Video point location retrieval method and device and nonvolatile storage medium

Abstract

The invention discloses a video point location retrieval method and device and a nonvolatile storage medium. The method comprises the steps that initial retrieval address information of video point position retrieval to be carried out is obtained, and a video point position is the position where a video collection device is located; a matching address set matched with standard retrieval address information of the initial retrieval address information is determined from an address base, and the standard retrieval address information comprises specific address information of administrative division levels; determining a candidate set formed by the video point location address information corresponding to the addresses in the matched address set; and determining a target video point location according to the similarity between the candidate set and the standard retrieval address information. The technical problem that a traditional camera calling mode mostly depends on position keyword information or map auxiliary information, and a camera cannot be quickly called under the condition that longitude and latitude are unknown is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and more specifically, to a video point retrieval method, device and non-volatile storage medium. Background Art

[0002] With the rapid development of information technology, video surveillance, as an important means of security, has been widely used in public safety, traffic management, commercial operations and other fields. However, with the continuous increase in the number of surveillance cameras, the scale of video data has also shown explosive growth. How to quickly retrieve the desired camera images from the massive surveillance equipment points is a pain point that needs to be solved. The traditional video point retrieval method mainly searches by video point name. Since the search of video point name description is simple, the retrieval efficiency is low, and it is easy to miss the retrieval results. The traditional camera calling method mostly relies on location keyword information or map auxiliary information. When the longitude and latitude are unknown, the camera cannot be quickly retrieved.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method, device and non-volatile storage medium for retrieving video points, so as to at least solve the technical problem that the traditional camera calling method mostly relies on location keyword information or map auxiliary information, and cannot quickly call the camera when the longitude and latitude are unknown.

[0005] According to one aspect of an embodiment of the present application, a method for retrieving video points is provided, including: obtaining initial retrieval address information for video point retrieval to be performed, wherein the video point is the location of a video acquisition device; determining a matching address set that matches the standard retrieval address information of the initial retrieval address information from an address library, wherein the standard retrieval address information contains specific address information at the administrative division level; determining a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and determining a target video point based on the similarity between the candidate set and the standard retrieval address information.

[0006] In some embodiments of the present application, after obtaining the initial search address information for video point retrieval, the method also includes: obtaining a dictionary tree consisting of all address texts in the database; determining supplementary address information of the initial search address information from the dictionary tree; and determining standard search address information corresponding to the initial search address information based on the initial search address information and the supplementary address information.

[0007] In some embodiments of the present application, the address library is determined in the following manner: obtaining the original address description information and longitude and latitude information of all video points; reverse geocoding the longitude and latitude information to obtain standard address information, wherein the standard address information includes specific address information at the administrative level; obtaining a preset number of landmark address information within a preset range of longitude and latitude information; expanding the landmark address information in the format of the standard address information to obtain expanded address information; splicing the original address description information and the expanded address information to obtain the address library, wherein the address library includes multi-dimensional semantic description information of all video points.

[0008] In some embodiments of the present application, determining a candidate set consisting of video point address information corresponding to addresses in a matching address set includes: obtaining a physical address of a video acquisition device that has a mapping relationship with the addresses in the matching address set; determining all video point address information corresponding to the addresses in the matching address set based on the physical address of the video acquisition device; and obtaining a candidate set based on all video point address information.

[0009] In some embodiments of the present application, the similarity between the candidate set and the standard retrieval address information is determined in the following manner: determining a first similarity between the address information in the candidate set and the standard retrieval address information, wherein the first similarity is used to represent the similarity between the address element characters of the address information in the candidate set and the standard retrieval address information; determining a second similarity between the address information in the candidate set and the standard retrieval address information, wherein the second similarity is used to represent the similarity between the address semantic information of the address information in the candidate set and the standard retrieval address information; and determining the similarity between the candidate set and the standard retrieval address information based on the first similarity and the second similarity.

