CN114359696A - Weather map feature type identification and similarity matching system and method - Google Patents
Weather map feature type identification and similarity matching system and method Download PDFInfo
- Publication number
- CN114359696A CN114359696A CN202210016445.3A CN202210016445A CN114359696A CN 114359696 A CN114359696 A CN 114359696A CN 202210016445 A CN202210016445 A CN 202210016445A CN 114359696 A CN114359696 A CN 114359696A
- Authority
- CN
- China
- Prior art keywords
- weather
- similarity
- weather map
- map
- module
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012545 processing Methods 0.000 claims description 37
- 238000012549 training Methods 0.000 claims description 25
- 238000012015 optical character recognition Methods 0.000 claims description 13
- 239000003086 colorant Substances 0.000 claims description 7
- 238000013500 data storage Methods 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 abstract description 15
- 230000002265 prevention Effects 0.000 abstract description 9
- 238000001514 detection method Methods 0.000 description 18
- 238000010586 diagram Methods 0.000 description 16
- 238000005516 engineering process Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- FNYLWPVRPXGIIP-UHFFFAOYSA-N Triamterene Chemical compound NC1=NC2=NC(N)=NC(N)=C2N=C1C1=CC=CC=C1 FNYLWPVRPXGIIP-UHFFFAOYSA-N 0.000 description 6
- 238000011156 evaluation Methods 0.000 description 5
- 238000007726 management method Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000013144 data compression Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of computers, in particular to a system and a method for identifying weather map feature types and matching similarity. The system comprises: the weather map uploading module is used for uploading the imported current weather map and the imported historical weather map; the weather map identification module is used for identifying the type of the uploaded current weather map to obtain a weather type identification result corresponding to the weather map; the similarity matching module is used for performing similarity matching on the identified current weather map and the historical weather map to obtain a matched optimal solution weather map; and the result display module is used for outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity. The method is convenient for the atmosphere correlation system to quickly and accurately identify and match the weather types of the weather map, and acquire the corresponding weather type information and matching degree score of the weather map for reference of atmosphere pollution prevention and emission reduction measures; the method is flexible to call, and the heavy work brought by artificial reference to weather map identification is greatly reduced.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a system and a method for identifying weather map feature types and matching similarity.
Background
The atmospheric environment monitoring is a difficult and complicated work, and needs to be continuously monitored closely for 24 hours all day long. The existing automatic monitoring station is not high in automation degree, manual monitoring is often needed, the weather type can be detected and identified only in a short period of time according to collected images, and if an accident situation is met, problems of bad weather or more image data and the like are encountered, the weather type cannot be automatically responded, and manual processing is needed. Resulting in failure to perform continuous monitoring without full automation.
Aiming at the current situation of the atmospheric environment industry, how to use big data and artificial intelligence technology to identify weather types by images and match the weather types with current historical cases so as to call the historical cases and adjust parameters to prevent and control pollution is an urgent problem to be solved.
Disclosure of Invention
Aiming at the problems of the defects, the embodiment of the invention aims to provide a system and a method which are specially used for identifying and matching the characteristic types of weather patterns, so that the weather patterns can be conveniently and accurately identified and matched by an atmosphere correlation system, and corresponding weather pattern weather type information and matching degree scores can be obtained for reference of measures for preventing and reducing emission of atmospheric pollution; the system and the method can be flexibly called by a related service system, and the heavy work brought by artificial participation in weather map identification is greatly reduced.
In order to achieve the above purpose, the embodiment of the present invention provides the following technical solutions:
in a first aspect, in an embodiment provided by the present invention, a weather map feature type identification and similarity matching system is provided, where the system includes:
the weather map uploading module is used for uploading the imported current weather map and the imported historical weather map;
the weather map identification module is used for identifying the type of the uploaded current weather map to obtain a weather type identification result corresponding to the weather map;
the similarity matching module is used for performing similarity matching on the identified current weather map and the historical weather map to obtain a matched optimal solution weather map; and
and the result display module is used for outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity.
The weather map feature type identification and similarity matching system is convenient for an atmosphere correlation system to quickly and accurately identify and match weather types of the weather map, and acquire corresponding weather type information and matching degree scores of the weather map for reference of measures for preventing and reducing emission of atmospheric pollution; the invention can be flexibly called by a related service system, and greatly lightens the heavy work brought by artificial reference to weather map identification.
In some embodiments provided by the present invention, the weather map feature type identification and similarity matching system further includes a micro-service module, and the micro-service module is separately connected to the weather map uploading module, the weather map identification module, the similarity matching module, and the result display module through interfaces.
The weather map feature type identification and similarity matching system disclosed by the invention utilizes the architecture of the micro-service module to manage the related interfaces, and the module connected with each interface of the micro-service module independently exists and can be independently deployed without undergoing full-service release every time a certain function is released; following a single function principle, the modules connected with the interface of the micro service module can be functionally decoupled through RESTFUL or RPC calling; the fine granularity is high in expandability, and each interface connected module can be expanded independently and load is balanced independently; the micro-service can be flexibly called by the related business system to meet the functional requirements of the business system.
In some embodiments provided by the invention, the weather map feature type identification and similarity matching system is further connected with an atmosphere forecast early warning system through a microservice module, and the atmosphere forecast early warning system comprises an atmosphere pollution prevention module and a case base; the atmospheric pollution control module is used for receiving the current pollution weather map uploaded by the micro-service module, returning the current weather type to the micro-service module after identifying and processing the current pollution weather map, the case base is used for uploading the corresponding weather map in the case to the micro-service module, returning the weather map with the highest similarity to the atmospheric pollution control module after performing similarity matching by the micro-service module, and forming the result data into a typical case by the atmospheric pollution control module to generate a big data case base.
In some embodiments provided by the present invention, the weather map identification module is configured to identify weather pictures with two different pressures, that is, 500hpa and surface _ pres, where the 500hpa pressure is used to identify a west airflow, a high altitude ridge, and a high side weather in the weather pictures, and the surface _ pres pressure is used to identify a high pressure, a pressure equalizing, and a typhoon weather in the weather pictures.
