WO2020238556A1 - 基于配置平台的数据传输方法、装置和计算机设备 - Google Patents

基于配置平台的数据传输方法、装置和计算机设备 Download PDF

Info

Publication number
WO2020238556A1
WO2020238556A1 PCT/CN2020/088049 CN2020088049W WO2020238556A1 WO 2020238556 A1 WO2020238556 A1 WO 2020238556A1 CN 2020088049 W CN2020088049 W CN 2020088049W WO 2020238556 A1 WO2020238556 A1 WO 2020238556A1
Authority
WO
WIPO (PCT)
Prior art keywords
text
picture
stored
designated
preset
Prior art date
Application number
PCT/CN2020/088049
Other languages
English (en)
French (fr)
Inventor
唐奥强
Original Assignee
深圳壹账通智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳壹账通智能科技有限公司 filed Critical 深圳壹账通智能科技有限公司
Publication of WO2020238556A1 publication Critical patent/WO2020238556A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/14—Image acquisition
    • G06V30/148—Segmentation of character regions
    • G06V30/153—Segmentation of character regions using recognition of characters or words
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition

Definitions

  • This application relates to the computer field, and in particular to a data transmission method, device, computer equipment and storage medium based on a configuration platform.
  • Recognizing the text in the picture is a technology often used in current production and life.
  • the traditional text recognition method is that the upload terminal directly uploads the picture including the text to the recognition terminal, then the recognition terminal recognizes the picture to obtain the recognized text, and the recognition terminal sends the recognized text to the upload terminal.
  • the inventor realizes that this traditional method requires repeated recognition for repeated pictures, which wastes computing power; and cannot send recognized text to third-party terminals in real time. If the recognized text is to be sent to third-party terminals, further uploading is required.
  • the terminal sends the recognized text to a third-party terminal, thereby adding another step to the information transmission process, wasting resources and being unable to display the recognized text in real time at the same time.
  • the main purpose of this application is to provide a data transmission method, device, computer equipment, and storage medium based on a configuration platform to reduce additional network overhead caused by repeated sending of text information corresponding to repeated pictures.
  • this application proposes a data transmission method based on a configuration platform, which includes the following steps: receiving a text recognition application sent by an upload terminal, where the text recognition application specifies a text display terminal, and the text recognition application carries There is a designated picture; the predetermined picture similarity judgment method is used to determine whether the designated picture is similar to the pre-stored picture; if the designated picture is not similar to the pre-stored picture, the designated picture is sent to the designated text recognition terminal, so The designated text recognition terminal is used to recognize the text content in the designated picture as picture text text; receive the picture text text sent by the designated text recognition terminal, and use a preset text similarity calculation method to calculate the The similarity value between the text of the picture and the prestored text, and determine whether the similarity value is greater than a preset similarity threshold, wherein the prestored text refers to the text obtained by performing text recognition on the prestored picture; If the similarity value is not greater than a preset similarity
  • This application provides a data transmission device based on a configuration platform, including: a text recognition application receiving unit for receiving a text recognition application sent by an upload terminal, the text recognition application specifies a text display terminal, and the text recognition application carries There are designated pictures; the picture similarity judgment unit is used to use a preset picture similarity judgment method to judge whether the designated picture is similar to the pre-stored picture; the designated picture sending unit is used if the designated picture is not similar to the pre-stored picture , The designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture as picture text text; the picture text text receiving unit is used to receive the designated text Recognize the picture text text sent by the terminal, calculate the similarity value between the picture text text and the pre-stored text by using a preset text similarity calculation method, and determine whether the similarity value is greater than a preset similarity threshold , Wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture
  • the present application provides a computer device including a memory and a processor, the memory stores a computer program, and the processor implements the steps of any one of the above methods when the computer program is executed.
  • the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.
  • This application reduces the additional network overhead caused by repeated sending of text information corresponding to repeated pictures, and realizes real-time display of recognized text.
  • FIG. 1 is a schematic flowchart of a data transmission method based on a configuration platform according to an embodiment of the application;
  • FIG. 2 is a schematic block diagram of the structure of a data transmission device based on a configuration platform according to an embodiment of the application;
  • FIG. 3 is a schematic block diagram of the structure of a computer device according to an embodiment of the application.
  • an embodiment of the present application provides a data transmission method based on a configuration platform, including the following steps:
  • Recognizing the text in the picture is a technology often used in current production and life.
  • the traditional text recognition method is that the upload terminal directly uploads the picture including the text to the recognition terminal, then the recognition terminal recognizes the picture to obtain the recognized text, and the recognition terminal sends the recognized text to the upload terminal.
  • This traditional method requires repeated recognition for duplicate pictures, wasting computing power; and it cannot send the recognized text to a third-party terminal in real time. If the recognized text is to be sent to a third-party terminal, the upload terminal needs to further send the recognized text. To a third-party terminal, there is an extra step of sending information, a waste of resources, and it is impossible to display recognized text in real time.
  • This application uses the configuration platform to control the data transfer process, and uses two similarity calculations to ensure that the specified picture is a new picture, and then the text of the picture recognized from the specified picture is sent to the specified text display terminal, thereby It has realized the technical effect of saving computing power, saving process, real-time and displaying recognized text at the same time.
  • a text recognition application sent by an upload terminal is received, the text recognition application specifies a text display terminal, and the text recognition application carries a specified picture.
  • the execution subject of this application is a configuration platform, which is used to control the data transfer process between the upload terminal, the recognition terminal, and the text display terminal, and is preferably a configuration platform based on big data technology, such as the big data engine spark , Using a large database HBase, etc., so that the configuration platform can store and process a large amount of data from a large number of users.
  • the upload terminal may be any terminal, such as a terminal with a scanning function, a terminal with a photographing function, etc., preferably a terminal that can perform image collection to obtain a specified picture.
  • the designated picture is a picture with text.
  • a preset picture similarity determination method is used to determine whether the designated picture is similar to a pre-stored picture.
  • the preset picture similarity judgment method is, for example: sequentially comparing the corresponding pixels in the specified picture and the pre-stored picture, and if the number of the same pixel points in all the pixels is greater than a predetermined threshold, then it is determined If the designated picture is similar to the pre-stored picture, it means that the designated picture is pre-stored in the configuration platform and is a duplicate picture.
  • the pre-stored picture Since the pre-stored picture has already used text recognition technology to recognize the text, there is no need to perform another recognition; if If the proportion of the number of the same pixels in the number of all pixels is not greater than a predetermined threshold, it is determined that the specified picture is not similar to the pre-stored picture, indicating that the specified picture is a new picture and text recognition technology is not used for recognition, Therefore, identification is required.
  • step S3 if the designated picture is not similar to the pre-stored picture, the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture as Picture text text. If the designated picture is not similar to the pre-stored picture, then the designated picture needs text recognition, so the designated picture is sent to the designated text recognition terminal.
  • This application adopts the setting of separating the designated text recognition terminal from the configuration platform to separate the text recognition function and the designated text recognition terminal for recognition.
  • the recognition terminal of the corresponding text recognition algorithm can be used for the recognition according to the specific picture. Hierarchical management, saving resources (simple pictures can be identified by simple algorithms), and easy error tracking.
  • the technology for recognizing the text content in the specified picture as the text text of the picture can use the existing mature recognition technology, such as OCR technology (OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition), which is available). (The image information is converted into usable computer input technology), which will not be repeated here.
  • OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition), which is available).
  • Optical Character Recognition Optical Character Recognition
  • the picture text text sent by the designated text recognition terminal is received, and the preset text similarity calculation method is used to calculate the similarity value between the picture text text and the pre-stored text, and determine the Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture.
  • the calculation of the similarity value between the image text text and the pre-stored text can be any method, for example: using WMD algorithm (word move's distance), based on cosine similarity, simhash algorithm, word frequency
  • the vector algorithm is based on the similarity value between the text of the picture and the pre-stored text.
  • the similarity threshold may be any value, such as 100%. Further, in order to avoid misjudgment, the similarity threshold may be set to be less than 100%, for example, greater than or equal to 80%.
  • the text text of the picture is sent to the text display terminal.
  • the similarity value is used to determine whether the text of the picture is the same as the pre-stored text, that is, whether the designated picture is the same as the pre-stored picture. If the similarity value is not greater than the preset similarity threshold, it indicates that the specified picture is a new picture. Therefore, the text display terminal does not save the text of the picture recognized by the specified picture, and converts the text of the picture accordingly The text is sent to the text display terminal to display the recognized text in real time and simultaneously. Further, if the similarity value is greater than the preset similarity threshold, it indicates that the specified picture is a duplicate picture.
  • the text display terminal saves the picture text text recognized by the specified picture, so there is no need to send the picture text text. It only needs to inform the text display terminal of the storage location of the picture text and text, and the text display terminal can directly call the local storage data, thereby saving network overhead.
  • the step S2 of judging whether the designated picture is similar to a pre-stored picture by using a preset picture similarity judgment method includes:
  • S201 Perform gray-scale processing on the designated picture and the pre-stored picture respectively to obtain a first gray-scale picture and a second gray-scale picture;
  • S202 Calculate the average value Am of the gray values of all pixels in the m-th column or the m-th row of the gray-scale picture, and calculate the average value B of the gray values of all the pixels in the gray-scale picture;
  • grayscale refers to the color representing a grayscale color.
  • the color represents a grayscale color
  • the gray scale range is, for example, 0-255 (when the values of R, G, and B are all 0-255, of course it will also change with the change of the value range of R, G, and B).