[0010] In some embodiments of the present application, determining a first similarity between address information in a candidate set and standard retrieval address information includes: dividing the first address information in the candidate set into multiple first character strings according to a preset length to obtain a first character string set, wherein the first address information is any address information in the candidate set; dividing the standard retrieval address information into multiple second character strings according to a preset length to obtain a second character string set; determining the intersection and union of the first character string set and the second character string set; and determining the first similarity based on the ratio of the intersection and the union.

[0011] In some embodiments of the present application, determining the second similarity between address information in a candidate set and standard retrieval address information includes: determining a first vector corresponding to a first address information in the candidate set, wherein the first address information is any one of the address information in the candidate set; determining a second vector corresponding to the standard retrieval address information; and determining the second similarity based on the first vector and the second vector.

[0012] According to another aspect of an embodiment of the present application, a device for retrieving video points is also provided, including: an acquisition module, used to obtain initial retrieval address information for video point retrieval to be performed, wherein the video point is the location of a video acquisition device; a first determination module, used to determine from an address library a matching address set that matches the standard retrieval address information of the initial retrieval address information, wherein the standard retrieval address information contains specific address information at the administrative division level; a second determination module, used to determine a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and a third determination module, used to determine the target video point based on the similarity between the candidate set and the standard retrieval address information.

[0013] According to another aspect of an embodiment of the present application, an electronic device is also provided, including: a memory for storing program instructions; a processor, connected to the memory, for executing program instructions to implement the following functions: obtaining initial retrieval address information for video point retrieval to be performed, wherein the video point is the location of a video acquisition device; determining from an address library a matching address set that matches the standard retrieval address information of the initial retrieval address information, wherein the standard retrieval address information contains specific address information at the administrative division level; determining a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and determining a target video point based on the similarity between the candidate set and the standard retrieval address information.

[0014] According to another aspect of an embodiment of the present application, a non-volatile storage medium is further provided, the non-volatile storage medium including a stored computer program, wherein the device where the non-volatile storage medium is located executes the above-mentioned video point retrieval method by running the computer program.

[0015] According to another aspect of an embodiment of the present application, a computer program product is also provided, including computer instructions, which implement the above-mentioned video point retrieval method when executed by a processor.

[0016] In an embodiment of the present application, the method of obtaining initial search address information for video point retrieval to be performed is adopted, wherein the video point is the location of the video acquisition device; determining a set of matching addresses that match the standard search address information of the initial search address information from an address library, wherein the standard search address information includes specific address information of the administrative division level; determining a candidate set consisting of the video point address information corresponding to the addresses in the matching address set; determining a target video point according to the similarity between the candidate set and the standard search address information, by determining the standard search address information corresponding to the initial search address information for video point retrieval, and then determining a set of matching addresses that match the standard search address information of the initial search address information from the address library to obtain a candidate set, calculating the similarity between each address in the candidate set and the standard search address information to find the target video point, thereby achieving the technical effect of quickly and accurately retrieving the camera, and further solving the technical problem that most traditional camera calling methods rely on location keyword information or map auxiliary information, and cannot quickly retrieve the camera when the longitude and latitude are unknown. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal for implementing a video point retrieval method provided according to an embodiment of the present application.

[0019] Figure 2 It is a flowchart of a video point retrieval method provided according to an embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of a process of building an address library according to an embodiment of the present application;

[0021] Figure 4 is a structural diagram of an information model provided according to an embodiment of the present application;

[0022] Figure 5 It is a structural schematic diagram of a two-step rapid retrieval module provided according to an embodiment of the present application;

[0023] Figure 6 It is a structural diagram of a video point retrieval device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] In order to better understand the embodiments of the present application, the technical terms involved in the embodiments of the present application are explained as follows:

[0027] N-gram: is an algorithm based on statistical language model. Its basic idea is to perform a sliding window operation of size N on the content of the text according to bytes, forming a sequence of byte segments of length N. Each byte segment is called a gram. The occurrence frequency of all grams is counted and filtered according to a pre-set threshold to form a key gram list, which is the vector feature space of the text. Each gram in the list is a feature vector dimension.