In some embodiments provided by the present invention, the similarity matching module is configured to match a most similar picture in a current weather pattern and a historical weather pattern, and the matching rule of the similarity matching module is to determine the overall similarity of the pictures according to the picture weather pattern, the visual similarity of the pictures, the air pressure size and distribution, the temperature size and distribution, the wind direction and distribution, and the rainfall size and distribution.
In some embodiments provided by the present invention, the weather pattern feature type identification and similarity matching system further includes a model training module, where the model training module is configured to input a weather picture image into a tensrflow model for training to generate a weather model, and identify a weather type in the weather picture by using the tensrflow weather model.
In some embodiments of the present invention, the weather map feature type identification and similarity matching system further includes an AI module, where the AI module includes:
the weather image processing and computer vision processing unit is used for processing the weather image by adopting an OpenCv algorithm; and
and the optical character recognition unit is used for OCR picture recognition based on Tesseract and is used for recognizing image files in multiple formats and converting the image files into texts.
In some embodiments of the present invention, the weather map feature type identification and similarity matching system further includes: and the data storage service module is used for storing and managing the weather pictures.
In a second aspect, in an embodiment provided by the present invention, a method for identifying a weather pattern feature type and matching similarity is provided, which includes the following steps:
acquiring an uploaded weather picture data set, wherein the weather picture data set comprises a current weather map data set and a historical weather map data set;
traversing all the pictures in the weather picture data set, classifying the pictures in the weather picture data set, and independently processing the classified pictures of different categories to obtain a weather type identification result corresponding to the weather map;
traversing a historical weather map data set, and performing similarity matching on the current weather map to obtain an optimal solution weather map which is matched with the current weather map and has the highest historical matching similarity;
and outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity.
In some embodiments provided by the present invention, before acquiring the uploaded weather picture data set, the method further includes:
acquiring a weather picture uploading request;
responding to the weather picture uploading request, and receiving data compression files of the current weather picture and the historical weather picture;
decompressing a historical pollution event weather map compressed file and a current pollution event weather map compressed file in the uploaded current weather map and historical weather map data;
the decompressed weather map is distributed and matched to be routed to a corresponding image identification and recognition service module through a Gateway for recognition processing, and a corresponding weather type is recognized.
In some embodiments provided by the present invention, when classifying the pictures in the weather picture data set, the method further includes: judging the type of the weather picture in the weather picture data set; the types of the weather pictures comprise two types of 500hpa and surface _ pres, the classified weather pictures are copied to the folders of the corresponding types to find out the weather pictures, and different colors are respectively extracted to obtain new pictures.
In some embodiments provided by the present invention, the 500hpa type new picture is processed by an OpenCv training model to extract feature information in the new picture, where the feature information includes one or more of a curve angle profile, a wind direction angle, an air pressure magnitude, and a similarity in the new picture; the new picture of surface _ pres type identifies typhoon type through the pre-trained TensorFlow weather model.
In a third aspect, in an embodiment provided by the present invention, a method for identifying a weather pattern feature type and matching similarity is provided, which includes the following steps:
acquiring a front-end weather picture uploading request, and distributing matched routing information to a corresponding image identification service module through a Gateway for processing;
the weather pictures are classified into 500hpa and surface _ pres to be processed independently, and colors in the classified weather pictures are extracted to generate new pictures;
processing and extracting curve angle outline, wind direction angle, air pressure and similarity information in the 500hpa picture through an OpenCv algorithm, and storing the curve angle outline, the wind direction angle, the air pressure and the similarity information in the picture to characteristic text data; identifying the typhoon type of the surface _ pres picture through a TensorFlow meteorological model and storing the typhoon type into type text data;
and acquiring text data of the 500hpa and surface _ pres pictures to judge the picture types, calculating scores according to the data, sorting, and returning final results to the browser to display result data.
The technical scheme provided by the invention has the following beneficial effects:
1. the method is convenient for the atmosphere correlation system to quickly and accurately identify and match the weather types of the weather map, and acquire the corresponding weather type information and matching degree score of the weather map for reference of atmosphere pollution prevention and emission reduction measures; the invention can be flexibly called by a related service system, and greatly lightens the heavy work brought by artificial reference to weather map identification.
2. The weather map feature type identification and similarity matching system disclosed by the invention utilizes the architecture of the micro-service module to manage the related interfaces, and the module connected with each interface of the micro-service module independently exists and can be independently deployed without undergoing full-service release every time a certain function is released; following a single function principle, the modules connected with the interface of the micro service module can be functionally decoupled through RESTFUL or RPC calling; the fine granularity is high in expandability, and each interface connected module can be expanded independently and load is balanced independently; the micro-service can be flexibly called by the related business system to meet the functional requirements of the business system.
3. According to the method, the cases of heavy air quality pollution, major events and weather types and key characteristics thereof are automatically identified through a pollution prevention image identification technology, the matching of the similarity of historical pollution events is carried out by using artificial intelligence technologies such as intelligent image identification and the like, the pollution process and element analysis are carried out on the current pollution process and the historical cases, and the rapid calling of emission reduction measures is realized. Therefore, an optimal control technical scheme and engineering measures are obtained according to the actual emission reduction situation so as to achieve the regional atmospheric environment quality control target.
These and other aspects of the invention are apparent from and will be elucidated with reference to the embodiments described hereinafter. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
In order to more clearly illustrate the embodiments of the present invention or technical solutions in the related art, the drawings, which are needed to be used in the description of the exemplary embodiments or related art, will be briefly described below, and are used for providing further understanding of the present invention and are a part of the specification, and together with the embodiments of the present invention, serve to explain the present invention without limiting the present invention. In the drawings:
fig. 1 is a general framework diagram of a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 2 is a system functional block diagram of a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a weather chart uploading interface in a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 4 is a schematic view of an identification operation interface in a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a current weather map identification result in the weather map feature type identification and similarity matching system according to the embodiment of the present invention.