  • the gray-scale processing method can be any method, such as the component method, the maximum method, the average method, and the weighted average method. Among them, since there are only 256 value ranges for the gray value, comparison of pictures on this basis can greatly reduce the amount of calculation. Then calculate the average value Am of the gray values of all pixels in the m-th column or the m-th row of the gray-scale picture, and calculate the average value B of the gray values of all the pixels in the gray-scale picture.
  • the process of calculating the average value Am of the gray values of all pixels in the m-th column or m-th row of the gray-scale picture includes: collecting all the pixels in the m-th column or m-th row of the gray-scale picture Add the gray values of all pixels in the mth column or mth row, and divide the sum of the gray values obtained by the summation by the mth column or The number of all pixels in the m rows is the average value Am of the gray values of all the pixels in the mth column or mth row of the grayscale image.
  • the process of calculating the average value B of the gray values of all pixels in the gray image includes: calculating the sum of the gray values of all pixels in the gray image, and dividing the sum of the gray values by According to the number of pixels, the average value B of the gray values of all pixels in the gray image is obtained.
  • the overall variance is used to measure the average of the gray values Am of the pixels in the m-th column or the m-th row of the gray-scale image and the average of the gray-scale values of all pixels in the gray-scale image. The difference between the value B.
  • the gray value of the mth column or row of the first grayscale image is the same or approximately the same as the gray value of the mth column or row of the second grayscale image (approximate judgment to save computing power , And because the overall variance of the two different pictures is generally not equal, the accuracy of the judgment is very high), on the contrary, the gray value of the mth column or mth row of the first grayscale image is the same as the second grayscale value.
  • the gray value of the mth column or mth row of the picture is different.
  • the return value is The maximum value in. If it is less than the preset variance error threshold, it is determined that the specified picture is similar to the pre-stored picture. Thus, approximate judgment is used (because all gray values of grayscale pictures converted from two different pictures are generally not equal, and all grayscale values of grayscale pictures converted from the same picture are generally equal), it achieves less consumption Under the premise of computing resources, it is determined whether the specified picture is similar to the pre-stored picture.
  • the step S2 of judging whether the designated picture is similar to a pre-stored picture by using a preset picture similarity judgment method includes:
  • S203 Determine whether the proportion of the same pixel points is greater than a preset proportion threshold
  • the use of a preset picture similarity judgment method is realized to judge whether the specified picture is similar to the pre-stored picture.
  • the method includes:
  • S22 Search for the text text corresponding to the pre-stored picture in all texts corresponding to the upload terminal, so as to obtain position information of the text text corresponding to the pre-stored picture in all texts corresponding to the upload terminal;
  • the location information is sent to the text display terminal.
  • the designated picture is similar to the pre-stored picture, only the position information of the text corresponding to the pre-stored picture in all the texts corresponding to the upload terminal is sent to the text display terminal, and the text display terminal uses the position information All the text corresponding to the upload terminal can be retrieved from local storage, and the text content corresponding to the designated picture can be obtained according to the location information.
  • a Kafka message queue can also be used to transmit location information, that is, the location information is put into a preset Kafka message queue for consumption by the text display terminal.
  • Kafka is a high-throughput distributed publish-subscribe message queue, which can process all action flow data in consumer-scale websites.
  • the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture Step S3 is the text of the picture, including:
  • the designated picture is not similar to the pre-stored picture, use a preset preliminary text recognition technology to recognize a predetermined area in the picture text text picture to obtain a preliminary recognized text, wherein the area of the predetermined area is smaller than that of the designated picture area;
  • S302 Obtain the text text corresponding to the pre-stored picture according to the corresponding relationship between the preset picture and the pre-stored text text;
  • this embodiment adopts preliminary text recognition technology to further determine whether the designated picture is similar to the pre-stored picture.
  • the preliminary character recognition technology can be any character recognition technology, for example, OCR recognition technology.
  • the area of the predetermined area is smaller than the area of the designated picture, for example, an edge area of the designated picture.
  • this embodiment can further determine whether the designated picture is similar to the pre-stored picture under the premise of only consuming a small amount of computing power, and when the text corresponding to the pre-stored picture does not contain the preliminary recognized text (that is, the designated The picture is not similar to the pre-stored picture), the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture as text text of the picture.
  • Step S4 includes:
  • S401 Receive the picture text text sent by the designated text recognition terminal
  • the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture
  • similarity is the similarity value
  • A is the word frequency of the picture text text Vector
  • B is the word frequency vector of the pre-stored text
  • Ai is the number of occurrences of the i-th word of the image text
  • Bi is the number of occurrences of the i-th word of the pre-stored text
  • S403 Determine whether the similarity value is greater than a preset similarity threshold.
  • the similarity algorithm is calculated based on the cosine similarity of the text content of the two nodes to reflect the degree of similarity between the text content of the two nodes. When the value of similarity is closer to 1, the more similar is; the closer to 0, the less similar. Based on this, it is determined whether the similarity value is greater than a preset similarity threshold.
  • the similarity threshold is preferably 100%.
  • the receiving the picture text text sent by the designated text recognition terminal and using a preset text similarity calculation method to calculate the similarity value between the picture text text and the pre-stored text, and determine Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture after step S4, including:
  • the location information is sent to the text display terminal. If the similarity value is greater than the preset similarity threshold, it indicates that the specified picture is similar to the pre-stored picture (the text of the picture is similar to the pre-stored text), so only the pre-stored text in all texts corresponding to the upload terminal
  • the location information can be sent to the text display terminal, and the text display terminal can retrieve all the text corresponding to the upload terminal from the local storage through the location information, and learn the text content corresponding to the specified picture according to the location information .
  • a Kafka message queue can also be used to transmit location information, that is, the location information is put into a preset Kafka message queue for consumption by the text display terminal.
  • Kafka is a high-throughput distributed publish-subscribe message queue, which can process all action flow data in consumer-scale websites.
  • the method includes:
  • the similarity threshold is obtained by reducing the value of the preset standard threshold.
  • this application adopts obtaining the flow value sent to the text display terminal within a predetermined time; determining whether the flow value is greater than a preset flow threshold; if the flow value is greater than the preset flow threshold, Then, the similarity threshold is obtained by reducing the value of the preset standard threshold, and the value is reduced on the basis of the standard threshold to reduce the number of times of sending text to the text display terminal, thereby reducing the pressure on the configuration platform.
  • the data transmission method based on the configuration platform of this application receives the text recognition application sent by the upload terminal; determines whether the specified picture is similar to the prestored picture; if the specified picture is not similar to the prestored picture, then the specified picture is sent to A designated text recognition terminal, the designated text recognition terminal is used to recognize the text content in the designated picture as picture text text; receive the picture text text sent by the designated text recognition terminal, calculate the picture text text and The similarity value of the pre-stored text; if the similarity value is not greater than the preset similarity threshold, the text of the picture is sent to the text display terminal. This reduces the additional network overhead caused by the repeated sending of text information corresponding to the repeated pictures, and realizes real-time display of recognized text.
  • an embodiment of the present application provides a data transmission device based on a configuration platform, including:
  • the text recognition application receiving unit 10 is configured to receive a text recognition application sent by an upload terminal, where the text recognition application specifies a text display terminal, and the text recognition application carries a specified picture;
  • the picture similarity judgment unit 20 is configured to use a preset picture similarity judgment method to judge whether the designated picture is similar to a pre-stored picture;
  • the designated picture sending unit 30 is configured to send the designated picture to a designated text recognition terminal if the designated picture is not similar to the prestored picture, and the designated text recognition terminal is used to recognize the text content in the designated picture Is the text of the picture;
  • the picture text text receiving unit 40 is configured to receive the picture text text sent by the designated text recognition terminal, and use a preset text similarity calculation method to calculate the similarity value between the picture text text and the prestored text, and Judging whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture;
  • the picture text text sending unit 50 is configured to send the picture text text to the text display terminal if the similarity value is not greater than a preset similarity threshold.
  • Recognizing the text in the picture is a technology often used in current production and life.
  • the traditional text recognition method is that the upload terminal directly uploads the picture including the text to the recognition terminal, then the recognition terminal recognizes the picture to obtain the recognized text, and the recognition terminal sends the recognized text to the upload terminal.
  • This traditional method requires repeated recognition for duplicate pictures, wasting computing power; and it cannot send the recognized text to a third-party terminal in real time. If the recognized text is to be sent to a third-party terminal, the upload terminal needs to further send the recognized text. To a third-party terminal, there is an extra step of sending information, a waste of resources, and it is impossible to display recognized text in real time.
  • This application uses the configuration platform to control the data transfer process, and uses two similarity calculations to ensure that the specified picture is a new picture, and then the text of the picture recognized from the specified picture is sent to the specified text display terminal, thereby It has realized the technical effect of saving computing power, saving process, real-time and displaying recognized text at the same time.
  • a text recognition application sent by an upload terminal is received, where the text recognition application specifies a text display terminal, and the text recognition application carries a specified picture.
  • the execution subject of this application is a configuration platform, which is used to control the data transfer process between the upload terminal, the recognition terminal, and the text display terminal, and is preferably a configuration platform based on big data technology, such as the big data engine spark , Using a large database HBase, etc., so that the configuration platform can store and process a large amount of data from a large number of users.
  • the upload terminal may be any terminal, such as a terminal with a scanning function, a terminal with a photographing function, etc., preferably a terminal that can perform image collection to obtain a specified picture.
