CN115168526A - Address positioning correction method, system and storage medium based on geocoding - Google Patents

Abstract

The invention belongs to the technical field of communication positioning, and provides an address positioning correction method, system and storage medium based on geocoding. The positioning correction method of the invention comprises the following steps: comparing the mobile equipment positioning information with the actual position information and judging whether the mobile equipment positioning information needs to be corrected or not; scanning a house code two-dimensional code or shooting a landmark building, uploading the house code two-dimensional code or shooting the landmark building to a server for analysis, and then acquiring position information of the house code two-dimensional code or shooting the landmark building; geocoding the position information to obtain longitude and latitude information, and comparing the longitude and latitude information with the actual position information; and judging whether the position information data accords with the actual address or not, and presenting final address text information after position correction. The method and the system of the invention correct the position information of houses and landmarks, avoid the problems that the actual address is not matched with the picture watermark address and the positioning information deviation is large, greatly improve the work efficiency of the bank when the bank checks the data of loan enterprises, and simultaneously reduce the rework operation of operators.

Description

Address positioning correction method, system and storage medium based on geocoding

Technical Field

The invention relates to the technical field of communication positioning, in particular to an address positioning correction method and system based on geocoding and a storage medium.

Background

At present, many banking related businesses (such as bank enterprise reconciliation, small loan and mortgage registration) all carry out outdoor operations. Data verification is needed after relevant outdoor operation, and the loan enterprise needs to shoot on site if the bill is sent and the position of the loan enterprise is correct, and position watermarks and longitudes are added, so that the authenticity of the loan enterprise is conveniently verified. At present, the positioning of mobile equipment is mainly based on GPS, the precision of the positioning and navigation equipment outdoors can reach 5-10 meters, but with the abundance of use scenes, the problems of the traditional positioning and navigation based on beacon environment, communication with a base station and the like are highlighted, and the positioning and navigation equipment mainly comprises the following aspects:

1. the indoor positioning navigation has influence on the precision due to the blockage of the house wall, and the accuracy of the geographic position information cannot be guaranteed;

2. most enterprises are in office buildings at present, the positioning information deviation is large, so that a bank has the problem that a plurality of verification data addresses are not matched with image watermark addresses in the actual verification process, and the verification passing rate is low;

3. the positioning is inaccurate, the frequency of the rework operation of bank operators is high, and the consumed time cost and the labor resource cost are high.

Therefore, an address positioning correction method, system and storage medium based on geocoding are needed to be developed, so that geographic position information of related houses and landmarks can be corrected, the problem that an audit data address is not matched with a picture watermark address is avoided, the working efficiency of a bank in auditing related data of a loan enterprise is improved, the reworking operation of operators is reduced, and meanwhile, indoor positioning is more accurate.

Disclosure of Invention

The present invention provides an address location correction method, system and storage medium based on geocoding, so as to solve the problems of the prior location technology mentioned in the above background art.

In order to achieve the purpose, the invention adopts the following technical scheme:

in a first aspect, an embodiment of the present invention provides a method for correcting address location based on geocoding, where the method includes the following specific steps:

comparing the mobile equipment positioning information with the actual position information and judging whether the mobile equipment positioning information needs to be corrected or not;

scanning the house code two-dimensional code or shooting a landmark building, uploading the house code two-dimensional code or shooting the landmark building to a server, analyzing and then acquiring position information of the house code two-dimensional code or shooting the landmark building;

geocoding the position information to obtain longitude and latitude information, and comparing the longitude and latitude information with actual position information;

and judging whether the position information data accords with the actual address or not, and displaying the final address text information after position correction.

In a second aspect, an embodiment of the present invention provides a system for correcting address location based on geocoding, where the system includes: the system comprises a bar code identification module, an image analysis module, a geocoding module and a result plotting module. Wherein:

the bar code identification module is used for identifying the house code or the house code two-dimensional code to acquire the position information in the house code or the house code two-dimensional code;

the image analysis module is used for extracting the characteristics of the shot landmark building image, mapping the characteristics of the landmark building image to a neural network for image recognition and classification, screening out pictures which accord with actual landmarks, and acquiring position information in the pictures;

the geocoding module is used for geocoding the geographic position information acquired from the bar code identification module or the image analysis module, acquiring longitude and latitude information therein and carrying out targeted position correction operation;

and the result plotting module is used for displaying the coordinates of the final result according to the geocoded data.