Fig. 6 is a schematic diagram of a historical weather map identification result in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 7 is another schematic diagram of a historical weather map recognition result in the weather map feature type recognition and similarity matching system according to the embodiment of the present invention.
Fig. 8 is a logic architecture diagram of a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 9 is a schematic diagram of a west airflow weather in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 10 is a schematic diagram of a weather vane for weather extraction of the westernish airflow in a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 11 is a schematic diagram of a wind direction of a westernward airflow weather vane in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 12 is a schematic view of a weather pattern feature type identification and similarity matching system in a high altitude ridge in the embodiment of the present invention.
Fig. 13 is a schematic diagram of isobars extracted in a system for identifying weather chart feature types and matching similarity provided in an embodiment of the present invention.
Fig. 14 is a schematic diagram of the isobar direction in the weather chart feature type identification and similarity matching system according to the embodiment of the present invention.
Fig. 15 is a schematic view of secondary high weather in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 16 is a schematic view of typhoon weather in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 17 is a schematic view of typhoon weather identified in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 18 is a schematic diagram of high-pressure weather in a weather map feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 19 is a schematic view of pressure-equalizing weather in a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Fig. 20 is a schematic diagram of a similarity matching process in the weather chart feature type identification and similarity matching system according to the embodiment of the present invention.
Fig. 21 is a flowchart of a method for identifying a weather pattern feature type and matching similarity according to an embodiment of the present invention.
Fig. 22 is a flowchart of data processing in a weather chart feature type identification and similarity matching system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In some of the flows described in the present specification and claims and in the above figures, a number of operations are included that occur in a particular order, but it should be clearly understood that these operations may be performed out of order or in parallel as they occur herein, with the order of the operations being indicated as 101, 102, etc. merely to distinguish between the various operations, and the order of the operations by themselves does not represent any order of performance. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor limit the types of "first" and "second" to be different.
The technical solutions in the exemplary embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the exemplary embodiments of the present invention, and it is apparent that the described exemplary embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Aiming at the current situation of the atmospheric environment industry, how to use big data and artificial intelligence technology to identify weather types by images and match the weather types with current historical cases so as to call the historical cases and adjust parameters to prevent and control pollution is an urgent problem to be solved. .
In order to solve the above problems, embodiments of the present invention provide a system and a method for identifying and matching similarity for weather patterns, which are specially used for identifying and matching weather patterns, and facilitate an atmosphere-related system to quickly and accurately identify and match the weather patterns, and obtain corresponding weather pattern information and matching degree scores for reference of measures for preventing and reducing emission of atmospheric pollution; the general assembly can be flexibly called by a related service system, and the heavy work brought by artificial reference to weather map identification is greatly reduced.
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
Referring to fig. 1, fig. 1 is a general framework diagram of a weather map feature type identification and similarity matching system according to the present invention. The invention provides a weather map feature type identification and similarity matching system, which mainly comprises a micro-service module, a weather map uploading module, a weather map identification module, a similarity matching module and a result display module, wherein the weather map uploading module, the weather map identification module, the similarity matching module and the result display module are independently connected with an interface of the micro-service module.
In this embodiment, the weather map uploading module is configured to upload the imported current weather map and historical weather map. And the weather map identification module is used for identifying the type of the uploaded current weather map to obtain a weather type identification result corresponding to the weather map. And the similarity matching module is used for performing similarity matching on the identified current weather map and the historical weather map to acquire a matched optimal solution weather map. And the result display module is used for outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity.
The weather map feature type identification and similarity matching system of the embodiment of the invention is convenient for an atmosphere correlation system to quickly and accurately identify and match weather types of the weather map, and acquire corresponding weather type information and matching degree scores of the weather map for reference of atmosphere pollution prevention and emission reduction measures; the invention can be flexibly called by a related service system, and greatly lightens the heavy work brought by artificial reference to weather map identification.
The microservice module is independently connected with the weather map uploading module, the weather map identification module, the similarity matching module and the result display module through interfaces. The weather map feature type identification and similarity matching system of the embodiment of the invention uses the architecture management related interface of the micro-service module, and the module connected with each interface of the micro-service module independently exists and can be separately deployed without undergoing full-service release every time a certain function is released; following a single function principle, the modules connected with the interface of the micro service module can be functionally decoupled through RESTFUL or RPC calling; the fine granularity is high in expandability, and each interface connected module can be expanded independently and load is balanced independently; the micro-service can be flexibly called by the related business system to meet the functional requirements of the business system.
The weather map feature type identification and similarity matching system is also connected with an atmospheric forecast early warning system through a micro-service module, and the atmospheric forecast early warning system comprises an atmospheric pollution prevention module and a case library; the atmospheric pollution control module is used for receiving the current pollution weather map uploaded by the micro-service module, returning the current weather type to the micro-service module after identifying and processing the current pollution weather map, the case base is used for uploading the corresponding weather map in the case to the micro-service module, returning the weather map with the highest similarity to the atmospheric pollution control module after performing similarity matching by the micro-service module, and forming the result data into a typical case by the atmospheric pollution control module to generate a big data case base.
Referring to fig. 1, when performing the feature type identification and similarity matching of the weather map, firstly, importing the current weather map and the historical weather map into a weather map uploading module of the weather map feature type identification and similarity matching system, then performing the weather type identification on the weather map in the weather map identification module, performing the similarity matching between the identified current weather map and the historical weather map, and finally obtaining the matched optimal solution; the related operation interfaces are uniformly managed by the micro-service module, and the micro-service can be called by other service systems.
In the embodiment, the atmosphere forecast early warning system can directly call an interface of a micro-service module deployed by the module, and is used for service logic processing of the atmosphere pollution prevention module; the result data can form a typical case generation large data case base.