  • the designated picture is a picture with text.
  • a preset picture similarity determination method is used to determine whether the designated picture is similar to a pre-stored picture.
  • the preset picture similarity judgment method is, for example: sequentially comparing the corresponding pixels in the specified picture and the pre-stored picture, and if the number of the same pixel points in all the pixels is greater than a predetermined threshold, then it is determined If the designated picture is similar to the pre-stored picture, it means that the designated picture is pre-stored in the configuration platform and is a duplicate picture.
  • the pre-stored picture Since the pre-stored picture has already used text recognition technology to recognize the text, there is no need to perform another recognition; if If the proportion of the number of the same pixels in the number of all pixels is not greater than a predetermined threshold, it is determined that the specified picture is not similar to the pre-stored picture, indicating that the specified picture is a new picture and text recognition technology is not used for recognition, Therefore, identification is required.
  • the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture as Picture text text. If the designated picture is not similar to the pre-stored picture, then the designated picture needs text recognition, so the designated picture is sent to the designated text recognition terminal.
  • This application adopts the setting of separating the designated text recognition terminal from the configuration platform to separate the text recognition function and the designated text recognition terminal for recognition.
  • the recognition terminal of the corresponding text recognition algorithm can be used for the recognition according to the specific picture. Hierarchical management, saving resources (simple pictures can be identified by simple algorithms), and easy error tracking.
  • the technology for recognizing the text content in the specified picture as the text text of the picture can use the existing mature recognition technology, such as OCR technology (OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition), which is available). (The image information is converted into usable computer input technology), which will not be repeated here.
  • OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition), which is available).
  • Optical Character Recognition Optical Character Recognition
  • the picture text text sent by the designated text recognition terminal is received, and a preset text similarity calculation method is used to calculate the similarity value between the picture text text and the pre-stored text, and determine the Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture.
  • the calculation of the similarity value between the image text text and the pre-stored text can be any method, for example: using WMD algorithm (word move's distance), based on cosine similarity, simhash algorithm, word frequency
  • WMD algorithm word move's distance
  • cosine similarity based on cosine similarity
  • simhash algorithm word frequency
  • the vector algorithm is based on the similarity value between the text of the picture and the pre-stored text.
  • the similarity threshold may be any value, such as 100%. Further, in order to avoid misjudgment, the similarity threshold may be set to be less than 100%, for example, greater than or equal to 80%.
  • the text of the picture is sent to the text display terminal.
  • the similarity value is used to determine whether the text of the picture is the same as the pre-stored text, that is, whether the designated picture is the same as the pre-stored picture. If the similarity value is not greater than the preset similarity threshold, it indicates that the specified picture is a new picture. Therefore, the text display terminal does not save the text of the picture recognized by the specified picture, and converts the text of the picture accordingly The text is sent to the text display terminal to display the recognized text in real time and simultaneously. Further, if the similarity value is greater than the preset similarity threshold, it indicates that the specified picture is a duplicate picture.
  • the text display terminal saves the picture text text recognized by the specified picture, so there is no need to send the picture text text. It only needs to inform the text display terminal of the storage location of the picture text and text, and the text display terminal can directly call the local storage data, thereby saving network overhead.
  • the picture similarity judgment unit 20 includes:
  • the gray-scale processing subunit is used to perform gray-scale processing on the designated picture and the pre-stored picture to obtain a first gray-scale picture and a second gray-scale picture;
  • the average gray value calculation subunit is used to calculate the average value Am of the gray values of all pixels in the m-th column or m-th row of the gray-scale image, and to calculate all the pixels in the gray-scale image
  • the overall variance gets the sub-unit, which is used according to the formula: Calculate the overall variance of the m-th column or m-th row of the grayscale image Where N is the total number of columns or rows in the grayscale picture;
  • the difference of the overall variance gets the sub-unit, which is used according to the formula: Obtain the difference between the overall variance of the m-th column or m-th row of the two grayscale images among them, Is the overall variance of the mth column or mth row of the first grayscale image, Is the overall variance of the mth column or mth row of the second grayscale image;
  • Variance error threshold judgment subunit used to judge Whether it is less than the preset variance error threshold
  • Picture similarity determination subunit used if If it is less than the preset variance error threshold, it is determined that the specified picture is similar to the pre-stored picture.
  • grayscale refers to the color representing a grayscale color.
  • the color represents a grayscale color
  • the gray scale range is, for example, 0-255 (when the values of R, G, and B are all 0-255, of course it will also change with the change of the value range of R, G, and B).
  • the gray-scale processing method can be any method, such as the component method, the maximum method, the average method, and the weighted average method. Among them, since there are only 256 value ranges for the gray value, comparison of pictures on this basis can greatly reduce the amount of calculation. Then calculate the average value Am of the gray values of all pixels in the m-th column or the m-th row of the gray-scale picture, and calculate the average value B of the gray values of all the pixels in the gray-scale picture.
  • the process of calculating the average value Am of the gray values of all pixels in the m-th column or m-th row of the gray-scale picture includes: collecting all the pixels in the m-th column or m-th row of the gray-scale picture Add the gray values of all pixels in the mth column or mth row, and divide the sum of the gray values obtained by the summation by the mth column or The number of all pixels in the m rows is the average value Am of the gray values of all the pixels in the mth column or mth row of the grayscale image.
  • the process of calculating the average value B of the gray values of all pixels in the gray image includes: calculating the sum of the gray values of all pixels in the gray image, and dividing the sum of the gray values by According to the number of pixels, the average value B of the gray values of all pixels in the gray image is obtained.
  • the overall variance is used to measure the average of the gray values Am of the pixels in the m-th column or the m-th row of the gray-scale image and the average of the gray-scale values of all pixels in the gray-scale image. The difference between the value B.
  • the gray value of the mth column or row of the first grayscale image is the same or approximately the same as the gray value of the mth column or row of the second grayscale image (approximate judgment to save computing power , And because the overall variance of the two different pictures is generally not equal, the accuracy of the judgment is very high), on the contrary, the gray value of the mth column or mth row of the first grayscale image is the same as the second grayscale value.
  • the gray value of the mth column or mth row of the picture is different.
  • the return value is The maximum value in. If it is less than the preset variance error threshold, it is determined that the specified picture is similar to the pre-stored picture. Thus, approximate judgment is used (because all gray values of grayscale pictures converted from two different pictures are generally not equal, and all grayscale values of grayscale pictures converted from the same picture are generally equal), it achieves less consumption Under the premise of computing resources, it is determined whether the specified picture is similar to the pre-stored picture.
  • the picture similarity judgment unit 20 includes:
  • Counting the same pixel points subunit used to sequentially compare the corresponding pixels in the designated picture and the pre-stored picture, and count the number of the same pixels
  • the proportion threshold judgment subunit is used to judge whether the proportion of the same pixel is greater than a preset proportion threshold
  • the similarity determination subunit is configured to determine that the specified picture is similar to a pre-stored picture if the proportion of the same pixel is greater than a preset proportion threshold.
  • the use of a preset picture similarity judgment method is realized to judge whether the specified picture is similar to the pre-stored picture.
  • the device includes:
  • the all text acquisition unit is configured to, if the designated picture is similar to the pre-stored picture, acquire all the text corresponding to the upload terminal, and obtain the corresponding relationship of the pre-stored picture according to the correspondence between the preset picture and the pre-stored text and text Text text
  • the location information obtaining unit is configured to search for the text text corresponding to the pre-stored picture in all texts corresponding to the upload terminal, so as to obtain the location information of the text text corresponding to the pre-stored picture in all texts corresponding to the upload terminal ;
  • the location information sending unit is configured to send the location information to the text display terminal.
  • the location information is sent to the text display terminal.
  • the designated picture is similar to the pre-stored picture, only the position information of the text corresponding to the pre-stored picture in all the texts corresponding to the upload terminal is sent to the text display terminal, and the text display terminal uses the position information All the text corresponding to the upload terminal can be retrieved from local storage, and the text content corresponding to the designated picture can be obtained according to the location information.
  • a Kafka message queue can also be used to transmit location information, that is, the location information is put into a preset Kafka message queue for consumption by the text display terminal.
  • Kafka is a high-throughput distributed publish-subscribe message queue, which can process all action flow data in consumer-scale websites.
  • the designated picture sending unit 30 includes:
  • the preliminary recognition text acquisition sub-unit is used for if the designated picture is not similar to the pre-stored picture, use a preset preliminary text recognition technology to recognize a predetermined area in the picture text text picture to obtain a preliminary recognition text.
  • the area is smaller than the area of the designated picture;
  • the text text obtaining subunit is used to obtain the text text corresponding to the pre-stored picture according to the corresponding relationship between the preset picture and the pre-stored text text;
  • the preliminary recognition text judgment subunit is used to compare the preliminary recognition text and the text text corresponding to the pre-stored picture, so as to determine whether the text text corresponding to the pre-stored picture contains the preliminary recognition text;
  • the designated picture sending subunit is configured to send the designated picture to a designated text recognition terminal if the text corresponding to the prestored picture does not contain the preliminary recognized text, and the designated text recognition terminal is used to send the designated text
  • the text content in the picture is recognized as picture text text.
  • this embodiment adopts preliminary text recognition technology to further determine whether the designated picture is similar to the pre-stored picture.
  • the preliminary character recognition technology can be any character recognition technology, for example, OCR recognition technology.
