessing the chart to be processed according to a preset image preprocessing method, the extracting and processing method further includes: acquiring a training chart for training and corresponding chart information, dividing the chart information into intertidal zone type information and non-intertidal zone type information, preprocessing the training chart according to a preset image preprocessing method, and acquiring a training characteristic data set; screening the intertidal zone samples from the intertidal zone category information according to a preset identification sample screening method, and screening the non-intertidal zone samples from the non-intertidal zone category information according to a preset reverse sample screening method; establishing a first training data set and a first verification data set according to the intertidal zone sample and the non-intertidal zone sample respectively; assimilating the first training data set and the first verification data set according to the image information of the training chart, and correspondingly obtaining the training data set and the verification data set; and training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model.
As an improvement of the above scheme, training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model, which specifically includes: according to the training data set and the training feature data set, training and debugging a preset random forest model to obtain a first chart information extraction model; according to the verification data set and the training feature data set, verifying the first chart information extraction model, calculating model precision and Kappa coefficient, and judging whether the model precision and the Kappa coefficient meet preset model output requirements; repeating the steps when the model precision and the Kappa coefficient do not meet the preset model output requirement; and outputting the first chart information extraction model as a chart information extraction and identification model when the model precision and the Kappa coefficient meet preset model output requirements.
The invention further provides an extraction processing device for the chart thematic information, which comprises a feature acquisition unit, an extraction recognition unit and a result correction unit, wherein the feature acquisition unit is used for acquiring a chart to be processed, preprocessing the chart to be processed according to a preset image preprocessing method, and acquiring a feature data set; the extraction and identification unit is used for extracting an identification model and a preset result requirement according to the characteristic data set, preset chart information, and identifying and determining an identification result; the result correction unit is used for correcting the identification result according to a preset grid information correction method so as to obtain a chart extraction result.
As an improvement of the above-described scheme, the result correction unit is further configured to: performing grid division on the identification result according to a preset unit division size to obtain a plurality of grid units; respectively carrying out weight assignment on the central grid and the non-central grid according to a preset weight assignment method so as to obtain the weight of the central grid and the weight of the non-central grid; the grid unit comprises a central grid and a non-central grid; respectively carrying out weighted proportion calculation on the intertidal zone class and the non-intertidal zone class according to the class of the central grid, the class of the non-central grid, the central grid weight and the non-central grid weight to obtain the intertidal zone class proportion and the non-intertidal zone class proportion; determining a correction type of the central grid according to the inter-tidal zone category specific gravity and the non-inter-tidal zone category specific gravity; and outputting a chart extraction result according to the correction type of each center grid.
As an improvement of the above-described scheme, the extraction and recognition unit is further configured to: inputting the characteristic data set into a preset chart information extraction and identification model to obtain a first identification result file and a second identification result file; and acquiring a recognition result according to a preset result requirement, the first recognition result file and the second recognition result file.
As an improvement of the above-described aspect, the feature acquisition unit is configured to: and acquiring the gray value of each pixel in each layer of the sea chart to be processed, and taking the gray value as a characteristic data set of the sea chart to be processed.
As an improvement of the above-described scheme, the extraction and recognition unit is further configured to: taking the first identification result file as a first identification result; dividing the second identification result file into a category of the intertidal zone and a category of the non-intertidal zone according to a preset dividing limit, and taking the divided content as a second identification result; threshold division is carried out on the second identification result file according to a preset interval threshold, and the divided content is used as a third identification result; and selecting and determining the identification result from the first identification result, the second identification result and the third identification result according to a preset result requirement.
As an improvement of the above-described aspect, the extraction processing apparatus further includes a model training unit for: acquiring a training chart for training and corresponding chart information, dividing the chart information into intertidal zone type information and non-intertidal zone type information, preprocessing the training chart according to a preset image preprocessing method, and acquiring a training characteristic data set; screening the intertidal zone samples from the intertidal zone category information according to a preset identification sample screening method, and screening the non-intertidal zone samples from the non-intertidal zone category information according to a preset reverse sample screening method; establishing a first training data set and a first verification data set according to the intertidal zone sample and the non-intertidal zone sample respectively; assimilating the first training data set and the first verification data set according to the image information of the training chart, and correspondingly obtaining the training data set and the verification data set; and training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model.
As an improvement to the above, the model training unit is further configured to: according to the training data set and the training feature data set, training and debugging a preset random forest model to obtain a first chart information extraction model; according to the verification data set and the training feature data set, verifying the first chart information extraction model, calculating model precision and Kappa coefficient, and judging whether the model precision and the Kappa coefficient meet preset model output requirements; repeating the steps when the model precision and the Kappa coefficient do not meet the preset model output requirement; and outputting the first chart information extraction model as a chart information extraction and identification model when the model precision and the Kappa coefficient meet preset model output requirements.
Another embodiment of the present invention provides a computer readable storage medium, where the computer readable storage medium includes a stored computer program, and when the computer program runs, controls a device where the computer readable storage medium is located to execute a method for extracting the chart topic information as described above.
Another embodiment of the present invention provides an extraction processing system for sea chart topic information, the extraction processing system including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the method for extracting sea chart topic information as described above when executing the computer program.