[0028] Jaccard similarity: also known as the Jaccard similarity coefficient, is used to compare the similarities and differences between finite sample sets. The larger the Jaccard coefficient value, the higher the sample similarity. Given two sets A and B, the Jaccard coefficient is defined as the ratio of the size of the intersection of A and B to the size of the union of A and B, and is defined as follows:

[0029]

[0030] When both sets A and B are empty, J(A,B) is defined as 1.

[0031] The Aho-Corasick (AC automaton) algorithm is mainly used to find multiple keywords in a text. Compared with the single keyword search algorithm (such as the KMP algorithm), the AC automaton can handle the search problem of multiple keywords more efficiently. The core is to build a multi-mode dictionary tree (also called a failure tree or automatic keyword search tree).

[0032] The information collected in the embodiments of the present application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or reject automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered.

[0033] In order to solve the problems existing in the related art, the present application embodiment provides a video point retrieval method, which can be run on Figure 1 In the computer terminal shown, the computer terminal is explained below.

[0034] The video point retrieval method embodiment provided in the embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal for implementing a video point retrieval method. Figure 1 As shown, the computer terminal 10 may include one or more (102a, 102b, ..., 102n are used to illustrate) processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions connected via a wired and / or wireless network. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art can understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0035] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0036] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the video point retrieval method in the embodiment of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned video point retrieval method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0037] The transmission module 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission module 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0038] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0039] It should be noted that, in some optional embodiments, the above Figure 1 The computer terminal shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of hardware elements and software elements. It should be noted that Figure 1 This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computer terminal described above.

[0040] In the above-mentioned operating environment, an embodiment of the present application provides an embodiment of a method for retrieving video points. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0041] Figure 2 is a flow chart of a video point retrieval method provided according to an embodiment of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0042] Step S202, obtaining initial search address information for video point search to be performed, wherein the video point is the location of the video acquisition device.

[0043] In the technical solution provided in step S202, after obtaining the initial search address information for the video point search, it is also necessary to obtain a dictionary tree consisting of all address texts in the database; determine the supplementary address information of the initial search address information from the dictionary tree; and determine the standard search address information corresponding to the initial search address information based on the initial search address information and the supplementary address information. By constructing a dictionary tree, the address information can be quickly retrieved and supplemented, improving the accuracy and efficiency of address resolution, and is suitable for the management and retrieval of large-scale address information, such as video surveillance point management in public places such as large shopping malls, airports, and railway stations.

[0044] The following is a specific embodiment:

[0045] The fuzzy address supplement module is called to perform fuzzy supplement on the initial search address information, and the fuzzy address information to be searched is converted into standard search address information based on the AC automaton algorithm. Specifically: first, all address texts in the address library are loaded, the root node of the dictionary tree is constructed, each address text is recursively read, and the address text is split into individual independent words, and then new nodes are inserted in sequence according to the continuous existence of the words, and the address text after omitting the keywords representing the administrative division level is used as the new address text, and then a new branch is established from the root node, and finally the failure pointer between nodes is further constructed using the breadth-first search algorithm. When matching, an empty list of candidate phrases and an empty string of candidate words are established, and each word in the initial search address information is read in turn. If it can match a continuous phrase in the tree structure constructed in the AC automaton algorithm, it will continue to execute. When the candidate word characters composed of the newly added words cannot match the continuous phrases in the tree structure, the original candidate word string is added to the candidate word list as a candidate phrase, and finally the standard search address information is obtained after screening.

[0046] It should be noted that there are many ways to determine the address library in the above steps. For example, the address library can be determined in the following ways: obtain the original address description information and longitude and latitude information of all video points; reverse geocode the longitude and latitude information to obtain standard address information, wherein the standard address information includes specific address information at the administrative level; obtain a preset number of landmark address information within a preset range of longitude and latitude information; expand the landmark address information in the format of the standard address information to obtain expanded address information; splice the original address description information and the expanded address information to obtain the address library, wherein the address library includes multi-dimensional semantic description information of all video points. This method can ensure the standardization and richness of address information, and is suitable for video surveillance systems that require precise geographic positioning, such as urban traffic monitoring, public safety monitoring, etc.