  • the area of the predetermined area is smaller than the area of the designated picture, for example, an edge area of the designated picture.
  • this embodiment can further determine whether the designated picture is similar to the pre-stored picture under the premise of only consuming a small amount of computing power, and when the text corresponding to the pre-stored picture does not contain the preliminary recognized text (that is, the designated The picture is not similar to the pre-stored picture), the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture as text text of the picture.
  • the picture text text receiving unit 40 includes:
  • the picture text text receiving subunit is configured to receive the picture text text sent by the designated text recognition terminal;
  • the similarity value calculation subunit is used to adopt the formula:
  • the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture
  • similarity is the similarity value
  • A is the word frequency of the picture text text Vector
  • B is the word frequency vector of the pre-stored text
  • Ai is the number of occurrences of the i-th word of the image text
  • Bi is the number of occurrences of the i-th word of the pre-stored text
  • the similarity threshold judgment subunit is used to judge whether the similarity value is greater than a preset similarity threshold.
  • the similarity algorithm is calculated based on the cosine similarity of the text content of the two nodes to reflect the degree of similarity between the text content of the two nodes. When the value of similarity is closer to 1, the more similar is; the closer to 0, the less similar. Based on this, it is determined whether the similarity value is greater than a preset similarity threshold.
  • the similarity threshold is preferably 100%.
  • the device includes:
  • An all text acquisition unit configured to acquire all the text corresponding to the upload terminal if the similarity value is greater than a preset similarity threshold
  • a location information acquiring unit configured to acquire location information of the pre-stored text in all texts corresponding to the upload terminal
  • the location information sending unit is configured to send the location information to the text display terminal.
  • the location information is sent to the text display terminal. If the similarity value is greater than the preset similarity threshold, it indicates that the specified picture is similar to the pre-stored picture (the text of the picture is similar to the pre-stored text), so only the pre-stored text in all texts corresponding to the upload terminal
  • the location information can be sent to the text display terminal, and the text display terminal can retrieve all the text corresponding to the upload terminal from the local storage through the location information, and learn the text content corresponding to the specified picture according to the location information .
  • a Kafka message queue can also be used to transmit location information, that is, the location information is put into a preset Kafka message queue for consumption by the text display terminal.
  • Kafka is a high-throughput distributed publish-subscribe message queue, which can process all action flow data in consumer-scale websites.
  • the device includes:
  • a flow value obtaining unit configured to obtain a flow value sent to the text display terminal within a predetermined time
  • a flow threshold judging unit configured to judge whether the flow value is greater than a preset flow threshold
  • the similarity threshold obtaining unit is configured to obtain the similarity threshold by reducing the value of the preset standard threshold if the flow value is greater than the preset flow threshold.
  • this application adopts acquiring the flow value sent to the text display terminal within a predetermined time; determining whether the flow value is greater than a preset flow threshold; if the flow value is greater than the preset flow threshold, Then, the similarity threshold is obtained by reducing the value of the preset standard threshold, and the value is reduced on the basis of the standard threshold to reduce the number of times of sending text to the text display terminal, thereby reducing the pressure on the configuration platform.
  • the data transmission device based on the configuration platform of this application receives the text recognition application sent by the upload terminal; determines whether the specified picture is similar to the pre-stored picture; if the specified picture is not similar to the pre-stored picture, then the specified picture is sent to A designated text recognition terminal, the designated text recognition terminal is used to recognize the text content in the designated picture as picture text text; receive the picture text text sent by the designated text recognition terminal, calculate the picture text text and Prestore the similarity value of the text; if the similarity value is not greater than the preset similarity threshold, the text of the picture is sent to the text display terminal. This reduces the additional network overhead caused by repeated sending of text information corresponding to the repeated pictures, and realizes real-time display of recognized text.
  • an embodiment of the present invention also provides a computer device.
  • the computer device may be a server, and its internal structure may be as shown in the figure.
  • the computer equipment includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the computer designed processor is used to provide calculation and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system, a computer program, and a database.
  • the memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium.
  • the database of the computer equipment is used to store the data used in the data transmission method based on the configuration platform.
  • the network interface of the computer device is used to communicate with an external terminal through a network connection.
  • the computer program is executed by the processor to realize a data transmission method based on the configuration platform.
  • the processor executes the data transmission method based on the configuration platform, including the following steps: receiving a text recognition application sent by an upload terminal, where the text recognition application specifies a text display terminal, and the text recognition application carries a specified picture; Set the picture similarity judgment method to determine whether the designated picture is similar to the pre-stored picture; if the designated picture is not similar to the pre-stored picture, the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal uses Recognizing the text content in the designated picture as picture text text; receiving the picture text text sent by the designated text recognition terminal, and using a preset text similarity calculation method to calculate the picture text text and pre-stored The similarity value of the text, and determine whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture; if the similarity value is not greater than The preset similarity threshold is then sent to the text display terminal.
  • the step of judging whether the designated picture is similar to a pre-stored picture by using a preset picture similarity judgment method includes: performing gray-scale processing on the designated picture and the pre-stored picture respectively, Obtain the first grayscale picture and the second grayscale picture; calculate the average value Am of the grayscale values of all pixels in the mth column or mth row of the grayscale picture, and calculate all pixels in the grayscale picture The average value B of the gray value of the point; according to the formula: Calculate the overall variance of the m-th column or m-th row of the grayscale image Where N is the total number of columns or rows in the grayscale picture; according to the formula: Obtain the difference between the overall variance of the m-th column or m-th row of the two grayscale images among them, Is the overall variance of the mth column or mth row of the first grayscale image, Is the overall variance of the mth column or mth row of the second grayscale image; judge Is it less than the preset variance error
  • the method includes: if the designated picture is similar to a pre-stored picture, acquiring the upload terminal Corresponding to all texts, and obtain the text text corresponding to the pre-stored picture according to the corresponding relationship between the preset picture and the pre-stored text text; search for the text text corresponding to the pre-stored picture in all texts corresponding to the upload terminal, In this way, the position information of the text corresponding to the pre-stored picture in all the texts corresponding to the upload terminal is acquired; the position information is sent to the text display terminal.
  • the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture
  • the step of being a picture text text includes: if the designated picture is not similar to the pre-stored picture, using a preset preliminary text recognition technology to recognize a predetermined area in the picture text text picture to obtain a preliminary recognized text, wherein The area is smaller than the area of the designated picture; the text text corresponding to the pre-stored picture is obtained according to the correspondence between the preset picture and the pre-stored text and text; the preliminary recognized text is compared with the text text corresponding to the pre-stored picture, thereby Determine whether the text text corresponding to the pre-stored picture contains the preliminary recognized text; if the text text corresponding to the pre-stored picture does not contain the preliminary recognized text, the specified picture is sent to a specified text recognition terminal, and the specified The text recognition terminal is used for recognizing the text content in the designated picture as picture text.
  • the receiving the picture text text sent by the designated text recognition terminal and using a preset text similarity calculation method to calculate the similarity value between the picture text text and the pre-stored text, and determine Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the step of text obtained by performing text recognition on the pre-stored picture, including: receiving the picture sent by the designated text recognition terminal Text text; use formula: Calculate the similarity value between the picture text text and the pre-stored text, where the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture; where similarity is the similarity value, and A is the word frequency of the picture text text Vector, B is the word frequency vector of the pre-stored text, Ai is the number of occurrences of the i-th word of the image text, Bi is the number of occurrences of the i-th word of the pre-stored text; judge whether the similarity value Greater than the preset similarity threshold.
  • the receiving the picture text text sent by the designated text recognition terminal and using a preset text similarity calculation method to calculate the similarity value between the picture text text and the pre-stored text, and determine Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture, after the step, including: if the similarity value is greater than the preset similarity If the threshold value is higher than the threshold value, all the characters corresponding to the upload terminal are acquired; the position information of the pre-stored characters in all the characters corresponding to the upload terminal is obtained; the position information is sent to the character display terminal.
  • the step includes: acquiring and sending it to all The text displays the flow value of the terminal; it is determined whether the flow value is greater than the preset flow threshold; if the flow value is greater than the preset flow threshold, the similarity is obtained by reducing the value of the preset standard threshold Threshold.
  • the computer device of this application receives a text recognition application sent by an upload terminal; determines whether the designated picture is similar to a pre-stored picture; if the designated picture is not similar to a pre-stored picture, sends the designated picture to a designated text recognition terminal,
  • the designated text recognition terminal is used for recognizing the text content in the designated picture as picture text text; receiving the picture text text sent by the designated text recognition terminal, and calculating the similarity between the picture text text and the prestored text Value; if the similarity value is not greater than a preset similarity threshold, the text of the picture is sent to the text display terminal. This reduces the additional network overhead caused by the repeated sending of text information corresponding to the repeated pictures, and realizes real-time display of recognized text.
  • An embodiment of the present application also provides a computer-readable storage medium.
  • the computer-readable storage medium may be non-volatile or volatile, and has a computer program stored thereon, which is realized when the computer program is executed by a processor.