Compared with the prior art, the technical scheme of the invention has at least one of the following beneficial effects:
the invention provides a method, a device, a computer readable storage medium and a system for extracting and processing chart thematic information, which automatically identify and extract chart information through a preset chart information extraction and identification model to acquire an identification result and further correct the identification result.
Further, the method, the device, the computer-readable storage medium and the system for extracting and processing the sea chart thematic information further improve the accuracy and the efficiency of automatic sea map information extraction by processing the layer gray value of the sea chart to be processed.
Drawings
FIG. 1 is a flow chart of a method for extracting and processing chart thematic information according to an embodiment of the invention;
fig. 2 is a schematic structural diagram of a device for extracting and processing chart thematic information according to an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Detailed description of the preferred embodiments
The embodiment of the invention firstly describes a method for extracting and processing chart thematic information. Fig. 1 is a flowchart illustrating a method for extracting and processing chart thematic information according to an embodiment of the invention.
As shown in fig. 1, the extraction processing method includes:
s1, acquiring a chart to be processed, and preprocessing the chart to be processed according to a preset image preprocessing method to acquire a characteristic data set.
To realize automatic extraction of the chart thematic information, firstly, digital processing is carried out on the chart, specifically, when the chart to be extracted is a paper chart, the paper chart is scanned to obtain an electronic chart, and further, geographic registration is carried out on the electronic chart to obtain the electronic chart with an accurate geographic space position (described as a 'chart to be processed' in the text). By carrying out geographic registration and then identifying and extracting, the result after identifying and extracting can directly have accurate geographic space position, so that the result after identifying and extracting is convenient for calculation, statistics and other works, and error accumulation caused by identifying and extracting and then geographic registration is avoided. The corresponding feature dataset may then be extracted for subsequent identification based on the registered electronic version of the chart (described herein as a "chart to be processed"). Specifically, the preprocessed chart has three layers of RGB, each pixel in each layer has a different gray value, the gray value of each pixel in each layer represents the information (plays an extremely important role in classification and extraction), and the gray value of each pixel in each layer is obtained and is used as a characteristic data set based on the gray value.
That is, in one embodiment, preprocessing the chart to be processed according to a preset image preprocessing method to obtain a feature data set, which specifically includes: and acquiring the gray value of each pixel in each layer of the sea chart to be processed, and taking the gray value as a characteristic data set of the sea chart to be processed.
Based on the operation processing of the gray values, the three layers of RGB of the pretreated chart do not need to be subjected to redundant operation processing. Compared with the prior art, the method does not need to carry out the processes of filtering, color convergence sorting, binary image segmentation, enhancement conversion of color space and the like, reduces the complexity of preparation work and enables non-professional staff to understand the method more easily. Based on the gray value operation processing, information loss, data redundancy and the like caused during conversion, enhancement and other processing are avoided, the data calculation amount of a computer is reduced, and the efficiency and the accuracy are improved. Based on the gray value operation processing, the gray value of each pixel in each image layer is taken as a processing object, so that the fine degree of recognition and extraction can be increased, and the fine deviation caused by manpower is avoided. The gray value of each pixel in each image layer is used as a processing object, so that the influence of color distortion and color deviation caused by individual pixels on the accuracy of the overall identification extraction is eliminated. When the gray value of each pixel in each image layer is used as a processing object and the information category consisting of thicker lines or thinner surfaces in the preprocessed chart is identified and extracted, confusion extraction or missing extraction can be effectively avoided, and the identification and extraction effect is optimal.
After the characteristic data set is acquired from the sea chart to be processed, the characteristic data set can be extracted and output according to a preset sea chart information extraction and identification model. In order to achieve the object, firstly, training a chart information extraction and identification model, specifically, constructing a training data set and a verification data set according to a training chart for training, then acquiring a characteristic data set of the training chart, then, adopting a random forest model as a basic model of the chart information extraction and identification model, carrying out parameter debugging, model training and model verification on the random forest model according to the training data set, the verification data set and the characteristic data set of the training chart, and outputting the model as the chart information extraction and identification model when the model precision and Kappa coefficient reach certain standards. The training chart for training is obtained through electronization and geographic registration. In one embodiment, the random forest model may be replaced with a maximum likelihood model, a support vector machine model, a logistic regression model, a deep learning model, or the like. When the data amount is too large, a maximum likelihood model may be preferable to reduce the calculation time of the computer while also ensuring accuracy.
Before constructing the training dataset, the training chart (default to an electronic chart, if not, conversion as described previously) needs to be subjected to intertidal zone information identification extraction, specifically, the information expressed in the training chart is divided into two categories: the intertidal zone category (information that needs to be extracted) and the non-intertidal zone category (other information). The adoption of the division mode is to distinguish two types of information to the greatest extent so as to achieve the optimal recognition and extraction effect. The larger the mutual difference between the two kinds of information is, the easier the information is to be identified, and the influence of color distortion and color deviation on the accuracy of identification and extraction is weakened, so that the identification and extraction effect is optimal.