[0047] The following is a specific embodiment:

[0048] Extract the video point information, separate the original address description information and longitude and latitude information for storage; call the map interface API for reverse geocoding based on the separated longitude and latitude information, convert the longitude and latitude into standard address information for storage, and expand the address information into a model containing city / district (county) / street (township) / community (administrative village) / specific address; enrich the semantics of the point information. The specific steps are as follows:

[0049] According to the separated longitude and latitude information, call the API to obtain 5 (the preset number mentioned above, which can be adjusted according to actual conditions) landmark address information POI (Point Of Interest) in the surrounding area (the preset range mentioned above, for example, the range within 5 kilometers from the selected address), and expand the address information on the basis of the standard address information to obtain the expanded address information; splice the original address description information of the video point with the expanded address information and store them in the address library to realize the multi-dimensional semantic information description of each video point.

[0050] Step S204: determining a set of matching addresses that match the standard search address information of the initial search address information from the address database, wherein the standard search address information includes specific address information at the administrative division level.

[0051] In the technical solution provided in step S204, determining the set of matching addresses that match the standard search address information of the initial search address information from the address library can be achieved in the following manner: according to the keyword matching method, the standard search address information is used as input, and the matching address information is quickly obtained in the address library to obtain the matching address set. For example, the following method can be used: extract key address elements for administrative division levels from the standard search address information, such as "city", "district (county)", "street (township)", "community (administrative village)", etc., and these keywords will be used in the subsequent matching process. In the address library, a keyword index is created for each stored address information. The index can be a keyword-based inverted list, so that address information containing specific keywords can be quickly located. Match the keywords of the standard search address information with the keyword index in the address library, and find the address information in the address library with a matching degree ranked in the front preset number of digits (for example, 1) as the matching address set.

[0052] Step S206: determine a candidate set consisting of video point address information corresponding to the addresses in the matching address set.

[0053] In the technical solution provided in step S206, there are multiple ways to determine the candidate set consisting of the video point address information corresponding to the addresses in the matching address set. For example, the candidate set consisting of the video point address information corresponding to the addresses in the matching address set can be determined in the following ways: obtaining the physical address of the video acquisition device that has a mapping relationship with the addresses in the matching address set; determining all the video point address information corresponding to the addresses in the matching address set based on the physical address of the video acquisition device; and obtaining the candidate set based on all the video point address information. This method can ensure the accuracy and comprehensiveness of the candidate set and is suitable for emergency monitoring systems that require rapid response, such as real-time monitoring of natural disasters such as fires and earthquakes.

[0054] In the above steps, the similarity between the candidate set and the standard search address information can be determined in the following manner: determining the first similarity between the address information in the candidate set and the standard search address information, wherein the first similarity is used to represent the similarity between the characters of the address elements in the candidate set and the standard search address information; determining the second similarity between the address information in the candidate set and the standard search address information, wherein the second similarity is used to represent the similarity between the address semantic information in the candidate set and the standard search address information; determining the similarity between the candidate set and the standard search address information based on the first similarity and the second similarity. This method combines character similarity and semantic similarity, can more accurately match the target video point, and is suitable for scenarios requiring high-precision retrieval.

[0055] There are many ways to determine the first similarity between the address information in the candidate set and the standard search address information, for example: divide the first address information in the candidate set into multiple first character strings according to a preset length to obtain a first character string set, wherein the first address information is any address information in the candidate set; divide the standard search address information into multiple second character strings according to a preset length to obtain a second character string set; determine the intersection and union of the first character string set and the second character string set; determine the first similarity based on the ratio of the intersection and the union. This method calculates the similarity by string matching, is suitable for the preliminary screening of address information, and improves the efficiency of subsequent semantic matching.

[0056] There are also many ways to determine the second similarity between the address information in the candidate set and the standard search address information, for example: determine the first vector corresponding to the first address information in the candidate set, where the first address information is any address information in the candidate set; determine the second vector corresponding to the standard search address information; determine the second similarity based on the first vector and the second vector. This method calculates semantic similarity through a vector space model, is suitable for deep understanding and matching of address information, and improves the accuracy of video point retrieval.