  • the data transmission method based on the configuration platform includes the following steps: receiving a text recognition application sent by an upload terminal, the text recognition application specifies a text display terminal, and the text recognition application carries a specified picture; using a preset picture similarity
  • the judging method is to judge whether the designated picture is similar to the pre-stored picture; if the designated picture is not similar to the pre-stored picture, the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to transfer the designated
  • the text content in the picture is recognized as picture text text; receiving the picture text text sent by the designated text recognition terminal, and using a preset text similarity calculation method to calculate the similarity value between the picture text text and the pre-stored text , And determine whether the similarity value is greater than a
  • the step of judging whether the designated picture is similar to a pre-stored picture by using a preset picture similarity judgment method includes: performing gray-scale processing on the designated picture and the pre-stored picture respectively, Obtain the first grayscale picture and the second grayscale picture; calculate the average value Am of the grayscale values of all pixels in the mth column or mth row of the grayscale picture, and calculate all pixels in the grayscale picture The average value B of the gray value of the point; according to the formula: Calculate the overall variance of the m-th column or m-th row of the grayscale image Where N is the total number of columns or rows in the grayscale picture; according to the formula: Obtain the difference between the overall variance of the m-th column or m-th row of the two grayscale images among them, Is the overall variance of the mth column or mth row of the first grayscale image, Is the overall variance of the mth column or mth row of the second grayscale image; judge Is it less than the preset variance error
  • the method includes: if the designated picture is similar to a pre-stored picture, acquiring the upload terminal Corresponding to all texts, and obtain the text text corresponding to the pre-stored picture according to the corresponding relationship between the preset picture and the pre-stored text text; search for the text text corresponding to the pre-stored picture in all texts corresponding to the upload terminal, In this way, the position information of the text corresponding to the pre-stored picture in all the texts corresponding to the upload terminal is acquired; the position information is sent to the text display terminal.
  • the designated picture is sent to a designated text recognition terminal, and the designated text recognition terminal is used to recognize the text content in the designated picture
  • the step of being a picture text text includes: if the designated picture is not similar to the pre-stored picture, using a preset preliminary text recognition technology to recognize a predetermined area in the picture text text picture to obtain a preliminary recognized text, wherein The area is smaller than the area of the designated picture; the text text corresponding to the pre-stored picture is obtained according to the correspondence between the preset picture and the pre-stored text and text; the preliminary recognized text is compared with the text text corresponding to the pre-stored picture, thereby Determine whether the text text corresponding to the pre-stored picture contains the preliminary recognized text; if the text text corresponding to the pre-stored picture does not contain the preliminary recognized text, the specified picture is sent to a specified text recognition terminal, and the specified The text recognition terminal is used for recognizing the text content in the designated picture as the text of
  • the receiving the picture text text sent by the designated text recognition terminal and using a preset text similarity calculation method to calculate the similarity value between the picture text text and the pre-stored text, and determine Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the step of text obtained by performing text recognition on the pre-stored picture, including: receiving the picture sent by the designated text recognition terminal Text text; use formula: Calculate the similarity value between the picture text text and the pre-stored text, where the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture; where similarity is the similarity value, and A is the word frequency of the picture text text Vector, B is the word frequency vector of the pre-stored text, Ai is the number of occurrences of the i-th word of the image text, Bi is the number of occurrences of the i-th word of the pre-stored text; judge whether the similarity value Greater than the preset similarity threshold.
  • the receiving the picture text text sent by the designated text recognition terminal and using a preset text similarity calculation method to calculate the similarity value between the picture text text and the pre-stored text, and determine Whether the similarity value is greater than a preset similarity threshold, wherein the pre-stored text refers to the text obtained by performing text recognition on the pre-stored picture, after the step, including: if the similarity value is greater than the preset similarity If the threshold value is higher than the threshold value, all the characters corresponding to the upload terminal are acquired; the position information of the pre-stored characters in all the characters corresponding to the upload terminal is obtained; the position information is sent to the character display terminal.
  • the step includes: acquiring and sending it to all The text displays the flow value of the terminal; it is determined whether the flow value is greater than the preset flow threshold; if the flow value is greater than the preset flow threshold, the similarity is obtained by reducing the value of the preset standard threshold Threshold.
  • the computer-readable storage medium of this application receives a text recognition application sent by an upload terminal; determines whether the designated picture is similar to a pre-stored picture; if the designated picture is not similar to the pre-stored picture, then the designated picture is sent to the designated text A recognition terminal, where the designated text recognition terminal is used to recognize the text content in the designated picture as picture text text; receive the picture text text sent by the designated text recognition terminal, and calculate the picture text text and pre-stored text If the similarity value is not greater than the preset similarity threshold, the text of the picture is sent to the text display terminal. This reduces the additional network overhead caused by the repeated sending of text information corresponding to the repeated pictures, and realizes real-time display of recognized text.