Sample selection is carried out on the types of the intertidal zones (information to be extracted) in the training chart, the selected intertidal zone samples are uniformly and randomly distributed in the pretreated chart, so that the selected intertidal zone samples have optimal representativeness, and the accuracy of subsequent recognition and extraction can be improved. Especially, the positions of the color distortion and the color deviation of the intertidal zone in the pretreated sea chart are required to be subjected to important sample selection, so that the influence of the color distortion and the color deviation on the recognition and extraction precision can be minimized. The selected intertidal zone samples are classified into the intertidal zone categories, and the corresponding assignment of the intertidal zone categories is 1 for the convenience of understanding.
Sample selection is performed on the non-intertidal zone categories (other information) in the training chart, and the selected non-intertidal zone samples comprise all the categories except the intertidal zone categories as far as possible. The more comprehensive the non-intertidal zone class sample is selected, the larger the mutual difference between the intertidal zone class and the non-intertidal zone class is, the easier the identification is, the influence of color distortion and color deviation on the identification extraction precision is weakened, and the identification extraction effect is optimal. The non-intertidal zone sample selection is also uniformly and randomly distributed in the pretreated chart so as to ensure that the non-intertidal zone sample has optimal representativeness. For ease of understanding, the corresponding assignment of non-intertidal zone information is "2".
And creating a training data set by using the samples of the two types selected by the method. The training data sets are uniformly and randomly distributed in the pretreated chart to achieve the best representativeness. The total area of the training data set is more than 5 percent, preferably 5 to 10 percent, of the total area of the pretreated chart, and the value range can ensure the recognition accuracy of the model, reduce the manual operation and also reduce the complexity of model calculation and the hardware requirement of a computer. When the area of the pretreated chart is very large, the selection area of the training set can be properly reduced.
In creating the validation dataset, sample selection is made of the intertidal and non-intertidal classes in the pre-processed chart. The value corresponding to the category of the intertidal zone is assigned to be "1", and the value corresponding to the category of the non-intertidal zone is assigned to be "2". And creating a verification data set by using the samples of the two types selected above. The verification data set is not required to be repeated with the training data set, so that the independence of the verification data set can be ensured, and the model can express accurate precision information when the verification data set verifies the model. The amount of data in the validation data set should be less than the amount of data in the training data set and the amount of data in the validation data set should be 40% -50%, preferably 43% of the amount of data in the training data set. The verification data set is selected in this way, so that the accuracy of model verification can be guaranteed, manual operation is reduced, and the complexity of model calculation and the hardware requirement of a computer are reduced.
After creating the training dataset and the validation dataset, the training dataset and the validation dataset are processed to have the same spatial size and resolution as the training chart, i.e. the same number of rows and columns (also described herein as "assimilation process"). In this way, the training data set and the verification data set can correspond to the correct category information on the preprocessed chart, and systematic errors generated by the model can be reduced in further operation.
When random forest model parameter debugging is carried out, the first parameter is as follows: the Number of decision trees (Number of Tree), which depends on the complexity of the data. The method can select 300 stable models, preferably 300-500 models, so that the stability of the recognition accuracy of the models can be ensured, the phenomenon of over-fitting of the models is avoided, and the complexity of model data calculation and the hardware requirement of a computer are reduced. However, in view of the change of the sea chart data amount, the test can be carried out by setting the sea chart data amount to 100, 300, 500 and 1000, so that the precision of the model under different conditions can be stabilized to obtain the optimal effect of the model; the second parameter: the number of features randomly selected on each node is selected as the square root of the number of all features. The decision trees in the random forest can be different from each other and independent of each other, the diversity of the decision trees is increased, factors with linear or nonlinear relations can be accurately classified, meanwhile, over-fitting is avoided, classification performance is improved, and model accuracy is higher; third parameter: stopping criteria (for node splitting), selecting a default value of 1 for the minimum number of samples in the node and calculating the minimum unreliability based on a coefficient of 0. The selection can enable the model to select the optimal characteristic division points to generate a decision tree, and further the classification result is more accurate. And verifying the trained model based on the characteristic data set and the verification data set, wherein when the overall accuracy of the model is more than 90% and the Kappa coefficient is more than 0.7, the model accuracy is reasonable and very reliable. Through optimizing and selecting the training data set according to the requirements and combining with the model parameter debugging part, the overall accuracy and Kappa coefficient of the model can be further improved, so that a more accurate and reliable recognition result can be obtained.
In one embodiment, before obtaining the chart to be processed and preprocessing the chart to be processed according to a preset image preprocessing method, the extracting and processing method further includes: acquiring a training chart for training and corresponding chart information, dividing the chart information into intertidal zone type information and non-intertidal zone type information, preprocessing the training chart according to a preset image preprocessing method, and acquiring a training characteristic data set; screening the intertidal zone samples from the intertidal zone category information according to a preset identification sample screening method, and screening the non-intertidal zone samples from the non-intertidal zone category information according to a preset reverse sample screening method; establishing a first training data set and a first verification data set according to the intertidal zone sample and the non-intertidal zone sample respectively; assimilating the first training data set and the first verification data set according to the image information of the training chart, and correspondingly obtaining the training data set and the verification data set; and training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model. Wherein the assimilation process causes the first training data set and the first verification data set to have the same number of rows and columns.