In the embodiment of the invention, the functional modules of the weather map feature type identification and similarity matching system are shown in fig. 2, and the functional modules comprise a weather map uploading module, a weather map identification module, a similarity matching module and a result display module, which are respectively used for weather map uploading, weather map identification, similarity matching and result display. The weather map uploading module is responsible for uploading a current weather map and a historical weather map; the weather map identification module is responsible for identifying weather pictures with two different pressures of 500hpa and surface _ pres, 500hpa can identify weather with partial west airflow, high altitude ridge and secondary high weather, and surface _ pres can identify weather with high pressure, pressure equalization and typhoon; the similarity matching module is used for matching a picture which is most similar to the current weather map from the historical weather map; the result display module may return the weather map weather type identification result and the history matching the most similar weather map.
In an embodiment of the present invention, the weather map uploading module includes a weather map uploading interface, and the weather map uploading interface mainly includes three buttons: uploading a historical weather map, uploading a current weather map, and starting matching.
Referring to fig. 3, the upload historical weather map button is used for uploading a historical weather map set (including two types of pictures, 500hpa and surface _ pres) for matching out a picture that is most similar to the current weather map. The upload current weather map button is used to upload a current weather map (including two types of pictures of 500hpa and surface _ pres). The starting matching button is used for clicking the starting matching button, and decompressing the historical pollution event weather map compressed file and the current pollution event weather map compressed file at first; then, identifying the decompressed weather map to identify a corresponding weather type; and finally, performing similarity matching on the identified current weather and historical event weather maps to find the most similar weather map.
In an embodiment of the invention, the weather map identification module is used for identifying a weather map, and the weather map identification part is mainly used for identifying the decompressed weather map and identifying a corresponding weather type. Referring to fig. 4, the intelligent identification interface display result of the weather map identification module is preferably shown, and in the embodiment of the present invention, the weather types used for identification include a westernized airflow, a high altitude ridge, a secondary altitude, a high pressure, a pressure equalizing weather, and a typhoon weather.
In one embodiment of the invention, the similarity matching module is used for matching the most similar picture in the current weather type weather map and the historical weather map; the matching rule mainly judges the overall similarity of the pictures according to the weather types of the pictures, the visual similarity of the pictures, the air pressure size and distribution, the temperature size and distribution, the wind direction and distribution and the rainfall size and distribution.
In one embodiment of the invention, the main result display interface of the result display module mainly comprises a current weather map display interface, a historical weather map display interface and a historical weather map switch button. Referring to fig. 5, the current weather map presentation interface area is used for presenting a current weather map weather type identification result, the left side of the current weather map presentation interface is a 500hpa portion, and the right side of the current weather map presentation interface is a surface _ pres portion; and displaying a weather type identification result and a current weather map identification result above the picture in the current weather map display interface.
Referring to fig. 6 and 7, a historical weather map display interface area is used for displaying a historical weather map weather type identification result, a left side of the historical weather map display interface is a 500hpa part, and a right side of the historical weather map display interface is a surface _ pres part; and displaying a weather type identification result and a ranking score above the pictures in the historical weather map display interface, wherein the display results are sorted from high to low according to the score.
Referring to fig. 8, the weather map feature type identification and similarity matching system further includes a model training module, an AI module, and a data storage service module.
The model training module is used for inputting the meteorological picture image into a TensorFlow model for training to generate a meteorological model, and the TensorFlow meteorological model is used for identifying the weather type in the meteorological picture.
The AI module includes:
the weather image processing and computer vision processing unit is used for processing the weather image by adopting an OpenCv algorithm; and
and the optical character recognition unit is used for OCR picture recognition based on Tesseract and is used for recognizing image files in multiple formats and converting the image files into texts.
The data storage service module is used for storing and managing the weather pictures.
In this embodiment, the TensorFlow framework used by the model training module trains the model based on the SSD algorithm. Since target detection is an important application of AI, a target object such as a person, an animal, an automobile, an airplane, etc. can be detected in an image through a target detection model, and even the outline of the object can be drawn, as in the following group of diagrams, typhoon is recognized as a target detection object.
In this embodiment, a data training target detection model is trained and generated based on a SSD (Single Shot multi box Detector) algorithm, and may be used to identify typhoons and the like in an image. The method for training the target detection model based on the SSD algorithm comprises the following steps: installing a labeling tool, labeling data, configuring an SSD, downloading a pre-trained model, training a model, and using the model.
When the marking tool is installed, the model is trained by using sample data, data marking is firstly carried out, and the object in the marked image and the position information of the object are used as the sample data for training the model; the method comprises the steps of configuring SSD parameters after data are labeled, downloading a pre-training model, providing a pre-trained model by SSD-Tensorflow, training the model based on a VGG model, training according to a label file and the SSD model, wherein the larger the value of batch _ size, the larger the batch _ size, the higher the requirement on the performance of a machine, wherein the learning rate learning _ rate is adjusted according to the actual condition, the smaller the learning rate is, the more accurate the learning rate is, the longer the training time is, the larger the learning rate is, the training time can be shortened, and the accuracy is reduced. And after the training of the SSD model is finished, identifying the target through the model execution, and finishing the training of the target detection model by using the sample data through the steps.
In the embodiment of the invention, when the OpenCv algorithm is used for processing the image, the OpenCv algorithm provides a set of intelligent algorithms for processing the image, and algorithms such as image similarity, contour detection, direction detection, line detection and the like are used for processing the image. Wherein:
in the image similarity, the OpenCv algorithm provides 5 image similarity algorithms, namely a hash algorithm, a difference hash algorithm, a perceptual hash algorithm, a three-histogram and a single-channel histogram, wherein the hash algorithm, the difference hash algorithm and the perceptual hash algorithm are smaller in value, the higher the similarity is, the value is 0-64, namely, the different hash values of 64 bits in the hamming distance are. The values of the three histograms and the single-channel histogram are 0-1, and the larger the value, the higher the similarity.
During contour detection, the OpenCV algorithm carries out binarization on a binary image, so the contour detection comprises the following steps: loading an image, graying, binaryzation and contour detection.
During direction detection, the outline of each workpiece is respectively obtained; processing each contour, and obtaining information such as the middle points, the main directions and the like of the set of all contour points by adopting a pca (principal component analysis) method; drawing and returning the result.