  • Non-volatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

Abstract

本申请涉及图像识别，揭示了一种基于配置平台的数据传输方法，包括：接收上传终端发送的文字识别申请；利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。从而减少了重复图片对应的文字信息的重复发送导致的额外网络开销。

Description

基于配置平台的数据传输方法、装置和计算机设备
本申请要求于2019年5月30日提交中国专利局，申请号为201910463920.X、发明名称为“基于配置平台的数据传输方法、装置和计算机设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及到计算机领域，特别是涉及到一种基于配置平台的数据传输方法、装置、计算机设备和存储介质。
背景技术
将图片中的文字识别出来，是当前生产生活中常使用的技术。传统的文字识别方法是上传终端将包括文字的图片直接上传给识别终端，再由识别终端识别图片以获得识别文字，识别终端再将识别文字发送给上传终端。发明人意识到，这种传统方法对于重复图片需要进行重复识别，浪费算力；并且不能实时将识别文字发给第三方终端，而若要将识别文字发给第三方终端，还需要进一步由上传终端将识别文字发给第三方终端，从而多了一步信息发送过程，浪费资源且无法做到实时且同时展示识别文字。
技术问题
本申请的主要目的为提供一种基于配置平台的数据传输方法、装置、计算机设备和存储介质，减少重复图片对应的文字信息的重复发送导致的额外网络开销。
技术解决方案
为了实现上述发明目的，本申请提出一种基于配置平台的数据传输方法，包括以下步骤：接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
本申请提供一种基于配置平台的数据传输装置，包括：文字识别申请接收单元，用于接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；图片相似度判断单元，用于利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；指定图片发送单元，用于若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；图片文字文本接收单元，用于接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；图片文字文本发送单元，用于若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
本申请提供一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，所述处理器执行所述计算机程序时实现上述任一项所述方法的步骤。
本申请提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述任一项所述的方法的步骤。
有益效果
本申请减少了重复图片对应的文字信息的重复发送导致的额外网络开销，并且实现了实时展示识别文字。
附图说明
图1为本申请一实施例的基于配置平台的数据传输方法的流程示意图；
图2为本申请一实施例的基于配置平台的数据传输装置的结构示意框图；
图3为本申请一实施例的计算机设备的结构示意框图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
本发明的最佳实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
参照图1，本申请实施例提供一种基于配置平台的数据传输方法，包括以下步骤：
S1、接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；
S2、利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；
S3、若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；
S4、接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；
S5、若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
将图片中的文字识别出来，是当前生产生活中常使用的技术。传统的文字识别方法是上传终端将包括文字的图片直接上传给识别终端，再由识别终端识别图片以获得识别文字，识别终端再将识别文字发送给上传终端。这种传统方法对于重复图片需要进行重复识别，浪费算力；并且不能实时将识别文字发给第三方终端，而若要将识别文字发给第三方终端，还需要进一步由上传终端将识别文字发给第三方终端，从而多了一步信息发送过程，浪费资源且无法做到实时且同时展示识别文字。本申请利用配置平台，对数据传递过程进行控制，利用两次相似度计算，以保证将指定图片为新的图片，再将从指定图片识别出的图片文字文本发送给指定的文字展示终端，从而实现了节省算力、节省流程、实时且同时展示识别文字的技术效果。
如上述步骤S1所述，接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片。本申请的执行主体为配置平台，所述配置平台用于在上传终端、识别终端、文字展示终端之间对数据传递过程进行控制，优选为基于大数据技术的配置平台，例如采用大数据引擎spark，采用大数据库HBase等，从而使配置平台能够存储并处理源于大量用户的大量数据。所述上传终端可为任意终端，例如为具有扫描功能的终端、具有拍照功能的终端等，优选为能够进行图像采集从而获得指定图片的终端。其中，所述指定图片为具有文字的图片。
如上述步骤S2所述，利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似。所述预设的图片相似度判断方法例如为：依次对比所述指定图片与预存图片中对应的像素点，若相同的像素点的数量在所有像素点数量中的占比大于预定阈值，则判定所述指定图片与预存图片相似，即表明所述指定图片预先存储在配置平台中，是重复图片，由于预存图片已经采用过文字识别技术识别出文字文本，因此没有必要再进行再一次识别；若相同的像素点的数量在所有像素点数量中的占比不大于预定阈值，则判定所述指定图片 与预存图片不相似，表明所述指定图片是新的图片，未采用文字识别技术进行识别，因此需要进行识别。
如上述步骤S3所述，若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。若所述指定图片与预存图片不相似，则所述指定图片有文字识别的需要，因此将所述指定图片发送给指定文本识别终端。本申请采用指定文本识别终端与配置平台相分离的设置，以将文字识别功能单独拆分出来由指定文本识别终端进行识别，可以根据具体的图片采用相应的文字识别算法的识别终端进行识别，实现分级管理、节省资源(简单图片由简单算法识别即可)、便于错误追踪。其中，所述将所述指定图片中的文字内容识别为图片文字文本的技术可采用现有的已成熟识别技术，例如OCR技术(OCR技术是光学字符识别的缩写(Optical Character Recognition)，是可用于图像信息转化为可以使用的计算机输入技术)，在此不再赘述。
如上述步骤S4所述，接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字。其中，利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值可为任意方法，例如：采用WMD算法(word mover’s distance)、基于余弦相似度、simhash算法、词频向量的算法等以所述图片文字文本与预存文字的相似度值。其中所述相似度阈值可为任意数值，例如为100％，进一步地，为了避免误判，可将相似度阈值设为小于100％，例如为大于等于80％。
如上述步骤S5所述，若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。相似度值用于判断所述图片文字文本与预存文字是否相同，即判断所述指定图片与预存图片是否相同。若所述相似度值不大于预设的相似度阈值，表明所述指定图片是新的图片，因此文字展示终端未保存由所述指定图片识别出的图片文字文本，据此将所述图片文字文本发送给所述文字展示终端，以实时且同时展示识别文字。进一步地，若所述相似度值大于预设的相似度阈值，表明所述指定图片是重复图片，因此文字展示终端保存由所述指定图片识别出的图片文字文本，因此无需发送图片文字文本，只需向所述文字展示终端告知所述图片文字文本存储位置，所述文字展示终端直接调用本地存储数据即可，从而节省了网络开销。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤S2,包括：
S201、分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；
S202、计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；
S203、根据公式：
Figure PCTCN2020088049-appb-000001
计算所述灰度图片的第m列或者第m行的总体方差
Figure PCTCN2020088049-appb-000002
其中N为所述灰度图片中的列或者行的总数量；
S204、根据公式:
Figure PCTCN2020088049-appb-000003
获得两张所述灰度图片的第m列或者第m行的总体方差之差
Figure PCTCN2020088049-appb-000004
其中，
Figure PCTCN2020088049-appb-000005
为第一张灰度图片的第m列或者第m行的总体方差，
Figure PCTCN2020088049-appb-000006
为第二张灰度图片的第m列或者第m行的总体方差；
S205、判断
Figure PCTCN2020088049-appb-000007
是否小于预设的方差误差阈值；
S206、若
Figure PCTCN2020088049-appb-000008
小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
如上所述，实现了利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似。其中，灰度化指将彩色表示一种灰度颜色，例如在在RGB模型中，如果R＝G＝B时，则彩色表示一种灰度颜色，其中R＝G＝B的值叫灰度值，因此，灰度图像每个像素只需一个字节存放灰度值(又称强度值、亮度值)，从而减少存储量。灰度范围例如为0-255(当R,G,B的取值均为0-255时，当然也会随R,G,B的取值范围的变化而变化)。采用灰度化处理的方法可以为任意方法，例如分量法、最大值法、平均值法、加权平均法等。其中，由于灰度值的取值范围只有256种，在此基础上进行图片对比能够大大减轻计算量。再计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B。其中，计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am的过程包括：采集所述灰度图片的第m列或者第m行的所有像素点的灰度值，对所述第m列或者第m行的所有像素点的灰度值进行加和处理，将进行过加和处理得到的灰度值之和除以所述第m列或者第m行的所有像素点的数量，得到所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am。计算所述灰度图片中所有像素点的灰度值的平均值B的过程包括：计算所述灰度图片中所有像素点的灰度值之和，再以所述灰度值之和除以所述像素点的数量，得到所述灰度图片中所有像素点的灰度值的平均值B。根据公式：
Figure PCTCN2020088049-appb-000009
计算所述灰度图片的第m列或者第m行的总体方差
Figure PCTCN2020088049-appb-000010
其中N为所述灰度图片中的列或者行的总数量。在本申请中，采用总体方差来衡量所述灰度图片的第m列或者第m行的像素点的灰度值的平均值Am与所述灰度图片中所有像素点的灰度值的平均值B之间的差异。
根据公式:
Figure PCTCN2020088049-appb-000011
获得两张所述灰度图片的第m列或者第m行的总体方差之差
Figure PCTCN2020088049-appb-000012
其中，
Figure PCTCN2020088049-appb-000013
为第一张灰度图片的第m列或者第m行的总体方差，
Figure PCTCN2020088049-appb-000014
为第二张灰度图片的第m列或者第m行的总体方差。总体方差之差
Figure PCTCN2020088049-appb-000015
反应了两张灰度图片的第m列或者第m行的灰度值的差异。当
Figure PCTCN2020088049-appb-000016
较小时，例如为0时，表明
Figure PCTCN2020088049-appb-000017
等于或者近似等于
Figure PCTCN2020088049-appb-000018
可视为第一张灰度图片第m列或者第m行的灰度值与第二张灰度图片第m列或者第m行的灰度值相同或者近似相同(近似判断，以节省算力，并且由于不同的两张图片的总体方差一般不相等，因此该判断的准确性很高)，反之认为第一张灰度图片第m列或者第m行的灰度值与第二张灰度图片第m列或者第m行的灰度值不相同。
判断
Figure PCTCN2020088049-appb-000019
是否小于预设的方差误差阈值。其中
Figure PCTCN2020088049-appb-000020
的返回值即为
Figure PCTCN2020088049-appb-000021
中的最大值。若
Figure PCTCN2020088049-appb-000022
小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。从而利用了近似判断(由于两张不同图片转化为的灰度图片的所有灰度值一般不相等，而相同图片转化为的灰度图片的所有灰度值一般相等)，实现了在消耗较少计算资源的前提下，判断所述指定图片与预存图片是否相似。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤S2,包括：
S201、依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；
S202、根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素 点的数量，计算出所述相同像素点占比；
S203、判断所述相同像素点占比是否大于预设的占比阈值；
S204、若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
如上所述，实现了利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似。为了精准判断所述指定图片与预存图片是否相似，本实施方式采用逐次比对像素点的方式进行判断。若两张图片是相同的，那么相同像素点的数量应当占绝大多数，即所述相同像素点占比趋近于1。据此，根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比，若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤S2之后,包括：
S21、若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
S22、在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；
S23、将所述位置信息发送给所述文字展示终端。