In one embodiment, training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model, which specifically includes: according to the training data set and the training feature data set, training and debugging a preset random forest model to obtain a first chart information extraction model; according to the verification data set and the training feature data set, verifying the first chart information extraction model, calculating model precision and Kappa coefficient, and judging whether the model precision and the Kappa coefficient meet preset model output requirements; repeating the steps when the model precision and the Kappa coefficient do not meet the preset model output requirement; and outputting the first chart information extraction model as a chart information extraction and identification model when the model precision and the Kappa coefficient meet preset model output requirements.
And S2, extracting an identification model and a preset result requirement according to the characteristic data set and preset chart information, and identifying and determining an identification result.
In one embodiment, the identifying and determining the identifying result according to the feature data set, the preset chart information extracting identifying model and the preset result requirement specifically includes: inputting the characteristic data set into a preset chart information extraction and identification model to obtain a first identification result file and a second identification result file; and acquiring a recognition result according to a preset result requirement, the first recognition result file and the second recognition result file.
Inputting the characteristic data set into a preset chart information extraction and identification model, and obtaining two files, namely a first identification result file and a second identification result file; wherein the first recognition result file is a result of two types of recognition and separation (the first file has high precision) and can be directly used as a classification recognition result (the first recognition result); the second recognition result file is a probability value file (namely, each grid value represents the probability of whether the grid is an intertidal zone, the probability value is between 0 and 1, the value of 0 represents the minimum probability of the grid being the intertidal zone, and the value of 1 represents the maximum probability of the grid being the intertidal zone) which is the same as the size of the feature data set; further, the second recognition result file is divided in a segmented mode, the second recognition result file is divided by taking 0.5 as a limit, all segments with grid probability values larger than 0.5 are classified into intertidal zone categories, all segments with grid probability values smaller than 0.5 are classified into non-intertidal zone categories, and a classification recognition result (second recognition result) with higher precision can be obtained; further, the second recognition result file is subjected to finer threshold division by comparing the preprocessed chart, the finer the threshold division of the second recognition result file is, the more accurate the classification of each grid is, and thus a more accurate classification recognition result (third recognition result) is obtained.
The first recognition result, the second recognition result and the third recognition result have high precision and can be used as classification results. The accuracy of the first recognition result is smaller than or equal to that of the second recognition result and smaller than or equal to that of the third recognition result, and one of the results can be selected and used correspondingly according to preset result requirements (the requirements for classification accuracy).
In one embodiment, the obtaining the identification result according to the preset result requirement, the first identification result file and the second identification result file specifically includes: taking the first identification result file as a first identification result; dividing the second identification result file into a category of the intertidal zone and a category of the non-intertidal zone according to a preset dividing limit, and taking the divided content as a second identification result; threshold division is carried out on the second identification result file according to a preset interval threshold, and the divided content is used as a third identification result; and selecting and determining the identification result from the first identification result, the second identification result and the third identification result according to a preset result requirement.
And S3, correcting the identification result according to a preset grid information correction method to obtain a chart extraction result.
In order to further improve the precision of extracting and identifying the chart information, the embodiment of the invention further removes a small number of grids which are wrongly classified based on the identification result, and further obtains a result with higher precision. This step is described by way of practical application, but not limitation, specifically, dividing the grids in the recognition result into a unit range of every 5×5 grids. In one cell range, the weight of the center grid may be assigned 1, and the weights of the remaining grids are 1. In a unit range, each grid category is multiplied by a weight to obtain the category with the largest number of the same categories, and the category is redefined as the category of the central grid of the unit range. And so on, traversing all grids in the recognition result. Exemplary: if the category of the central grid in one unit range is a non-intertidal category, the remaining grids have 22 intertidal categories and 2 non-intertidal categories. Then the inter-tidal band class x 22 > the non-inter-tidal band class x 2, then the class of the central grid of this cell range is redefined as the inter-tidal band class. The number of grids in a unit range and the weight of the central grid can be adjusted according to requirements, the number of grids in the unit range is selected to be odd multiplied by odd, the optimal selection is the same odd multiplied by the same odd, namely, every 5 multiplied by 5 grids are a unit range, the weight of the central grid can be assigned to be 1, and the weights of the other grids are 1. The selection can not influence the accuracy of removing the error grid because of the overlarge or the overlarge small of one unit range, the center grid can be positioned in the middle of one unit range, the representativeness is the most, and the weight of the center grid can not excessively determine the whole category of one unit range.
In order to further improve the accuracy of extracting and identifying chart information, as an alternative, the embodiment of the invention can also apply a cluster processing (Clump) method. The clustering process is to cluster and combine adjacent similar classification areas by using morphological operators. Classified images often lack spatial continuity (the presence of spots or holes in the classified area). Although low pass filtering can be used to smooth these images, the class information is often disturbed by the coding of neighboring classes, and clustering can solve this problem. The selected classifications are first merged together using an expansion operation, and then the classified images are checked for erosion operations using a transform of specified size.