Based on the above scheme, the house code is a sequence number issued by the administration office, which is generally a house code organized by the local property center according to the relevant national regulations, and is composed of 25 digits and can be classified into 8 levels. The first level of house coding usually has 2 codes, which represent the province and the autonomous region where the house is located; the second level is also 2 codes, which represent the city of the house; the third level of house coding has a 3-bit code, which represents the county or district where the house is located; the eighth level of house coding has 6 codes, which represent the rooms of the house, etc. The house code is usually determined, unique and unchangeable and is equivalent to the identity card number of the house.

The house code acquiring method specifically comprises the following steps:

checking whether house coding two-dimensional code stickers exist on a house door or a nearby wall surface;

and if so, directly scanning by a mobile phone to obtain house information to obtain the house code.

If not, the user needs to log in the office of the local community grid management office and input the corresponding building, address and management organization (such as xx district xx street) for query.

Based on the foregoing solution, the image analysis module includes: the device comprises an image preprocessing unit, a feature extraction unit, a neural network mapping unit and an identification and classification unit. Wherein:

the image preprocessing unit is configured to: the server carries out a series of processing such as deletion, smoothing, binarization, amplitude normalization and the like on useless information in original data contained in the shot and uploaded landmark building image to obtain various characteristics of the landmark picture, including but not limited to characteristics such as color, texture, shape, spatial relationship and the like;

the feature extraction unit described above, configured to: extracting various characteristics of the landmark picture through a neural network image recognition system, and performing final characteristic description display;

the neural network mapping unit is configured to: the server inputs the final feature description into a neural network, and the neural network is compared with the labels stored in the system to form a mapping relation;

the above-mentioned recognition and classification unit is configured to: and the server performs image recognition and classification according to the mapping result so as to screen out the pictures which accord with the actual landmarks.

Based on the foregoing solution, the geocoding module includes: the device comprises a data preprocessing unit, a search matching unit and a positioning correction unit. Wherein:

the data preprocessing unit is configured to: the server screens and fuses from each platform database according to the geographic position information to acquire more data information related to the geographic position; and the number of the first and second groups,

one or more of the following processing modes are carried out on the data information: denoising, standardizing and structuring to obtain the geographic position characteristic data;

the search matching unit is configured to: the search engine inquires related data in the address database and matches the related data with the geographic position characteristic data, so that the geographic position longitude and latitude information contained in the house code or the landmark picture is accurately taken out;

the positioning correction unit is configured to: the server judges whether the search matching result is consistent with the actual address or not, and performs targeted position correction operation.

Based on the scheme, the key points of the search matching are the segmentation of the input text and the scoring of the recall result. If the segmentation effect is not good, the proper record in the database cannot be found; if the scoring rules are set poorly, the output result may not be the desired result.

Based on the scheme, before the coordinate display of the final result is carried out, if the matched result does not completely match the input address, interpolation (including single-side offset and terminal offset) is carried out based on line elements; if the reference element is a surface or a point, the centroid of the surface is acquired to correct the coordinates.

In a third aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the processing method in any implementation manner of the foregoing first aspect or second aspect.

Compared with the prior art, the invention has the following advantages and positive effects:

(1) According to the invention, the geographic position information acquired from the house code or the landmark picture is subjected to production processing, screening and fusion, and searching and matching are carried out, so that the final produced geographic position longitude and latitude information is more accurate, and the problem of larger deviation of indoor positioning information is effectively solved.

(2) In the image analysis process, the invention carries out a series of processing such as deleting, smoothing, binaryzation, amplitude normalization and the like on useless information in original data contained in the landmark image to obtain various characteristics of the image, screens out the image which accords with the actual landmark according to the characteristics, avoids the problem that an audit data address is not matched with an image watermark address when a bank audits, improves the audit passing rate and reduces the rework operation of operators.