[0057] The following is a specific embodiment:

[0058] The device identification address mapping module is called for the matching address set to perform device mapping to obtain device information (i.e. the physical address of the above-mentioned video acquisition device, such as device_mac), and then all address information at the point where the matching address set is located is reversely retrieved based on the device_mac information and included in the candidate set.

[0059] Perform WSC similarity calculation on each address information in the candidate set and the standard search address information, where WSC similarity calculation is weighted similarity calculation, namely Weighted Similarity Calculations, referred to as WSC, which is an independently designed address similarity calculation module. It adds the cosine similarity of address element characters and address semantic information and normalizes them. The address semantic similarity calculation model is a comprehensive model that integrates the N-gram-based method with the address semantic cosine similarity calculation. The N-gram-based method can more accurately calculate the similarity between different types of address elements (characters), while the address semantic similarity calculation can make up for the defects of the address element character similarity calculation and obtain the similarity value of the overall semantics of the address. The calculation process of the WSC (Weighted Similarity Calculations) model is as follows:

[0060] (1) Determine the first similarity between the address information in the candidate set and the standard retrieval address information. Let addr_1 represent the standard retrieval address information and addr_2 represent the address information of any address in the candidate set. Perform a similarity matching calculation between the two. The first similarity is obtained through the following formula:

[0061]

[0062] Among them, Simi_1(addr1, addr2) represents the first similarity, G_N(addr_1) represents the first string, G_N(addr_2) represents the second string, addr_1 represents the standard retrieval address information, addr_2 represents the address information of any address in the candidate set, and |.| represents the number of elements in the set. The value range of the Simi_1 similarity is between 0 and 1, and the larger the value, the more similar the two address strings are.

[0063] The above G_N(addr_1) and G_N(addr_2) can be determined through the following formula:

[0064] The calculation of the character similarity of address elements adopts the N-gram-based method. The N-gram-based method extracts features by splitting the text into a sequence of consecutive N items (usually characters or words). In the calculation of the character similarity of address elements, the N-gram method can effectively measure the similarity between two address strings:

[0065] Suppose S = {s1, s2... s n} is an address string of length n, where s i represents the i-th character in the string. For any integer N (usually N < n), the N-gram set G_N(S) can be defined as:

[0066] G_N(S) = {s i ... s i+N-1 | 1 <= i <= n - N + 1},

[0067] Among them, {s i ... s i+N-1} represents the substring starting from the character s i and with a length of N.

[0068] Similarly, the N-gram sets (G_N(addr_1)) and (G_N(addr_2)) of the two address strings (addr_1) and (addr_2) are calculated based on the above formula. Then, the intersection and union of these two sets are calculated to measure the similarity between the two address strings. The intersection (G_N(addr_1)∩G_N(addr_2)) contains the N-grams shared by the two address strings, while the union (G_N(addr_1)∩G_N(addr_2)) contains all the N-grams in the two address strings. The first similarity can be obtained by calculating the ratio of the intersection and the union.

[0069] (2) Determine the second similarity between the address information in the candidate set and the standard search address information: Calculate the cosine similarity (second similarity) of the address semantic information of addr_1 and addr_2. The second similarity can be obtained by the following formula:

[0070]

[0071] Among them, Simi_2(addr_1, addr_2) represents the second similarity, k i ,q i They are the address semantic matrices corresponding to addr_1 and addr_2 respectively, k and q represent the second vector corresponding to the standard retrieval address information and the first vector corresponding to the first address information in the candidate set respectively, and the first address information is any address information in the candidate set.

[0072] (3) Add the address element character similarity and the cosine similarity of the address semantic information and normalize them to obtain the final address similarity value. This can be expressed by the following formula:

[0073] addr_simi(addr_1, addr_2)=(Simi_1+Simi_2) / 2

[0074] Wherein, addr_simi(addr_1, addr_2) represents the similarity between the candidate set and the standard retrieval address information.

[0075] Step S208, determining the target video point according to the similarity between the candidate set and the standard search address information.