In order to further improve the accuracy of extracting and identifying the chart information, as an alternative, the embodiment of the invention can also apply a filtering processing (Sieve) method. The filtering process solves the islanding problem that occurs in classified images. The filtering process uses a blob grouping method to eliminate these isolated classified pels. The category screening method is to determine whether a pixel is in the same group as surrounding pixels by analyzing surrounding 4 or 8 pixels. If the number of pixels analyzed in a class is less than the threshold of the input. These pixels are deleted from the class and the deleted pixels are classified as Unclassified pixels (Unclassified). The two parallel schemes can achieve the effects of removing a small number of grids which are incorrectly classified and re-classifying the incorrect grids, so that a result with higher precision is obtained.
Compared with the classification precision of the first recognition result, the second recognition result and the third recognition result, the chart extraction result obtained through the steps is higher, the influence of a small amount of grid information which is wrongly classified is removed, and the recognition effect of the two categories is smoother.
In one embodiment, the correcting the identification result according to a preset grid information correction method to obtain a chart extraction result specifically includes: performing grid division on the identification result according to a preset unit division size to obtain a plurality of grid units; respectively carrying out weight assignment on the central grid and the non-central grid according to a preset weight assignment method so as to obtain the weight of the central grid and the weight of the non-central grid; the grid unit comprises a central grid and a non-central grid; respectively carrying out weighted proportion calculation on the intertidal zone class and the non-intertidal zone class according to the class of the central grid, the class of the non-central grid, the central grid weight and the non-central grid weight to obtain the intertidal zone class proportion and the non-intertidal zone class proportion; determining a correction type of the central grid according to the inter-tidal zone category specific gravity and the non-inter-tidal zone category specific gravity; and outputting a chart extraction result according to the correction type of each center grid.
In one embodiment, the extraction processing method further includes: and converting the chart extraction result into a vector surface file, and carrying out finishing treatment on the vector surface file to obtain a first chart extraction result. Specifically, converting the chart extraction result into a grid file, converting the grid file into a vector surface file, and generating tiny, blocked and scattered surfaces with intervals in the discontinuous grid vector surface file after converting the grid file into the vector surface file; the redundant tiny discontinuous broken surfaces are removed, the missing parts which do not accurately reach the boundary are supplemented, the finishing treatment can enable the classification results of the two categories in the vector surface file to be more continuous, the boundary part is smoother, and the accuracy of the results is higher; further, the categories with the same attribute in the first chart extraction result can be fused to obtain a continuous and integral second chart extraction result, so that the precision is further improved; furthermore, the second chart extraction result is compared with the pretreated chart, the fused vector surface file is shaped and modified, and the operations of editing, moving, fine tuning and the like are carried out on the boundary line or the characteristic folding point, so that a more accurate identification result and a more accurate boundary effect (third chart extraction result) can be obtained; furthermore, the third sea chart extraction result can be converted into a vector line file to obtain a fourth sea chart extraction result comprising sea-land boundary lines, so that the boundaries of the two categories are more accurate and smooth, the boundaries are obvious, and the precision is higher.
When the two kinds of information are extracted through the steps, boundary lines of the two kinds can be directly obtained, and the boundary lines are not required to be extracted independently by using other methods, so that the complexity of work is reduced. Furthermore, if the boundary line of the plurality of kinds of required category information can be directly obtained when the plurality of kinds of required category information are extracted at one time, the working efficiency is greatly improved.
According to the method steps, the information such as land area range, 2m equal depth line range, 5m equal depth line range, sea area range and the like in the paper edition chart can be extracted. And by analogy, the dividing lines of the information such as land area range, 2m equal depth line range, 5m equal depth line range, sea area range and the like in the paper edition chart can be obtained.
The embodiment of the invention describes an extraction processing method of sea chart thematic information, which automatically identifies and extracts sea chart information through a preset sea chart information extraction and identification model to obtain an identification result, and further corrects the identification result; furthermore, the extraction processing method of the sea chart thematic information further improves the accuracy and efficiency of automatic sea map information extraction by processing each pixel in each layer in the sea chart to be processed.
Second embodiment
Besides the method, the embodiment of the invention also discloses a device for extracting and processing the chart thematic information. Fig. 2 is a schematic structural diagram of a device for extracting and processing chart thematic information according to an embodiment of the invention.
As shown in fig. 2, the extraction processing apparatus includes a feature acquisition unit 11, an extraction recognition unit 12, and a result correction unit 13.
The feature acquisition unit 11 is configured to acquire a chart to be processed, and perform preprocessing on the chart to be processed according to a preset image preprocessing method, so as to acquire a feature data set.
In one embodiment, the feature acquisition unit 11 is configured to: and acquiring the gray value of each pixel in each layer of the sea chart to be processed, and taking the gray value as a characteristic data set of the sea chart to be processed.
The extraction and identification unit 12 is configured to extract an identification model and a preset result requirement according to the feature data set and preset chart information, and identify and determine an identification result.
In an embodiment, the extraction and identification unit 12 is further configured to: inputting the characteristic data set into a preset chart information extraction and identification model to obtain a first identification result file and a second identification result file; and acquiring a recognition result according to a preset result requirement, the first recognition result file and the second recognition result file.