(3) Before the coordinate display of the final result is carried out, the invention carries out coordinate correction on the search matching result which is not completely matched with the input address, ensures the accuracy of the geographic position information, is convenient for a bank to verify the authenticity of the related data of the loan enterprise during the auditing and provides convenience for the bank and operators.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without inventive efforts, wherein:

fig. 1 is a flowchart illustrating a geocoding-based address location correction method according to an embodiment of the present invention;

FIG. 2 is a flow chart of an image parsing method according to an embodiment of the present invention;

FIG. 3 is a flow chart of a geocoding method provided by an embodiment of the invention;

fig. 4 shows a block diagram of a geocoding-based address location correction system according to an embodiment of the present invention.

Detailed Description

To more clearly illustrate the objects, technical solutions and advantages of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments, and the example embodiments can be implemented in various forms and should not be construed as being limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known methods, devices, implementations or operations have not been shown or described in detail to avoid obscuring aspects of the invention.

The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.

The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.

The present invention is described in detail below with reference to specific examples:

example 1

As shown in fig. 1, the present embodiment provides a geocoding-based address location correction method, which includes the following specific steps:

s1: comparing the mobile equipment positioning information with the actual position information and judging whether the mobile equipment positioning information needs to be corrected or not;

preferably, in this embodiment, when the bank operator performs data verification on the location of the loan enterprise, the bank operator may first use the mobile device to obtain location information by positioning, and compare the location information with the current actual location information to determine whether the target geographic location information needs to be corrected. If the deviation is within the negligible range, directly passing the audit; if the deviation is large, subsequent position correction operation is required;

s2: scanning a house code two-dimensional code or shooting a landmark building, uploading the house code two-dimensional code or shooting the landmark building to a server for analysis, and then acquiring position information of the house code two-dimensional code or shooting the landmark building;

preferably, the house code refers to a feature combination code assigned to a house basic unit according to a related house code rule to distinguish recognition and service management, how to establish a scientific and reasonable house code plays an important role in house information acquisition and use, and the scientific house code is not only used for recognition, but also has some important information, such as spatial position, affiliated area, orientation relation, characteristics convenient to memorize and compare and the like.

Specifically, four types of information, namely 'house address, house code, contact telephone (comprehensive station in the district to which the house code belongs), and online transaction (website)' generally appear when the house code two-dimensional code is scanned. The bank auditor analyzes the house coding information through the server, and can further accurately take out the longitude and latitude information of the address through the geocoding process after acquiring the position data (such as province, city, district, town, countryside, street, building and house number) of the target address.

Preferably, in the embodiment, the server parses the landmark building image to obtain the location information, wherein the location information is obtained by means of GPS positioning information in an image file format Exif under a neural network image recognition technology.

Specifically, after identifying the GPS positioning information in the landmark building image, the server inputs the data and the image into the neural network system, thereby extracting the features of the image. Further, the system classifies the image according to the comparison between the extracted image features and the storage labels; after the neural network screens out the pictures which accord with the actual landmarks, the images are interpreted and a human-readable sentence (namely address text) is returned, so that the geographical position information of the house or the landmarks is obtained.

S3: geocoding the position information to obtain longitude and latitude information, and comparing the longitude and latitude information with the actual position information;

preferably, in this embodiment, the obtaining, by the server, the longitude and latitude information of the geographic location through the geocoding process includes: three links of data preprocessing, search matching and plotting.

Specifically, when the server performs data preprocessing on the geographic position information, the server needs to perform structuring and standardization processing on the address text, and the process needs massive data and a dictionary to perform bottom layer data support. The mass data can be screened and fused on each platform besides self-acquisition; the dictionary can be mined and generated in a system database besides daily accumulation of the server. Furthermore, after the server obtains the characteristic data information of the target geographic position through a data preprocessing link, a search engine is started to search relevant data and is matched with the description information of the position coordinates stored in the database, and therefore the geographic target corresponding to the address information is identified; when the accurate positioning can not be carried out, a certain range of geographic targets matched with the address information are determined. Preferably, the geographic object in this embodiment is expressed by a coordinate vector in the form of a point, a line, or a surface. Therefore, the bank operator can accurately take out the longitude and latitude information of the geographic position contained in the house code or the landmark picture.