[0076] In the technical solution provided in step S208, the final distance queue is determined based on the similarity between the candidate set and the standard retrieval address information (i.e., the similarity is sorted from high to low), and the preset number of items (e.g., 5 pieces of location information) are obtained as candidate cameras (i.e., the above-mentioned target video points).

[0077] Through the above steps S202-S208, an address standardization method is proposed: by extracting video point information, separating the original address description information and latitude and longitude information, calling the map interface API for reverse geocoding and POI location extraction, and realizing the semantic information description of each video point through expansion and splicing to form an address library; a similarity algorithm (WSC) is proposed to integrate the address element similarity calculation and the address semantic information similarity calculation to obtain the final similarity, and obtain more accurate relevant camera point information. A two-step fast retrieval method is proposed: using keyword retrieval for the first step of preliminary screening to obtain an alternative address set, reducing the calculation amount of the similarity calculation module, and in the second step, calling the similarity calculation module in the alternative address set to obtain the closest address information, thereby improving the retrieval efficiency and accuracy. Through the initial screening of keywords, the number of addresses that need to be calculated for similarity is greatly reduced, thereby significantly improving the retrieval efficiency. Refined similarity calculation is performed in the alternative address set to ensure the accuracy and speed of retrieving cameras. Standardize and expand the camera point information to form a unified address library. By standardizing and expanding camera location information, the problem of single and inconsistent addresses was solved, forming a unified, easy-to-manage and searchable address library.

[0078] It solves the technical problem that the traditional camera calling method mostly relies on location keyword information or map auxiliary information, and cannot quickly call the camera when the longitude and latitude are unknown. Due to the diversity and non-standardization of address expressions, traditional address recognition methods based on string matching often have problems with mismatches and missed matches, and cannot accurately call the desired camera images from a large number of cameras in a short period of time. The method proposed in this application solves the above problems. In scenarios such as urban security monitoring, traffic management, and emergency response, it can quickly and accurately locate video surveillance points in specific areas, thereby improving monitoring efficiency and response speed.

[0079] Figure 3 is a schematic diagram of a process for constructing an address library according to an embodiment of the present application, such as Figure 3 As shown, first obtain the original information and longitude and latitude information of the video point to be retrieved, standardize the above original information and longitude and latitude information through the address standardization module, convert them into standard address information, and expand the address through the address expansion module, expand the landmark address information according to the format of standard address information, obtain the expanded address information, and display it through the information module; splice the original address description information and the expanded address information to obtain the address library, wherein the address library includes multi-dimensional semantic description information of all video points. The specific standardization and expansion methods have been described in detail in the description of the above embodiments and will not be repeated here. Figure 3In the example, E / SS / N represents longitude and latitude information. The original information and longitude and latitude information are reverse geocoded through the map retrieval interface, and the longitude and latitude are converted into standard address information for storage. The address information is expanded into a model containing city / district (county) / street (township) / community (administrative village) / specific address. The API is called based on the longitude and latitude information to obtain the surrounding 5 landmark address information. The address information is expanded based on the standard address information, and the original address description information of the video point is spliced ​​with the expanded address information to achieve multi-dimensional semantic information description of each video point. Figure 3 In the example, the original information is XX Building, and the longitude and latitude information is 117.489157, 30.655037. After reverse geocoding through the map retrieval interface, the longitude and latitude are converted into standard address information storage, and the address information is expanded to become No. XX, XX Road, XX District, XX City, XX Province. Five landmark address information containing original information, landmark buildings and historical information are expanded: XX Building, XX Primary School, XX Bookstore, XX Supermarket, XX Shopping Mall, and the device (camera) corresponding to the standard address information is mapped to BC-0F-F4-7A-E6-FF through device retrieval. The structure of the information model is as follows Figure 4 As shown, it includes device identification, standard address information and extended information.