In an embodiment, the extraction and identification unit 12 is further configured to: taking the first identification result file as a first identification result; dividing the second identification result file into a category of the intertidal zone and a category of the non-intertidal zone according to a preset dividing limit, and taking the divided content as a second identification result; threshold division is carried out on the second identification result file according to a preset interval threshold, and the divided content is used as a third identification result; and selecting and determining the identification result from the first identification result, the second identification result and the third identification result according to a preset result requirement.
The result correcting unit 13 is configured to correct the identification result according to a preset grid information correction method to obtain a chart extraction result.
In an embodiment, the result correction unit 13 is further configured to: performing grid division on the identification result according to a preset unit division size to obtain a plurality of grid units; respectively carrying out weight assignment on the central grid and the non-central grid according to a preset weight assignment method so as to obtain the weight of the central grid and the weight of the non-central grid; the grid unit comprises a central grid and a non-central grid; respectively carrying out weighted proportion calculation on the intertidal zone class and the non-intertidal zone class according to the class of the central grid, the class of the non-central grid, the central grid weight and the non-central grid weight to obtain the intertidal zone class proportion and the non-intertidal zone class proportion; determining a correction type of the central grid according to the inter-tidal zone category specific gravity and the non-inter-tidal zone category specific gravity; and outputting a chart extraction result according to the correction type of each center grid.
In one embodiment, the extraction processing apparatus further comprises a model training unit for: acquiring a training chart for training and corresponding chart information, dividing the chart information into intertidal zone type information and non-intertidal zone type information, preprocessing the training chart according to a preset image preprocessing method, and acquiring a training characteristic data set; screening the intertidal zone samples from the intertidal zone category information according to a preset identification sample screening method, and screening the non-intertidal zone samples from the non-intertidal zone category information according to a preset reverse sample screening method; establishing a first training data set and a first verification data set according to the intertidal zone sample and the non-intertidal zone sample respectively; assimilating the first training data set and the first verification data set according to the image information of the training chart, and correspondingly obtaining the training data set and the verification data set; and training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model.
In an embodiment, the model training unit is further for: according to the training data set and the training feature data set, training and debugging a preset random forest model to obtain a first chart information extraction model; according to the verification data set and the training feature data set, verifying the first chart information extraction model, calculating model precision and Kappa coefficient, and judging whether the model precision and the Kappa coefficient meet preset model output requirements; repeating the steps when the model precision and the Kappa coefficient do not meet the preset model output requirement; and outputting the first chart information extraction model as a chart information extraction and identification model when the model precision and the Kappa coefficient meet preset model output requirements.
Wherein the units integrated by the extraction processing means may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by instructing related hardware by a computer program, where the computer program may be stored in a computer readable storage medium, and the computer program may implement the steps of each of the method embodiments described above when executed by a processor. That is, another embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program controls a device in which the computer-readable storage medium is located to execute the method for extracting the sea chart thematic information as described above when running.
The computer program comprises computer program code which may be in source code form, object code form, executable file or in some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.
It should be noted that the above-described apparatus embodiments are merely illustrative, and the units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the embodiment of the device provided by the invention, the connection relation between the units indicates that the units have communication connection, and the connection relation can be specifically realized as one or more communication buses or signal lines. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The embodiment of the invention describes an extraction processing device and a computer readable storage medium of chart thematic information, which automatically identify and extract chart information through a preset chart information extraction and identification model to obtain an identification result and further correct the identification result, and the extraction processing device and the computer readable storage medium realize automation of chart information extraction, thereby improving the accuracy and efficiency of automatic ocean map information extraction; furthermore, the extracting and processing device and the computer-readable storage medium for the sea chart thematic information described in the embodiment of the invention further improve the accuracy and efficiency of automatic extraction of the sea map information by processing each pixel in each layer in the sea chart to be processed.
Detailed description of the preferred embodiments
Besides the method and the device, the embodiment of the invention also describes a system for extracting and processing the chart thematic information.
The extraction processing system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, which when executed implements the extraction processing method of the chart thematic information as described above.
The processor may be a central processing unit (Central Processing Unit, CPU), other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like that is the control center of the device, connecting the various parts of the overall device using various interfaces and lines.
The memory may be used to store the computer program and/or modules, and the processor may implement various functions of the apparatus by running or executing the computer program and/or modules stored in the memory, and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data (such as audio data, phonebook, etc.) created according to the use of the handset, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
The embodiment of the invention describes an extraction processing system for chart thematic information, which automatically identifies and extracts chart information through a preset chart information extraction and identification model to obtain an identification result, and further corrects the identification result; furthermore, the extracting and processing system for the sea chart thematic information further improves the accuracy and efficiency of automatic extracting of the sea map information by processing each pixel in each layer in the sea chart to be processed.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the invention, such changes and modifications are also intended to be within the scope of the invention.