In this embodiment, the actual location information is compared with the acquired information, mainly by a latitude and longitude deviation value. The bank operator can accurately judge the position information of the target address according to the error between the target address and the image watermark address, and the problem that the audit data address is not matched with the image watermark address when the bank audits is effectively solved.

S4: and judging whether the position information data accords with the actual address or not, and presenting final address text information after position correction.

Preferably, the basis for judging whether the position information data conforms to the actual address is a result of searching for a match in step S3. If the geographic position information is matched with the geographic position information, the program is ended, and the geographic position information is returned; if the actual address is not completely matched, the position of the server needs to be corrected.

Specifically, the key points of searching for matching in the step S3 are the segmentation of the input text and the scoring of the recall result. If the segmentation effect is not good, the proper record in the database cannot be found; if the scoring rules are set poorly, the output result may not be the desired result. Further, before the coordinate display of the final result is performed, if the search matching result does not completely match the actual address input in the system, the position information correction processing is required. The position information correction comprises the following two aspects: on one hand, if the reference element is a line, interpolation (including unilateral offset and terminal offset) is needed based on the line element; on the other hand, if the reference element is a surface or a point, the centroid of the surface needs to be acquired to perform coordinate correction.

In the embodiment, whether the actual address is met or not is judged based on the result of searching matching, so that the judgment accuracy is guaranteed. Meanwhile, for the address which does not completely accord with the actual address, the bank operator can pertinently adopt a corresponding position correction method according to the actual situation, so that the finally displayed address text is more accurate.

Example 2

As shown in fig. 2, the present embodiment provides an image analysis method, which specifically includes the following steps:

s20: the server identifies the landmark building image and carries out preprocessing;

in this embodiment, after the bank operator uploads the captured landmark building image to the server, the server first identifies the image, analyzes the image into data information readable by the computer system, and performs a series of processing such as deleting, smoothing, binarizing, and amplitude normalizing on useless information in the original data to obtain preliminary feature data of the image.

S21: extracting the characteristic data of the image by using a neural network image recognition system;

preferably, in this embodiment, the extracting the image features by the neural network image recognition system includes: a characteristic extraction part and a non-characteristic extraction part.

Specifically, for the feature extraction part, the features include, but are not limited to, color, texture, shape, spatial relationship, and the like. This section mainly uses human experience to obtain pattern features and neural network classification capability to identify target images. And for the part without the characteristic extraction, the characteristic extraction is omitted, and the whole image is directly used as the input of the neural network.

S22: and the server maps the image characteristics to the neural network for image recognition and classification, and screens out the pictures which accord with the actual landmarks.

In this embodiment, preferably, before the server maps the image features to the neural network, a neural network mapping unit needs to be constructed in the system.

Specifically, by importing the landmark building picture and the extracted image feature data into a neural network system, the system outputs data information (such as the city, the region, the belonging type, the surrounding environment and the like where the picture is located) related to the landmark image after identifying and analyzing the landmark building picture; adjusting parameters of the neural network, repeating training until the parameters converge or the maximum iteration times is reached, and updating data of the neural network image recognition system; finally, the output data information is compared with the label stored in the system, so that the computer can understand the mapping relation between the landmark building image and the system storage data. Thus, a simple neural network mapping unit is constructed. Further, the neural network image recognition system maps the extracted image features to the neural network and carries out classification processing after comparing the extracted image features with the labels stored by the server; after the system screens out pictures that match the actual landmark, the system interprets the images and returns a human-readable sentence (i.e., address text) so that the bank operator can obtain the geographic location information of the house or landmark.