如上所述，实现了将所述位置信息发送给所述文字展示终端。若所述指定图片与预存图片相似，则只需将预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息发送给文字展示终端即可，文字展示终端通过所述位置信息即可从本地存储中调取所述上传终端对应的所有文字，并根据所述位置信息获知所述指定图片对应的文字内容。其中，为了适应大数据处理的需要，进一步地，还可以采用Kafka消息队列进行位置信息的传输，即将所述位置信息放入预设的Kafka消息队列，以待文字展示终端进行消费。其中，Kafka是一种高吞吐量的分布式发布订阅消息队列，它可以处理消费者规模的网站中的所有动作流数据。
在一个实施方式中，所述若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本的步骤S3,包括：
S301、若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；
S302、根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
S303、对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；
S304、若所述预存图片对应的文字文本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
如上所述，实现了在保证少量计算的前提下，进一步确定图片是否相同。本实施方式中，为了防止前述图片判断可能出现的误判，本实施方式采用初步文字识别技术进一步确定指定图片与预存图片是否相似。其中，所述初步文字识别技术可以为任意文字识别技术，例如为OCR识别技术。所述预定区域的面积小于所述指定图片的面积，例如为所述指定图片的边缘区域。由于扫描图片等需要移动摄像头，例如向右平移摄像头，将导致左侧边缘区域与预存图片对应的文字文本不同，从而仅识别左侧边缘区域即可判断所述指定图片与预存图片是否相似。因此本实施方式在仅消耗少量算力的前提下，即可进一步确定指定图片与预存图片是否相似，并且在所述预存图片对应的文字文本不包含所述初步识别文字之 时(即所述指定图片与预存图片不相似)，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
在一个实施方式中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤S4,包括：
S401、接收所述指定文本识别终端发送的所述图片文字文本；
S402、采用公式：
Figure PCTCN2020088049-appb-000023
计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；
S403、判断所述相似度值是否大于预设的相似度阈值。
如上所述，实现了计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值。所述词频向量是以文字内容中的各词出现的次数(频率)作为向量的维度数值，所构成的多维向量。即A＝(A1,A2,…,An),其中An为最后一个词(共有n个词)的词频。所述相似度算法是根据两个节点的文字内容的余弦相似度进行计算得到，以反应两个节点的文字内容间的相似程度。当similarity的值越接近于1，表明越相似；越接近于0，表明越不相似。据此判断所述相似度值是否大于预设的相似度阈值。其中所述相似度阈值优选为100％。
在一个实施方式中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤S4之后,包括：
S41、若所述相似度值大于预设的相似度阈值，则获取所述上传终端对应的所有文字；
S42、获取所述预存文字在所述上传终端对应的所有文字中的位置信息；
S43、将所述位置信息发送给所述文字展示终端。
如上所述，实现了将所述位置信息发送给所述文字展示终端。若所述相似度值大于预设的相似度阈值，表明所述指定图片与预存图片相似(图片文字文本与预存文字相似)，因此只需将预存文字在所述上传终端对应的所有文字中的位置信息发送给文字展示终端即可，文字展示终端通过所述位置信息即可从本地存储中调取所述上传终端对应的所有文字，并根据所述位置信息获知所述指定图片对应的文字内容。其中，为了适应大数据处理的需要，进一步地，还可以采用Kafka消息队列进行位置信息的传输，即将所述位置信息放入预设的Kafka消息队列，以待文字展示终端进行消费。其中，Kafka是一种高吞吐量的分布式发布订阅消息队列，它可以处理消费者规模的网站中的所有动作流数据。
在一个实施方式中，所述若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端的步骤S5之前，包括：
S41、获取预定时间内发送至所述文字展示终端的流量值；
S42、判断所述流量值是否大于预设的流量阈值；
S43、若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
如上所述，实现了设置相似度阈值。为了进一步减轻配置平台压力，本申请采用获取预定时间内发送至所述文字展示终端的流量值；判断所述流量值是否大于预设的流量阈值；若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值的方式，在标准阈值的基础上减小数值，以减少向文字展示终端发送文字文本的次数，进而减轻配置平台压力。
本申请的基于配置平台的数据传输方法，接收上传终端发送的文字识别申请；判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，计算所述图片文字文本与预存文字的相似度值；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。从而减少了重复图片对应的文字信息的重复发送导致的额外网络开销，并且实现了实时展示识别文字。
参照图2，本申请实施例提供一种基于配置平台的数据传输装置，包括：
文字识别申请接收单元10，用于接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；
图片相似度判断单元20，用于利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；
指定图片发送单元30，用于若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；
图片文字文本接收单元40，用于接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；
图片文字文本发送单元50，用于若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
将图片中的文字识别出来，是当前生产生活中常使用的技术。传统的文字识别方法是上传终端将包括文字的图片直接上传给识别终端，再由识别终端识别图片以获得识别文字，识别终端再将识别文字发送给上传终端。这种传统方法对于重复图片需要进行重复识别，浪费算力；并且不能实时将识别文字发给第三方终端，而若要将识别文字发给第三方终端，还需要进一步由上传终端将识别文字发给第三方终端，从而多了一步信息发送过程，浪费资源且无法做到实时且同时展示识别文字。本申请利用配置平台，对数据传递过程进行控制，利用两次相似度计算，以保证将指定图片为新的图片，再将从指定图片识别出的图片文字文本发送给指定的文字展示终端，从而实现了节省算力、节省流程、实时且同时展示识别文字的技术效果。
如上述单元10所述，接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片。本申请的执行主体为配置平台，所述配置平台用于在上传终端、识别终端、文字展示终端之间对数据传递过程进行控制，优选为基于大数据技术的配置平台，例如采用大数据引擎spark，采用大数据库HBase等，从而使配置平台能够存储并处理源于大量用户的大量数据。所述上传终端可为任意终端，例如为具有扫描功能的终端、具有拍照功能的终端等，优选为能够进行图像采集从而获得指定图片的终端。其中，所述指定图片为具有文字的图片。
如上述单元20所述，利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似。所述预设的图片相似度判断方法例如为：依次对比所述指定图片与预存图片中对应的像素点，若相同的像素点的数量在所有像素点数量中的占比大于预定阈值，则判定所述指定图片与预存图片相似，即表明所述指定图片预先存储在配置平台中，是重复图片， 由于预存图片已经采用过文字识别技术识别出文字文本，因此没有必要再进行再一次识别；若相同的像素点的数量在所有像素点数量中的占比不大于预定阈值，则判定所述指定图片与预存图片不相似，表明所述指定图片是新的图片，未采用文字识别技术进行识别，因此需要进行识别。
如上述单元30所述，若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。若所述指定图片与预存图片不相似，则所述指定图片有文字识别的需要，因此将所述指定图片发送给指定文本识别终端。本申请采用指定文本识别终端与配置平台相分离的设置，以将文字识别功能单独拆分出来由指定文本识别终端进行识别，可以根据具体的图片采用相应的文字识别算法的识别终端进行识别，实现分级管理、节省资源(简单图片由简单算法识别即可)、便于错误追踪。其中，所述将所述指定图片中的文字内容识别为图片文字文本的技术可采用现有的已成熟识别技术，例如OCR技术(OCR技术是光学字符识别的缩写(Optical Character Recognition)，是可用于图像信息转化为可以使用的计算机输入技术)，在此不再赘述。
如上述单元40所述，接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字。其中，利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值可为任意方法，例如：采用WMD算法(word mover’s distance)、基于余弦相似度、simhash算法、词频向量的算法等以所述图片文字文本与预存文字的相似度值。其中所述相似度阈值可为任意数值，例如为100％，进一步地，为了避免误判，可将相似度阈值设为小于100％，例如为大于等于80％。
如上述单元50所述，若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。相似度值用于判断所述图片文字文本与预存文字是否相同，即判断所述指定图片与预存图片是否相同。若所述相似度值不大于预设的相似度阈值，表明所述指定图片是新的图片，因此文字展示终端未保存由所述指定图片识别出的图片文字文本，据此将所述图片文字文本发送给所述文字展示终端，以实时且同时展示识别文字。进一步地，若所述相似度值大于预设的相似度阈值，表明所述指定图片是重复图片，因此文字展示终端保存由所述指定图片识别出的图片文字文本，因此无需发送图片文字文本，只需向所述文字展示终端告知所述图片文字文本存储位置，所述文字展示终端直接调用本地存储数据即可，从而节省了网络开销。
在一个实施方式中，所述图片相似度判断单元20,包括：
灰度化处理子单元，用于分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；
灰度值的平均值计算子单元，用于计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；
总体方差获取子单元，用于根据公式：
Figure PCTCN2020088049-appb-000024
计算所述灰度图片的第m列或者第m行的总体方差
Figure PCTCN2020088049-appb-000025
其中N为所述灰度图片中的列或者行的总数量；
总体方差之差获取子单元，用于根据公式:
Figure PCTCN2020088049-appb-000026
获得两张所述灰度图片的第m列或者第m行的总体方差之差
Figure PCTCN2020088049-appb-000027
其中，
Figure PCTCN2020088049-appb-000028
为第一张灰度图片的第m列或者第m行的总体方差，
Figure PCTCN2020088049-appb-000029
为第二张灰度图片的第m列或者第m行的总体方差；
方差误差阈值判断子单元，用于判断
Figure PCTCN2020088049-appb-000030
是否小于预设的方差误差阈值；
图片相似判定子单元，用于若
Figure PCTCN2020088049-appb-000031
小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
如上所述，实现了利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似。其中，灰度化指将彩色表示一种灰度颜色，例如在在RGB模型中，如果R＝G＝B时，则彩色表示一种灰度颜色，其中R＝G＝B的值叫灰度值，因此，灰度图像每个像素只需一个字节存放灰度值(又称强度值、亮度值)，从而减少存储量。灰度范围例如为0-255(当R,G,B的取值均为0-255时，当然也会随R,G,B的取值范围的变化而变化)。采用灰度化处理的方法可以为任意方法，例如分量法、最大值法、平均值法、加权平均法等。其中，由于灰度值的取值范围只有256种，在此基础上进行图片对比能够大大减轻计算量。再计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B。其中，计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am的过程包括：采集所述灰度图片的第m列或者第m行的所有像素点的灰度值，对所述第m列或者第m行的所有像素点的灰度值进行加和处理，将进行过加和处理得到的灰度值之和除以所述第m列或者第m行的所有像素点的数量，得到所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am。计算所述灰度图片中所有像素点的灰度值的平均值B的过程包括：计算所述灰度图片中所有像素点的灰度值之和，再以所述灰度值之和除以所述像素点的数量，得到所述灰度图片中所有像素点的灰度值的平均值B。根据公式：
Figure PCTCN2020088049-appb-000032
计算所述灰度图片的第m列或者第m行的总体方差
Figure PCTCN2020088049-appb-000033
其中N为所述灰度图片中的列或者行的总数量。在本申请中，采用总体方差来衡量所述灰度图片的第m列或者第m行的像素点的灰度值的平均值Am与所述灰度图片中所有像素点的灰度值的平均值B之间的差异。
根据公式:
Figure PCTCN2020088049-appb-000034
获得两张所述灰度图片的第m列或者第m行的总体方差之差
Figure PCTCN2020088049-appb-000035
其中，
Figure PCTCN2020088049-appb-000036
为第一张灰度图片的第m列或者第m行的总体方差，
Figure PCTCN2020088049-appb-000037
为第二张灰度图片的第m列或者第m行的总体方差。总体方差之差
Figure PCTCN2020088049-appb-000038
反应了两张灰度图片的第m列或者第m行的灰度值的差异。当
Figure PCTCN2020088049-appb-000039
较小时，例如为0时，表明
Figure PCTCN2020088049-appb-000040
等于或者近似等于
Figure PCTCN2020088049-appb-000041
可视为第一张灰度图片第m列或者第m行的灰度值与第二张灰度图片第m列或者第m行的灰度值相同或者近似相同(近似判断，以节省算力，并且由于不同的两张图片的总体方差一般不相等，因此该判断的准确性很高)，反之认为第一张灰度图片第m列或者第m行的灰度值与第二张灰度图片第m列或者第m行的灰度值不相同。
判断
Figure PCTCN2020088049-appb-000042
是否小于预设的方差误差阈值。其中
Figure PCTCN2020088049-appb-000043