Claims (8)
1. The extraction processing method of the chart thematic information is characterized by comprising the following steps of:
acquiring a chart to be processed, preprocessing the chart to be processed according to a preset image preprocessing method, and acquiring a characteristic data set;
Extracting an identification model and a preset result requirement according to the characteristic data set and preset chart information, and identifying and determining an identification result;
correcting the identification result according to a preset grid information correction method to obtain a chart extraction result;
extracting an identification model and a preset result requirement according to the characteristic data set and preset chart information, and identifying and determining an identification result, wherein the identification method specifically comprises the following steps of: inputting the characteristic data set into a preset chart information extraction and identification model to obtain a first identification result file and a second identification result file; acquiring a recognition result according to a preset result requirement, the first recognition result file and the second recognition result file; wherein the first identification result file is a result that two categories have been identified and separated, the two categories being an intertidal category and a non-intertidal category; the second recognition result file is a probability value file with the same size as the characteristic data set, and each grid value represents the probability of whether the tidal zone exists or not;
the method for obtaining the identification result comprises the following steps of: taking the first identification result file as a first identification result; dividing the second identification result file into a category of the intertidal zone and a category of the non-intertidal zone according to a preset dividing limit, and taking the divided content as a second identification result; threshold division is carried out on the second identification result file according to a preset interval threshold, and the divided content is used as a third identification result;
The preset result requirement is a requirement on classification accuracy, and one of the first recognition result, the second recognition result and the third recognition result is selected as the recognition result.
2. The method for extracting and processing chart specific information according to claim 1, wherein the identifying result is corrected according to a preset grid information correction method to obtain chart extraction results, specifically comprising:
performing grid division on the identification result according to a preset unit division size to obtain a plurality of grid units;
respectively carrying out weight assignment on the central grid and the non-central grid according to a preset weight assignment method so as to obtain the weight of the central grid and the weight of the non-central grid; the grid unit comprises a central grid and a non-central grid;
respectively carrying out weighted proportion calculation on the intertidal zone class and the non-intertidal zone class according to the class of the central grid, the class of the non-central grid, the central grid weight and the non-central grid weight to obtain the intertidal zone class proportion and the non-intertidal zone class proportion;
determining a correction type of the central grid according to the inter-tidal zone category specific gravity and the non-inter-tidal zone category specific gravity;
And outputting a chart extraction result according to the correction type of each center grid.
3. The method for extracting and processing the chart thematic information according to claim 2, wherein preprocessing the chart to be processed according to a preset image preprocessing method, and obtaining a feature data set, specifically comprises:
and acquiring the gray value of each pixel in each layer of the sea chart to be processed, and taking the gray value as a characteristic data set of the sea chart to be processed.
4. The method for extracting and processing chart specific information according to claim 3, wherein before obtaining a chart to be processed and preprocessing the chart to be processed according to a preset image preprocessing method, the method for extracting and processing further comprises:
acquiring a training chart for training and corresponding chart information, dividing the chart information into intertidal zone type information and non-intertidal zone type information, preprocessing the training chart according to a preset image preprocessing method, and acquiring a training characteristic data set;
screening the intertidal zone samples from the intertidal zone category information according to a preset identification sample screening method, and screening the non-intertidal zone samples from the non-intertidal zone category information according to a preset reverse sample screening method;
Establishing a first training data set and a first verification data set according to the intertidal zone sample and the non-intertidal zone sample respectively;
assimilating the first training data set and the first verification data set according to the image information of the training chart, and correspondingly obtaining the training data set and the verification data set;
and training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model.
5. The method for extracting and processing chart topic information according to claim 4, wherein training and verifying a preset random forest model according to the training data set, the verification data set and a preset model output requirement to obtain a chart information extraction and identification model, specifically comprising:
according to the training data set and the training feature data set, training and debugging a preset random forest model to obtain a first chart information extraction model;
according to the verification data set and the training feature data set, verifying the first chart information extraction model, calculating model precision and Kappa coefficient, and judging whether the model precision and the Kappa coefficient meet preset model output requirements;
Repeating the steps when the model precision and the Kappa coefficient do not meet the preset model output requirement;
and outputting the first chart information extraction model as a chart information extraction and identification model when the model precision and the Kappa coefficient meet preset model output requirements.
6. The extracting and processing device for the sea chart thematic information is characterized by comprising a characteristic acquisition unit, an extracting and identifying unit and a result correcting unit, wherein,
the characteristic acquisition unit is used for acquiring a chart to be processed, preprocessing the chart to be processed according to a preset image preprocessing method and acquiring a characteristic data set;
the extraction and identification unit is used for extracting an identification model and a preset result requirement according to the characteristic data set, preset chart information, and identifying and determining an identification result;
the result correction unit is used for correcting the identification result according to a preset grid information correction method so as to obtain a chart extraction result;
the extraction and identification unit is further used for: inputting the characteristic data set into a preset chart information extraction and identification model to obtain a first identification result file and a second identification result file; acquiring a recognition result according to a preset result requirement, the first recognition result file and the second recognition result file; wherein the first identification result file is a result that two categories have been identified and separated, the two categories being an intertidal category and a non-intertidal category; the second recognition result file is a probability value file with the same size as the characteristic data set, and each grid value represents the probability of whether the tidal zone exists or not;
The extraction and identification unit is further used for: taking the first identification result file as a first identification result; dividing the second identification result file into a category of the intertidal zone and a category of the non-intertidal zone according to a preset dividing limit, and taking the divided content as a second identification result; threshold division is carried out on the second identification result file according to a preset interval threshold, and the divided content is used as a third identification result; selecting and determining a recognition result from the first recognition result, the second recognition result and the third recognition result according to a preset result requirement;
the preset result requirement is a requirement on classification accuracy, and one of the first recognition result, the second recognition result and the third recognition result is selected as the recognition result.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored computer program, wherein the computer program, when run, controls a device in which the computer-readable storage medium is located to execute the method for extracting sea chart topic information according to any one of claims 1 to 5.