[0080] Figure 5 is a schematic diagram of a two-step rapid retrieval module according to an embodiment of the present application. Figure 5 As shown, STEP1 is the first step of detection, first input the information to be retrieved (i.e. the above-mentioned initial retrieval address information), supplement the fuzzy address of the information to be retrieved, obtain the standard retrieval information (i.e. the above-mentioned standard retrieval address information), and perform keyword matching with the address library to obtain a matching set (i.e. the above-mentioned matching address set); STEP2 is the second step of detection, first map the matching set with the device through the device identification mapping module, so as to obtain a candidate set (i.e. the candidate set consisting of the video point address information corresponding to the address in the above-mentioned determined matching address set), calculate the similarity between the candidate set and the standard retrieval information, so as to obtain a distance queue, and take the first preset number of the smallest distance in the distance queue as the final target video point. The implementation method of the above steps has been described in detail in the description of the above embodiments, and will not be repeated here.

[0081] Figure 6 is a schematic diagram of a structure of a video point retrieval device provided according to an embodiment of the present application, such as Figure 6 As shown, including:

[0082] The acquisition module 602 is used to acquire the initial search address information for the video point search to be performed, wherein the video point is the location of the video acquisition device.

[0083] The acquisition module 602 is also used to obtain a dictionary tree consisting of all address texts in the database; determine the supplementary address information of the initial search address information from the dictionary tree; and determine the standard search address information corresponding to the initial search address information based on the initial search address information and the supplementary address information.

[0084] The acquisition module 602 is also used to obtain the original address description information and longitude and latitude information of all video points; reverse geocode the longitude and latitude information to obtain standard address information, wherein the standard address information includes specific address information at the administrative level; obtain a preset number of landmark address information within a preset range of longitude and latitude information; expand the landmark address information in the format of the standard address information to obtain expanded address information; splice the original address description information and the expanded address information to obtain an address library, wherein the address library includes multi-dimensional semantic description information of all video points.

[0085] The first determination module 604 is used to determine a set of matching addresses that match the standard search address information of the initial search address information from the address database, wherein the standard search address information includes specific address information of the administrative division level.

[0086] The second determination module 606 is used to determine a candidate set consisting of video point address information corresponding to the addresses in the matching address set.

[0087] The second determination module 606 is also used to obtain the physical address of the video acquisition device that has a mapping relationship with the addresses in the matching address set; determine all video point address information corresponding to the addresses in the matching address set based on the physical address of the video acquisition device; and obtain a candidate set based on all video point address information.

[0088] The third determination module 608 is used to determine the target video point according to the similarity between the candidate set and the standard search address information.

[0089] The third determination module 608 is also used to determine a first similarity between the address information in the candidate set and the standard retrieval address information, wherein the first similarity is used to represent the similarity between the address element characters of the address information in the candidate set and the standard retrieval address information; determine a second similarity between the address information in the candidate set and the standard retrieval address information, wherein the second similarity is used to represent the similarity between the address semantic information of the address information in the candidate set and the standard retrieval address information; and determine the similarity between the candidate set and the standard retrieval address information based on the first similarity and the second similarity.

[0090] It should be noted that Figure 6 The video point retrieval device shown is used to perform Figure 2 The retrieval method of the video point shown is therefore Figure 2The relevant explanations in the retrieval method of the video point in are also applicable to the retrieval device of the video point, and will not be repeated here.

[0091] An embodiment of the present application also provides an electronic device, which includes a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute program instructions to implement the following functions: obtaining initial retrieval address information for video point retrieval to be performed, wherein the video point is the location of the video acquisition device; determining a matching address set that matches the standard retrieval address information of the initial retrieval address information from an address library, wherein the standard retrieval address information contains specific address information at the administrative division level; determining a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and determining the target video point based on the similarity between the candidate set and the standard retrieval address information.

[0092] An embodiment of the present application also provides a non-volatile storage medium, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the following video point retrieval method by running the computer program: obtaining initial retrieval address information for video point retrieval to be performed, wherein the video point is the location of the video acquisition device; determining a matching address set that matches the standard retrieval address information of the initial retrieval address information from an address library, wherein the standard retrieval address information includes specific address information at the administrative division level; determining a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and determining the target video point based on the similarity between the candidate set and the standard retrieval address information.

[0093] The embodiment of the present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the video point retrieval method in each embodiment of the present application.

[0094] The embodiments of the present application also provide a computer program, which, when executed by a processor, implements the steps of the video point retrieval method in each embodiment of the present application.