Example 3

As shown in fig. 3, the present embodiment provides a geocoding method, which specifically includes the following steps:

s30: scanning a house code two-dimensional code or shooting a landmark building, uploading the house code two-dimensional code or shooting the landmark building to a server for analysis, and then acquiring position information of the house code two-dimensional code or shooting the landmark building;

preferably, the bank operator can obtain the location data of the target address (such as province, city, district, town, county, street, building, house number, etc.) by analyzing the house code or landmark building by using the server. The house code two-dimensional code is generally pasted on a door or a nearby wall surface, and when the house code two-dimensional code sticker cannot be obtained, a bank worker can also input a corresponding building, an address and a management mechanism (such as xx street) and the like for query by logging in a local community grid management office official network. In the embodiment, the server analyzes the landmark building image to obtain the position information, wherein the position information is realized in the image recognition system by utilizing a neural network system based on the GPS positioning information in the image file format Exif. The specific procedure was as described in example 2.

S31: the server carries out data preprocessing on the geographic position information;

in this embodiment, preferably, the data preprocessing includes both data production processing and address text preprocessing. In addition to processing self-collected mass data, data production and processing also need to screen and fuse databases of various platforms (such as high-grade, baidu and the like) to acquire more information data related to the geographic position, and the better the data quality and the larger the magnitude, the more accurate the result is. Further, the server obtains the characteristic data of the geographic position by adopting one or more processing modes of 'denoising, standardization, structuralization' and the like for the data information after production and processing.

S32: the search engine inquires related data in the address database and matches the related data, and geographic position latitude and longitude information contained in the house code or the landmark picture is taken out;

in the present embodiment, the focus of search matching is the segmentation of the input text and scoring of the recall result. If the segmentation effect is not good, the proper record in the database cannot be found; if the scoring rules are set poorly, the output result may not be the desired result. Specifically, after the server obtains the characteristic data information of the target geographic position through a data preprocessing link, a search engine is started to search relevant data and is matched with description information of position coordinates stored in a database, and therefore the geographic target corresponding to the address information is identified; when the accurate positioning can not be carried out, a certain range of geographic targets matched with the address information are determined. The geographic target can be presented by a coordinate vector in a point, line or plane form, so that a bank operator can accurately acquire the longitude and latitude information of the geographic position contained in the house code or the landmark picture.

S33: and judging whether the search matching result is consistent with the actual address or not, and performing targeted position correction operation.

After acquiring the geographic position latitude and longitude information contained in the house code or the landmark picture, the bank operator compares the geographic position latitude and longitude information with actual address data input in the system, if the geographic position latitude and longitude information is consistent with the actual address data, correction is not needed, and the program is ended; if the coordinate is not completely matched with the coordinate, the position correction operation needs to be performed in a targeted manner according to the reference elements before the coordinate display of the final result is performed.

Example 4

As shown in fig. 4, the present embodiment provides a geocoding-based address location correction system, which includes: a barcode identification module 401, an image parsing module 402, a geocoding module 403, and a result plotting module 404. Wherein:

the barcode identification module 401 is configured to identify a house code or a house code two-dimensional code to obtain position information therein;

in this embodiment, the server analyzes the location information (such as province, city, district, town, county, street, building, house number, etc.) of the house with the target address through the barcode recognition module, and the bank operator can search in the database according to the information to obtain more geographic location data of the address and the address attachment.

The image analysis module 402 is used for extracting the characteristics of the shot landmark building image, mapping the characteristics of the landmark building image to a neural network for image recognition and classification, screening out pictures which accord with actual landmarks, and acquiring position information in the pictures;

the image parsing module 402 includes: an image preprocessing unit 4021, a feature extraction unit 4022, a neural network mapping unit 4023, and an identification classification unit 4024. Wherein:

the above-described image preprocessing unit 4021 is configured to: the server carries out a series of processing such as deletion, smoothing, binarization, amplitude normalization and the like on useless information in original data contained in the shot and uploaded landmark building image to obtain various characteristics of the landmark picture, including but not limited to characteristics such as color, texture, shape, spatial relationship and the like;

the above-described feature extraction unit 4022 is configured to: extracting various characteristics of the landmark picture through a neural network image recognition system, and performing final characteristic description display;

the neural network mapping unit 4023 is configured to: the server inputs the final feature description into a neural network, and the neural network is compared with the labels stored in the system to form a mapping relation;

the above-mentioned identification and classification unit 4024 is configured to: and the server performs image recognition and classification according to the mapping result so as to screen out the pictures which accord with the actual landmarks.