When detecting lines: hough Transform (Hough Transform) Hough Transform is one of basic methods for recognizing geometric shapes from images in image processing, is widely applied, and has a plurality of improved algorithms. Mainly for separating geometric shapes (e.g., lines, curves, lines, circles, etc.) having certain identical characteristics from the image. The most basic hough transform is the detection of lines (curves) from black and white images.
In the embodiment of the invention, when the OCR picture based on Tesseract is recognized, Tesseract is an open source OCR (Optical character recognition) engine maintained by Google developed by HP laboratories, and is characterized by being open source, free, multi-language-supporting and multi-platform, and being capable of recognizing image files in various formats and converting the image files into texts.
In an embodiment of the invention, 500hpa, when identifying a 500hp weather pattern, may identify a westernized airflow, high ridges, and sub-high weather. When recognizing the westward airflow, referring to fig. 9, for example, the wind direction passing through the zhejiang river is from the leftmost side of the picture to the west and east, the isotherms, the isobars and the wind direction are parallel, and the overall trend is horizontal. Referring to fig. 9, the wind vanes which are extracted from west to east and pass through Zhejiang are shown in fig. 10, the wind direction of each wind vane is calculated by using an OpenCv direction detection algorithm, referring to fig. 11, and finally whether the weather map belongs to the weather of the off-west airflow is comprehensively evaluated according to the wind direction of each wind vane.
When the high-altitude spine is identified, as shown in fig. 12, the wind direction is from the northwest to the southeast, and the isotherm, the isobars and the wind direction are parallel; the isobars closest to Zhejiang are opened downwards, and the opening orientation angle is larger than 180 degrees and smaller than 360 degrees. Referring to fig. 12, isobars are extracted, referring to fig. 13, the direction of each isobar is calculated by using an OpenCv line detection algorithm, referring to fig. 14, and finally whether the weather map belongs to high-altitude ridge weather or not is comprehensively evaluated according to each isobar.
When the weather is identified as high, see fig. 15, for example, the 500hPa isobars in the zhejiang range are equal to or greater than 588. The identification method comprises the steps of extracting the value of an isobar in a weather map, identifying a pressure value based on an OCR character identification technology of Tesseract, finally obtaining the pressure value and distribution information, and comprehensively evaluating whether the weather map belongs to high weather.
In the embodiment of the invention, when the surface _ pres weather map is used for identifying the map, the surface _ pres can identify high-voltage, pressure-equalizing and typhoon weather. When typhoon weather is identified, as shown in fig. 16, rainfall generally occurs in typhoon; the wind arrows are relatively long, and are distributed counterclockwise around the contour line when viewed from the direction of the arrow; the lines are dense and compact, the central value is minimum, and the lines gradually increase outwards; for example, the typhoon center is east in Anhui, south in Shandong, and Japan is excluded.
The method for identifying the typhoon weather comprises the following steps: referring to fig. 17, since the typhoon features are obvious, the typhoon features can be identified by using a target detection technology; and detecting typhoon by using a TensorFlow frame based on an SSD algorithm training model, and finally, comprehensively evaluating whether the weather map belongs to typhoon weather within the influence range of Zhejiang according to the recognition result and typhoon position information.
When identifying high pressures, see FIG. 18, for example, Zhejiang has a high pressure center (with isobar values greater than 1012 and increasing in order) or near Zhejiang. The method of identifying the high voltage is: extracting an isobaric line value, identifying the pressure value based on an OCR character recognition technology of Tesseract, and finally comprehensively evaluating whether the weather map belongs to high-altitude ridge weather or not according to each isobaric line.
When identifying the voltage sharing, see fig. 19, for example, the isobars near the Zhejiang are sparse, and there is no or only one isobar. And extracting pressure lines in the Zhejiang range, calculating the number and distribution of the contours by using an OpenCv contour detection algorithm, and finally, comprehensively evaluating whether the weather map belongs to pressure-equalizing weather or not according to contour data in the Zhejiang range.
In the embodiment of the invention, when carrying out intelligent similarity matching, referring to fig. 20, firstly, whether the two compared graphs are of the same weather type is judged, then, the similarity of the graphs is matched by using an Opencv similarity algorithm, and then, the west airflow, the high altitude ridge, the sub-height, the high pressure, the pressure equalizing and the typhoon are scored, and then, the scoring is carried out according to the rules of different weather graphs. The rules of different weather maps are respectively:
partial west airflow: wind direction number of the wind vane from west to east.
High-altitude ridge: the opening is oriented at an angle greater than 180 and less than 360.
The auxiliary height is as follows: the 500hPa isobars near Zhejiang are greater than 584.
High pressure: a high pressure center position.
Pressure equalizing: the number of isobars in the Zhejiang range.
Typhoon: typhoon similarity, typhoon intensity and typhoon position.
And finally, acquiring the highest score according to the score ranking as the picture with the highest matching degree.
Therefore, the weather map feature type identification and similarity matching system of the embodiment of the invention manages the relevant interfaces by using a micro service architecture, the micro service can be independently deployed and operated and can be flexibly called by a relevant service system, a deep learning model is trained on a historical weather map data set by using TensorFlow, the weather type of a future weather map (such as typhoon weather) can be identified by the model, the weather map is subjected to image processing by using an AI intelligent algorithm provided by OpenCv, important parameter information (such as wind direction, isothermal isobars and rainfall) of the weather map is obtained, and an open-source framework Tesseract is used for Optical Character Recognition (OCR) to identify the temperature value and the air pressure value of the weather map.
In addition, in the weather map feature type identification and similarity matching system of the embodiment, the relevant service interfaces respectively issue four micro services of a picture uploading service, a weather identification service, an intelligent matching service and a result obtaining service, and the micro services can be called by an atmospheric system; each micro service exists independently, can be deployed independently, is expanded independently, has balanced load independently and is convenient to call.
The system of the embodiment of the invention can identify the partial west airflow, high-altitude ridge, secondary height, high pressure, pressure equalization, typhoon weather and the like.