8. A system for extracting and processing sea chart topic information, characterized in that the system comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the method for extracting and processing sea chart topic information according to any one of claims 1 to 5 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211069968.0A CN115393884B (en) | 2022-09-02 | 2022-09-02 | Method, device and system for extracting and processing chart thematic information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211069968.0A CN115393884B (en) | 2022-09-02 | 2022-09-02 | Method, device and system for extracting and processing chart thematic information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115393884A CN115393884A (en) | 2022-11-25 |
| CN115393884B true CN115393884B (en) | 2023-05-02 |
Family
ID=84123719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211069968.0A Active CN115393884B (en) | 2022-09-02 | 2022-09-02 | Method, device and system for extracting and processing chart thematic information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115393884B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023128A (en) * | 2016-05-26 | 2016-10-12 | 天津市测绘院 | Method and device for fusing topographic map data and chart data |
| CN107564078A (en) * | 2016-06-30 | 2018-01-09 | 中国电力科学研究院 | A kind of grid block plan Automatic Vector method with interference pixel |
| 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 |
| CN114972581A (en) * | 2022-05-10 | 2022-08-30 | 上海商汤智能科技有限公司 | Remote sensing image labeling method, device, system, equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101116253B1 (en) * | 2010-02-08 | 2012-03-09 | 한국해양연구원 | System for updating electronic navigational chart according to tide |
| CN102800052B (en) * | 2012-06-13 | 2014-12-24 | 浙江大学 | Semi-automatic digital method of non-standard map |
| CN103177255B (en) * | 2013-03-15 | 2016-01-20 | 浙江大学 | A kind of intertidal zone extraction method based on multiresolution digital elevation model |
| CN108875659B (en) * | 2018-06-26 | 2022-04-22 | 上海海事大学 | Sea chart cultivation area identification method based on multispectral remote sensing image |
| CN112722156B (en) * | 2021-01-18 | 2021-10-29 | 大连海事大学 | Intelligent ship single mooring anchor position selection method based on decision tree |
-
2022
- 2022-09-02 CN CN202211069968.0A patent/CN115393884B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023128A (en) * | 2016-05-26 | 2016-10-12 | 天津市测绘院 | Method and device for fusing topographic map data and chart data |
| CN107564078A (en) * | 2016-06-30 | 2018-01-09 | 中国电力科学研究院 | A kind of grid block plan Automatic Vector method with interference pixel |
| 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 |
| CN114972581A (en) * | 2022-05-10 | 2022-08-30 | 上海商汤智能科技有限公司 | Remote sensing image labeling method, device, system, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115393884A (en) | 2022-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111860533A (en) | Image recognition method and device, storage medium and electronic device | |
| CN111027297A (en) | Method for processing key form information of image type PDF financial data | |
| KR20190143192A (en) | Automated defect classification method based on machine learning | |
| CN113919442B (en) | Tobacco maturity state identification method based on convolutional neural network | |
| CN116310882B (en) | Forestry information identification method based on high-resolution remote sensing image | |
| CN111695373B (en) | Zebra stripes positioning method, system, medium and equipment | |
| CN111161281A (en) | Face region identification method and device and storage medium | |
| CN113255434A (en) | Apple identification method fusing fruit features and deep convolutional neural network | |
| CN112348831A (en) | Shale SEM image segmentation method based on machine learning | |
| CN116721121A (en) | Plant phenotype color image feature extraction method | |
| CN111597845A (en) | Two-dimensional code detection method, device and equipment and readable storage medium | |
| CN111738310B (en) | Material classification method, device, electronic equipment and storage medium | |
| CN115393884B (en) | Method, device and system for extracting and processing chart thematic information | |
| CN113469233A (en) | Tobacco leaf automatic grading method and system based on deep learning | |
| CN110874835B (en) | Crop leaf disease resistance identification method and system, electronic equipment and storage medium | |
| CN116543325A (en) | Unmanned aerial vehicle image-based crop artificial intelligent automatic identification method and system | |
| CN115601747A (en) | Method and system for calculating confluency of adherent cells | |
| CN111415360B (en) | Tobacco leaf image cutting method, device, equipment and medium | |
| CN110348452B (en) | Image binarization processing method and system | |
| CN114299299A (en) | Tree leaf feature extraction method and device, computer equipment and storage medium | |
| CN113537253A (en) | Infrared image target detection method and device, computing equipment and storage medium | |
| Borianne et al. | Automated valuation of leaves area for large-scale analysis needing data coupling or petioles deletion | |
| CN113450355A (en) | Method for extracting image features based on multi-membrane CT image and 3DCNN network | |
| CN112200275A (en) | Artificial neural network quantification method and device | |
| CN113361530A (en) | Image semantic accurate segmentation and optimization method using interaction means |
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 |