A geocoding module 403, configured to geocode the geographic location information obtained in the barcode identification module or the image analysis module, obtain longitude and latitude information therein, and perform a targeted location correction operation;

a geocoding module 403 comprising: a data preprocessing unit 4031, a search matching unit 4032, and a positioning correction unit 4033. Wherein:

the data preprocessing unit 4031 is configured to: the server screens and fuses from each platform database according to the geographic position information to acquire more data information related to the geographic position; and the number of the first and second groups,

one or more of the following processing modes are carried out on the data information: denoising, standardizing and structuring to obtain the geographic position characteristic data;

the search matching unit 4032 is configured to: the search engine inquires related data in an address database and matches the geographic position characteristic data, so that the geographic position latitude and longitude information contained in the house code or the landmark picture is accurately taken out;

the above-described positioning correction unit 4033 is configured to: the server judges whether the search matching result is consistent with the actual address or not, and performs targeted position correction operation.

And a result plotting module 404, configured to perform coordinate display of a final result according to the geocoded data.

In this embodiment, the bank operator checks the actual address information based on the geocoding result, and after confirming that there is no error, the program is ended, and the accurate location information of the target address is obtained.

Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims. It will be understood that the invention is not limited to the precise arrangements that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.

Claims (10)

1. An address location correction method based on geocoding is characterized by comprising the following steps:

comparing the mobile equipment positioning information with the actual position information and judging whether the mobile equipment positioning information needs to be corrected or not;

scanning the house code two-dimensional code or shooting a landmark building, uploading the house code two-dimensional code or shooting the landmark building to a server, analyzing and then acquiring position information of the house code two-dimensional code or shooting the landmark building;

geocoding the position information to obtain longitude and latitude information, and comparing the longitude and latitude information with the actual position information;

and judging whether the position information data accords with the actual address or not, and presenting final address text information after position correction.

2. The address location correction method based on geocoding of claim 1, wherein the geocoding of the location information to obtain longitude and latitude information specifically comprises the following steps:

the server carries out data preprocessing on the geographic position information;

the search engine inquires relevant data in the address database and matches the relevant data, and geographic position longitude and latitude information contained in the house code or the landmark picture is taken out;

and judging whether the search matching result is consistent with the actual address or not, and displaying the coordinates of the final result.

3. The geocoding-based address location correction method of claim 1, wherein the server parses the landmark building image to obtain the location information therein by means of the GPS location information in the image file format Exif under the neural network image recognition technology.

4. The geocoding-based address location correction method of claim 2, wherein the data preprocessing comprises both data production processing and address text preprocessing.

5. The geocoding-based address location fix retrieval method of claim 2, wherein the search match is focused on segmentation of the input text and scoring of the recall result.

6. The address location correction method based on geocoding of claim 2, wherein before displaying the coordinates of the final result, if the search matching result does not completely match the input address, then interpolation is needed based on line elements; if the reference element is a surface or a point, the centroid of the surface is required to be acquired for coordinate correction.

7. The address location correction method based on geocoding of claim 3, wherein before the server returns the geographical location information, the neural network image recognition technology firstly extracts the image features in the system, and then maps the image features to the neural network for image recognition classification, thereby screening out the pictures which are in line with the current landmark building.

8. The address location correction method based on geocoding of claim 4, wherein the data production processing comprises processing of self-collected mass data and screening and obtaining on each platform to obtain the relevant data of the geographic location.

9. A geocoding-based address location correction system, comprising:

the bar code identification module is used for identifying the house code or the house code two-dimensional code to acquire position information in the house code or the house code two-dimensional code;

the image analysis module is used for extracting the characteristics of the shot landmark building image, mapping the characteristics of the landmark building image to a neural network for image recognition and classification, screening out pictures which accord with actual landmarks, and then acquiring position information in the pictures;

the geographic coding module is used for performing geographic coding on the geographic position information acquired in the bar code identification module or the image analysis module, acquiring longitude and latitude information in the geographic position information, and performing targeted position correction operation;

and the result plotting module is used for displaying the coordinates of the final result according to the geocoded data.

10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the processing method according to any one of claims 1 to 8.

Images