In a preferred embodiment provided by the invention, a method for identifying weather map feature types and matching similarity comprises the following steps:
acquiring an uploaded weather picture data set, wherein the weather picture data set comprises a current weather map data set and a historical weather map data set;
traversing all the pictures in the weather picture data set, classifying the pictures in the weather picture data set, and independently processing the classified pictures of different categories to obtain a weather type identification result corresponding to the weather map;
step three, traversing a historical weather map data set, and performing similarity matching on the current weather map to obtain an optimal solution weather map which is matched with the current weather map and has the highest historical matching similarity;
and step four, outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity.
In this embodiment, before acquiring the uploaded weather picture data set, the method further includes:
acquiring a weather picture uploading request;
responding to the weather picture uploading request, and receiving data compression files of the current weather picture and the historical weather picture;
decompressing a historical pollution event weather map compressed file and a current pollution event weather map compressed file in the uploaded current weather map and historical weather map data;
the decompressed weather map is distributed and matched to be routed to a corresponding image identification and recognition service module through a Gateway for recognition processing, and a corresponding weather type is recognized.
When classifying the pictures in the weather picture data set, the method further comprises the following steps: judging the type of the weather picture in the weather picture data set; the types of the weather pictures comprise two types of 500hpa and surface _ pres, the classified weather pictures are copied to the folders of the corresponding types to find out the weather pictures, and different colors are respectively extracted to obtain new pictures.
Extracting feature information in the 500hpa type new picture through OpenCv training model processing, wherein the feature information comprises one or more of curve angle outline, wind direction angle, air pressure and similarity in the new picture; the new picture of surface _ pres type identifies typhoon type through the pre-trained TensorFlow weather model.
In an embodiment provided by the present invention, referring to fig. 21, a method for identifying a weather map feature type and matching similarity is provided, which includes the following steps:
acquiring a front-end weather picture uploading request, and distributing matched routing information to a corresponding image identification service module through a Gateway for processing;
the weather pictures are classified into 500hpa and surface _ pres to be processed independently, and colors in the classified weather pictures are extracted to generate new pictures;
processing and extracting curve angle outline, wind direction angle, air pressure and similarity information in the 500hpa picture through an OpenCv algorithm, and storing the curve angle outline, the wind direction angle, the air pressure and the similarity information in the picture to characteristic text data; identifying the typhoon type of the surface _ pres picture through a TensorFlow meteorological model and storing the typhoon type into type text data;
and acquiring text data of the 500hpa and surface _ pres pictures to judge the picture types, calculating scores according to the data, sorting, and returning final results to the browser to display result data.
In this embodiment, a front-end weather picture uploading request is distributed to a corresponding image recognition service for processing through a Gateway distribution matching route; then, the pictures are classified into 500hpa and surface _ pres for independent processing, and various colors are extracted to generate new pictures; 500hpa extracts information storage texts such as curve angle outlines, wind direction angles, air pressure magnitudes, similarity and the like in the pictures through OpenCv processing; the surface _ pres picture identifies the typhoon type through the running call of a weather model trained by TensorFlow; the two types of pictures judge the picture types according to the text data, calculate scores through the data and sort the scores, and return the final result to the browser.
As shown in fig. 22, the data processing flow is traversed after the weather map is uploaded, the weather map is classified according to two types, namely 500hpa and surface _ pres, and new pictures (isobaric pictures, isothermal pictures, wind maps and the like) are extracted according to different colors in the pictures; identifying the type and parameter information (wind direction, pressure, position and the like) of the extracted new picture by using OpenCv, Tesserac and Tensorflow training models; and finally, performing similarity matching ranking according to the type and the parameter information, and returning and displaying the result data.
When the weather map feature type identification and similarity matching system is applied, the method comprises the following steps:
1. and identifying the weather type.
The atmospheric system can generate pollution events according to information such as the exceeding of factors and the like, a weather map in the current pollution event time period is obtained, and the system weather map generates and displays corresponding weather types.
2. The weather type is used as an evaluation parameter of the atmospheric pollution event.
The weather type is used as an important evaluation parameter of the atmospheric pollution event for preventing and controlling the atmospheric pollution event.
3. Case matching obtains the best historical similar events.
And comparing and scoring the current pollution event evaluation parameters with historical pollution event evaluation parameters in the case, wherein the weather type is used as an important evaluation index.
4. And adjusting parameters according to historical events to prevent and treat pollution.
Acquiring corresponding historical emission reduction arrangement from a pollution event case in historical matching, optimizing and adjusting an emission reduction scheme, and then calling a model to simulate the emission reduction scheme to acquire a future factor concentration trend; and obtaining an optimal emission reduction scheme by continuously adjusting parameters and simulating.
5. Case base.
After the current pollution event emission reduction scheme is completed, the current pollution event emission reduction scheme can be stored in a case library as a typical case for future similar pollution event emission reduction reference, and the cases in the case library can be checked in detail, wherein the cases comprise a current weather map of the pollution event and an identified weather type.
In this embodiment, the method for identifying weather pattern feature types and matching similarity implements the process of identifying weather pattern feature types and matching similarity of the weather pattern feature type and matching similarity system in the foregoing embodiments. Therefore, the operation process of the weather map feature type identification and similarity matching method in this embodiment will not be described in detail.
In summary, the technical scheme provided by the invention has the following advantages:
1. the method is convenient for the atmosphere correlation system to quickly and accurately identify and match the weather types of the weather map, and acquire the corresponding weather type information and matching degree score of the weather map for reference of atmosphere pollution prevention and emission reduction measures; the invention can be flexibly called by a related service system, and greatly lightens the heavy work brought by artificial reference to weather map identification.
2. The weather map feature type identification and similarity matching system disclosed by the invention utilizes the architecture management related interface of the micro-service management module, and the module connected with each interface of the micro-service management module independently exists and can be independently deployed without undergoing full-service release every time a certain function is released; following a single function principle, the modules connected with the interface of the micro service management module can be functionally decoupled through RESTFUL or RPC calling; the fine granularity is high in expandability, and each interface connected module can be expanded independently and load is balanced independently; the micro-service can be flexibly called by the related business system to meet the functional requirements of the business system.