[0095] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0096] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0097] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0098] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0099] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0100] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0101] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A video point retrieval method, characterized in that: include: Acquire initial search address information for a video point to be searched, wherein the video point is a location where a video acquisition device is located; Determine from the address database a set of matching addresses that match the standard search address information of the initial search address information, wherein the standard search address information includes specific address information at the administrative division level; Determine a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and determine a target video point based on the similarity between the candidate set and the standard search address information.

2. The method according to claim 1, characterized in that: After obtaining the initial search address information for the video point search, the method further includes: Get the dictionary tree consisting of all address texts in the database; Determining supplementary address information of the initial search address information from the dictionary tree; The standard search address information corresponding to the initial search address information is determined according to the initial search address information and the supplementary address information.

3. The method according to claim 1, characterized in that: The address library is determined by: Get the original address description information and longitude and latitude information of all video points; Reverse geocoding the latitude and longitude information to obtain standard address information, wherein the standard address information includes specific address information at the administrative division level; Acquire a preset number of landmark address information within a preset range of the latitude and longitude information; Expanding the landmark address information according to the format of the standard address information to obtain expanded address information; The original address description information and the expanded address information are spliced ​​to obtain the address library, wherein the address library includes multi-dimensional semantic description information of all video points.

4. The method according to claim 1, characterized in that Determining a candidate set consisting of video point address information corresponding to the addresses in the matching address set includes: Acquire the physical address of the video acquisition device that has a mapping relationship with the addresses in the matching address set; Determine all video point address information corresponding to the addresses in the matching address set according to the physical address of the video acquisition device; The candidate set is obtained according to the address information of all video points.

5. The method according to claim 1, characterized in that The similarity between the candidate set and the standard search address information is determined in the following manner: Determine a first similarity between the address information in the candidate set and the standard search address information, wherein the first similarity is used to represent the similarity between characters of address elements of the address information in the candidate set and the standard search address information; Determine a second similarity between the address information in the candidate set and the standard search address information, wherein the second similarity is used to represent the similarity of address semantic information between the address information in the candidate set and the standard search address information; The similarity between the candidate set and the standard search address information is determined according to the first similarity and the second similarity.

6. The method according to claim 5, characterized in that Determining a first similarity between the address information in the candidate set and the standard search address information includes: Divide the first address information in the candidate set into a plurality of first character strings according to a preset length to obtain a first character string set, wherein the first address information is any one of the address information in the candidate set; Dividing the standard search address information into a plurality of second character strings according to the preset length to obtain a second character string set; Determine the intersection and union of the first string set and the second string set; The first similarity is determined according to the ratio of the intersection to the union.

7. The method according to claim 5, characterized in that Determining a second similarity between the address information in the candidate set and the standard search address information includes: Determine a first vector corresponding to first address information in the candidate set, wherein the first address information is any address information in the candidate set; Determine a second vector corresponding to the standard search address information; The second similarity is determined according to the first vector and the second vector.

8. A video point retrieval device, characterized in that: include: An acquisition module, used to acquire initial search address information for a video point search to be performed, wherein the video point is a location where a video acquisition device is located; A first determination module is used to determine a set of matching addresses that match the standard search address information of the initial search address information from an address database, wherein the standard search address information includes specific address information of the administrative division level; A second determination module is used to determine a candidate set consisting of video point address information corresponding to the addresses in the matching address set; The third determination module is used to determine the target video point according to the similarity between the candidate set and the standard search address information.

9. An electronic device, characterized in that: include: A memory for storing program instructions; A processor is connected to the memory and is used to execute program instructions to implement the following functions: obtaining initial search address information for video point retrieval to be performed, wherein the video point is the location of a video acquisition device; determining a matching address set that matches the standard search address information of the initial search address information from an address library, wherein the standard search address information contains specific address information at the administrative division level; determining a candidate set consisting of video point address information corresponding to the addresses in the matching address set; and determining a target video point based on a similarity between the candidate set and the standard search address information.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the video point retrieval method described in any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the video point retrieval method described in any one of claims 1 to 7 is implemented.

Images