3. According to the method, the cases of heavy air quality pollution, major events and weather types and key characteristics thereof are automatically identified through a pollution prevention image identification technology, the matching of the similarity of historical pollution events is carried out by using artificial intelligence technologies such as intelligent image identification and the like, the pollution process and element analysis are carried out on the current pollution process and the historical cases, and the rapid calling of emission reduction measures is realized. Therefore, an optimal control technical scheme and engineering measures are obtained according to the actual emission reduction situation so as to achieve the regional atmospheric environment quality control target.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A weather map feature type identification and similarity matching system is characterized by comprising:
the weather map uploading module is used for uploading the imported current weather map and the imported historical weather map;
the weather map identification module is used for identifying the type of the uploaded current weather map to obtain a weather type identification result corresponding to the weather map;
the similarity matching module is used for performing similarity matching on the identified current weather map and the historical weather map to obtain a matched optimal solution weather map; and
and the result display module is used for outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity.
2. The weather pattern feature type identification and similarity matching system of claim 1, further comprising a micro-service module, the micro-service module being separately connected to the weather pattern uploading module, the weather pattern identification module, the similarity matching module, and the result display module via interfaces.
3. The weather pattern feature type identification and similarity matching system of claim 2, wherein the weather pattern feature type identification and similarity matching system is further connected with an atmospheric forecast warning system through a microservice module, the atmospheric forecast warning system comprising an atmospheric pollution control module and a case base; the atmospheric pollution control module is used for receiving the current pollution weather map uploaded by the micro-service module, returning the current weather type to the micro-service module after identifying and processing the current pollution weather map, the case base is used for uploading the corresponding weather map in the case to the micro-service module, returning the weather map with the highest similarity to the atmospheric pollution control module after performing similarity matching by the micro-service module, and forming the result data into a typical case by the atmospheric pollution control module to generate a big data case base.
4. The weather map feature type identification and similarity matching system of claim 1, wherein the weather map identification module is configured to identify weather pictures at two different pressures, namely 500hpa and surface _ pres, wherein the 500hpa pressure is used to identify those weather pictures with higher weather, higher ridges and higher secondary weather, and the surface _ pres pressure is used to identify those weather pictures with higher pressure, and typhoon weather.
5. The weather pattern feature type identification and similarity matching system of claim 4, wherein the similarity matching module is configured to match a most similar picture in the current weather pattern and the historical weather pattern, and the matching rule of the similarity matching module is to determine the overall similarity of the pictures from the picture weather pattern, the visual similarity of the pictures, the air pressure size and distribution, the temperature size and distribution, the wind direction size and distribution, and the rainfall size and distribution.
6. The weather pattern feature type identification and similarity matching system of claim 5, wherein the weather pattern feature type identification and similarity matching system further comprises a model training module, the model training module is configured to input a weather picture image into a TensorFlow model for training to generate a weather model, and the TensorFlow weather model is used to identify a weather type in the weather picture.
7. The weather pattern feature type identification and similarity matching system of claim 1 or 2, wherein the weather pattern feature type identification and similarity matching system further comprises an AI module, the AI module comprising:
the weather image processing and computer vision processing unit is used for processing the weather image by adopting an OpenCv algorithm; and
and the optical character recognition unit is used for OCR picture recognition based on Tesseract and is used for recognizing image files in multiple formats and converting the image files into texts.
8. The weather pattern feature type identification and similarity matching system of claim 7, wherein the weather pattern feature type identification and similarity matching system further comprises: and the data storage service module is used for storing and managing the weather pictures.
9. A method for identifying weather map feature types and matching similarity is characterized in that the method for identifying weather map feature types and matching similarity is used for realizing the feature type identification and similarity matching in the system for identifying weather map feature types and matching similarity as claimed in any one of claims 1 to 7; the weather map feature type identification and similarity matching method comprises the following steps:
acquiring an uploaded weather picture data set, wherein the weather picture data set comprises a current weather map data set and a historical weather map data set;
traversing all the pictures in the weather picture data set, classifying the pictures in the weather picture data set, and independently processing the classified pictures of different categories to obtain a weather type identification result corresponding to the weather map;
traversing a historical weather map data set, and performing similarity matching on the current weather map to obtain an optimal solution weather map which is matched with the current weather map and has the highest historical matching similarity;
and outputting the weather type identification result of the weather map and the optimal solution weather map with the highest historical matching similarity.
10. A method for identifying weather map feature types and matching similarity is characterized in that the method for identifying weather map feature types and matching similarity is used for realizing the feature type identification and similarity matching in the system for identifying weather map feature types and matching similarity as claimed in any one of claims 1 to 7; the weather map feature type identification and similarity matching method comprises the following steps:
acquiring a front-end weather picture uploading request, and distributing matched routing information to a corresponding image identification service module through a Gateway for processing;
the weather pictures are classified into 500hpa and surface _ pres to be processed independently, and colors in the classified weather pictures are extracted to generate new pictures;
processing and extracting curve angle outline, wind direction angle, air pressure and similarity information in the 500hpa picture through an OpenCv algorithm, and storing the curve angle outline, the wind direction angle, the air pressure and the similarity information in the picture to characteristic text data; identifying the typhoon type of the surface _ pres picture through a TensorFlow meteorological model and storing the typhoon type into type text data;
and acquiring text data of the 500hpa and surface _ pres pictures to judge the picture types, calculating scores according to the data, sorting, and returning final results to the browser to display result data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210016445.3A CN114359696A (en) | 2022-01-07 | 2022-01-07 | Weather map feature type identification and similarity matching system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210016445.3A CN114359696A (en) | 2022-01-07 | 2022-01-07 | Weather map feature type identification and similarity matching system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114359696A true CN114359696A (en) | 2022-04-15 |
Family
ID=81107244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210016445.3A Pending CN114359696A (en) | 2022-01-07 | 2022-01-07 | Weather map feature type identification and similarity matching system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114359696A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101088082B1 (en) * | 2010-11-29 | 2011-11-30 | (주)에스이랩 | System of searching similar weather map by principle component analysis and the method thereof |
| KR20130030550A (en) * | 2011-09-19 | 2013-03-27 | 대한민국(기상청장) | System for searching similar weather chart and method thereof |
| US9411072B1 (en) * | 2013-03-15 | 2016-08-09 | Exelis, Inc. | Real-time adaptive weather surveillance system and method |
| US20170357920A1 (en) * | 2016-06-09 | 2017-12-14 | Douglas Stewart | Method and system for estimating and mapping weather risk |
| CN108491959A (en) * | 2018-02-11 | 2018-09-04 | 周名扬 | A kind of similar weather forecast method of intelligence and system, information data processing terminal |
| US20190171986A1 (en) * | 2017-12-06 | 2019-06-06 | Vladislav Michael Beznos | System and method for evaluating images to support multiple risk applications |
| CN110766333A (en) * | 2019-10-29 | 2020-02-07 | 北京依派伟业数码科技有限公司 | Intelligent processing method and system for weather phenomenon information |
| CN111612245A (en) * | 2020-05-18 | 2020-09-01 | 北京中科三清环境技术有限公司 | Atmospheric pollution condition prediction method and device, electronic equipment and storage medium |
| CN112052852A (en) * | 2020-09-09 | 2020-12-08 | 国家气象信息中心 | Character recognition method of handwritten meteorological archive data based on deep learning |
| CN112815936A (en) * | 2020-12-29 | 2021-05-18 | 中国人民解放军国防科技大学 | Rapid all-sky-domain star map identification method and system for noise robustness |
-
2022
- 2022-01-07 CN CN202210016445.3A patent/CN114359696A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101088082B1 (en) * | 2010-11-29 | 2011-11-30 | (주)에스이랩 | System of searching similar weather map by principle component analysis and the method thereof |
| KR20130030550A (en) * | 2011-09-19 | 2013-03-27 | 대한민국(기상청장) | System for searching similar weather chart and method thereof |
| US9411072B1 (en) * | 2013-03-15 | 2016-08-09 | Exelis, Inc. | Real-time adaptive weather surveillance system and method |
| US20170357920A1 (en) * | 2016-06-09 | 2017-12-14 | Douglas Stewart | Method and system for estimating and mapping weather risk |
| US20190171986A1 (en) * | 2017-12-06 | 2019-06-06 | Vladislav Michael Beznos | System and method for evaluating images to support multiple risk applications |
| CN108491959A (en) * | 2018-02-11 | 2018-09-04 | 周名扬 | A kind of similar weather forecast method of intelligence and system, information data processing terminal |
| CN110766333A (en) * | 2019-10-29 | 2020-02-07 | 北京依派伟业数码科技有限公司 | Intelligent processing method and system for weather phenomenon information |
| CN111612245A (en) * | 2020-05-18 | 2020-09-01 | 北京中科三清环境技术有限公司 | Atmospheric pollution condition prediction method and device, electronic equipment and storage medium |
| CN112052852A (en) * | 2020-09-09 | 2020-12-08 | 国家气象信息中心 | Character recognition method of handwritten meteorological archive data based on deep learning |
| CN112815936A (en) * | 2020-12-29 | 2021-05-18 | 中国人民解放军国防科技大学 | Rapid all-sky-domain star map identification method and system for noise robustness |
Non-Patent Citations (3)
| Title |
|---|
| 周庭泽, 梁平德, 刘爱霞: "天气图相似识别系统及其在天气预报中的应用", 气象, no. 06, 21 June 1991 (1991-06-21), pages 1 - 4 * |
| 张云港;杨金华;: "天气图特征提取研究", 云南大学学报(自然科学版), no. 2, 15 September 2007 (2007-09-15) * |
| 王继光;孙鹤泉;鞠增彪;: "基于最小面积准则的天气图等值线相似算法", 解放军理工大学学报(自然科学版), no. 03, 15 June 2013 (2013-06-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108830188B (en) | Vehicle detection method based on deep learning | |
| US10628890B2 (en) | Visual analytics based vehicle insurance anti-fraud detection | |
| CN105574550A (en) | Vehicle identification method and device | |
| CN110598693A (en) | Ship plate identification method based on fast-RCNN | |
| CN106610969A (en) | Multimodal information-based video content auditing system and method | |
| CN111401418A (en) | Employee dressing specification detection method based on improved Faster r-cnn | |
| CN111402224B (en) | Target identification method for power equipment | |
| CN111522951A (en) | Sensitive data identification and classification technical method based on image identification | |
| CN111242899A (en) | Image-based flaw detection method and computer-readable storage medium | |
| CN115631472B (en) | Intelligent detection method for pedestrian intrusion on expressway | |
| CN110909598A (en) | Deep learning-based method for recognizing illegal traffic driving of non-motor vehicle lane | |
| CN114998234A (en) | Self-supervision spring defect detection method based on data enhancement strategy | |
| CN115147380A (en) | Small transparent plastic product defect detection method based on YOLOv5 | |
| CN115019294A (en) | Pointer instrument reading identification method and system | |
| CN111339927A (en) | Intelligent work state identification system for personnel in electric power business hall | |
| CN111461143A (en) | Picture copying identification method and device and electronic equipment | |
| CN114694130A (en) | Method and device for detecting telegraph poles and pole numbers along railway based on deep learning | |
| CN109190451B (en) | Remote sensing image vehicle detection method based on LFP characteristics | |
| CN111339926A (en) | Intelligent recognition system for behavior specification of personnel in electric power business hall | |
| CN114359696A (en) | Weather map feature type identification and similarity matching system and method | |
| CN111160756A (en) | Scenic spot assessment method and model based on secondary artificial intelligence algorithm | |
| CN115601747A (en) | Method and system for calculating confluency of adherent cells | |
| CN110765900A (en) | DSSD-based automatic illegal building detection method and system | |
| CN111325076A (en) | Aviation ground building extraction method based on U-net and Seg-net network fusion | |
| CN115512331A (en) | Traffic sign detection method and device, computer equipment and computer-readable storage medium |
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 |