的返回值即为
Figure PCTCN2020088049-appb-000044
中的最大值。若
Figure PCTCN2020088049-appb-000045
小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。从而利用了近似判断(由于两张不同图片转化为的灰度图片的所有灰度值一般不相等，而相同图片转化为的灰度图片的所有灰度值一般相等)，实现了在消耗较少计算资源的前提下，判断所述指定图片与预存图片是否相似。
在一个实施方式中，所述图片相似度判断单元20,包括：
统计相同像素点子单元，用于依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；
相同像素点占比计算子单元，用于根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比；
占比阈值判断子单元，用于判断所述相同像素点占比是否大于预设的占比阈值；
相似判定子单元，用于若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
如上所述，实现了利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似。为了精准判断所述指定图片与预存图片是否相似，本实施方式采用逐次比对像素点的方式进行判断。若两张图片是相同的，那么相同像素点的数量应当占绝大多数，即所述相同像素点占比趋近于1。据此，根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比，若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
在一个实施方式中，所述装置,包括：
所有文字获取单元，用于若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
位置信息获取单元，用于在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；
位置信息发送单元，用于将所述位置信息发送给所述文字展示终端。
如上所述，实现了将所述位置信息发送给所述文字展示终端。若所述指定图片与预存图片相似，则只需将预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息发送给文字展示终端即可，文字展示终端通过所述位置信息即可从本地存储中调取所述上传终端对应的所有文字，并根据所述位置信息获知所述指定图片对应的文字内容。其中，为了适应大数据处理的需要，进一步地，还可以采用Kafka消息队列进行位置信息的传输，即将所述位置信息放入预设的Kafka消息队列，以待文字展示终端进行消费。其中，Kafka是一种高吞吐量的分布式发布订阅消息队列，它可以处理消费者规模的网站中的所有动作流数据。
在一个实施方式中，所述指定图片发送单元30,包括：
初步识别文字获取子单元，用于若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；
文字文本获取子单元，用于根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
初步识别文字判断子单元，用于对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；
指定图片发送子单元，用于若所述预存图片对应的文字文本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
如上所述，实现了在保证少量计算的前提下，进一步确定图片是否相同。本实施方式中，为了防止前述图片判断可能出现的误判，本实施方式采用初步文字识别技术进一步确定指定图片与预存图片是否相似。其中，所述初步文字识别技术可以为任意文字识别技术，例如为OCR识别技术。所述预定区域的面积小于所述指定图片的面积，例如为所述指定图片的边缘区域。由于扫描图片等需要移动摄像头，例如向右平移摄像头，将导致左侧边缘 区域与预存图片对应的文字文本不同，从而仅识别左侧边缘区域即可判断所述指定图片与预存图片是否相似。因此本实施方式在仅消耗少量算力的前提下，即可进一步确定指定图片与预存图片是否相似，并且在所述预存图片对应的文字文本不包含所述初步识别文字之时(即所述指定图片与预存图片不相似)，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
在一个实施方式中，所述图片文字文本接收单元40,包括：
图片文字文本接收子单元，用于接收所述指定文本识别终端发送的所述图片文字文本；
相似度值计算子单元，用于采用公式：
Figure PCTCN2020088049-appb-000046
计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；
相似度阈值判断子单元，用于判断所述相似度值是否大于预设的相似度阈值。
如上所述，实现了计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值。所述词频向量是以文字内容中的各词出现的次数(频率)作为向量的维度数值，所构成的多维向量。即A＝(A1,A2,…,An),其中An为最后一个词(共有n个词)的词频。所述相似度算法是根据两个节点的文字内容的余弦相似度进行计算得到，以反应两个节点的文字内容间的相似程度。当similarity的值越接近于1，表明越相似；越接近于0，表明越不相似。据此判断所述相似度值是否大于预设的相似度阈值。其中所述相似度阈值优选为100％。
在一个实施方式中，所述装置,包括：
所有文字获取单元，用于若所述相似度值大于预设的相似度阈值，则获取所述上传终端对应的所有文字；
位置信息获取单元，用于获取所述预存文字在所述上传终端对应的所有文字中的位置信息；
位置信息发送单元，用于将所述位置信息发送给所述文字展示终端。
如上所述，实现了将所述位置信息发送给所述文字展示终端。若所述相似度值大于预设的相似度阈值，表明所述指定图片与预存图片相似(图片文字文本与预存文字相似)，因此只需将预存文字在所述上传终端对应的所有文字中的位置信息发送给文字展示终端即可，文字展示终端通过所述位置信息即可从本地存储中调取所述上传终端对应的所有文字，并根据所述位置信息获知所述指定图片对应的文字内容。其中，为了适应大数据处理的需要，进一步地，还可以采用Kafka消息队列进行位置信息的传输，即将所述位置信息放入预设的Kafka消息队列，以待文字展示终端进行消费。其中，Kafka是一种高吞吐量的分布式发布订阅消息队列，它可以处理消费者规模的网站中的所有动作流数据。
在一个实施方式中，所述装置，包括：
流量值获取单元，用于获取预定时间内发送至所述文字展示终端的流量值；
流量阈值判断单元，用于判断所述流量值是否大于预设的流量阈值；
相似度阈值获取单元，用于若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
如上所述，实现了设置相似度阈值。为了进一步减轻配置平台压力，本申请采用获取预定时间内发送至所述文字展示终端的流量值；判断所述流量值是否大于预设的流量阈值； 若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值的方式，在标准阈值的基础上减小数值，以减少向文字展示终端发送文字文本的次数，进而减轻配置平台压力。
本申请的基于配置平台的数据传输装置，接收上传终端发送的文字识别申请；判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，计算所述图片文字文本与预存文字的相似度值；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。从而减少了重复图片对应的文字信息的重复发送导致的额外网络开销，并且实现了实时展示识别文字。
参照图3，本发明实施例中还提供一种计算机设备，该计算机设备可以是服务器，其内部结构可以如图所示。该计算机设备包括通过系统总线连接的处理器、存储器、网络接口和数据库。其中，该计算机设计的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统、计算机程序和数据库。该内存器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的数据库用于存储基于配置平台的数据传输方法所用数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现一种基于配置平台的数据传输方法。
上述处理器执行上述基于配置平台的数据传输方法，包括以下步骤：接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；根据公式：
Figure PCTCN2020088049-appb-000047
计算所述灰度图片的第m列或者第m行的总体方差
Figure PCTCN2020088049-appb-000048
其中N为所述灰度图片中的列或者行的总数量；根据公式:
Figure PCTCN2020088049-appb-000049
获得两张所述灰度图片的第m列或者第m行的总体方差之差
Figure PCTCN2020088049-appb-000050
其中，
Figure PCTCN2020088049-appb-000051
为第一张灰度图片的第m列或者第m行的总体方差，
Figure PCTCN2020088049-appb-000052
为第二张灰度图片的第m列或者第m行的总体方差；判断
Figure PCTCN2020088049-appb-000053
是否小于预设的方差误差阈值；若
Figure PCTCN2020088049-appb-000054
小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比；判断所述相同像素点占比是否大于预设的占比阈值；若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片 相似。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤之后,包括：若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；将所述位置信息发送给所述文字展示终端。
在一个实施方式中，所述若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本的步骤,包括：若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；若所述预存图片对应的文字文本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
在一个实施方式中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤,包括：接收所述指定文本识别终端发送的所述图片文字文本；采用公式：
Figure PCTCN2020088049-appb-000055
计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；判断所述相似度值是否大于预设的相似度阈值。
在一个实施方式中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤之后,包括：若所述相似度值大于预设的相似度阈值，则获取所述上传终端对应的所有文字；获取所述预存文字在所述上传终端对应的所有文字中的位置信息；将所述位置信息发送给所述文字展示终端。
在一个实施方式中，所述若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端的步骤之前，包括：获取预定时间内发送至所述文字展示终端的流量值；判断所述流量值是否大于预设的流量阈值；若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
本领域技术人员可以理解，图中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定。
本申请的计算机设备，接收上传终端发送的文字识别申请；判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，计算所述图片文字文本与预存文字的相似度值；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所 述文字展示终端。从而减少了重复图片对应的文字信息的重复发送导致的额外网络开销，并且实现了实时展示识别文字。
本申请一实施例还提供一种计算机可读存储介质，所述计算机可读存储介质可以是非易失性，也可以是易失性，其上存储有计算机程序，计算机程序被处理器执行时实现基于配置平台的数据传输方法，包括以下步骤：接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；根据公式：
Figure PCTCN2020088049-appb-000056
计算所述灰度图片的第m列或者第m行的总体方差
Figure PCTCN2020088049-appb-000057
其中N为所述灰度图片中的列或者行的总数量；根据公式:
Figure PCTCN2020088049-appb-000058
获得两张所述灰度图片的第m列或者第m行的总体方差之差
Figure PCTCN2020088049-appb-000059
其中，
Figure PCTCN2020088049-appb-000060
为第一张灰度图片的第m列或者第m行的总体方差，
Figure PCTCN2020088049-appb-000061
为第二张灰度图片的第m列或者第m行的总体方差；判断
Figure PCTCN2020088049-appb-000062
是否小于预设的方差误差阈值；若
Figure PCTCN2020088049-appb-000063
小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比；判断所述相同像素点占比是否大于预设的占比阈值；若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
在一个实施方式中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤之后,包括：若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；将所述位置信息发送给所述文字展示终端。
在一个实施方式中，所述若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本的步骤,包括：若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；若所述预存图片对应的文字文 本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
在一个实施方式中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤,包括：接收所述指定文本识别终端发送的所述图片文字文本；采用公式：
Figure PCTCN2020088049-appb-000064
计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；判断所述相似度值是否大于预设的相似度阈值。
在一个实施方式中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤之后,包括：若所述相似度值大于预设的相似度阈值，则获取所述上传终端对应的所有文字；获取所述预存文字在所述上传终端对应的所有文字中的位置信息；将所述位置信息发送给所述文字展示终端。
在一个实施方式中，所述若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端的步骤之前，包括：获取预定时间内发送至所述文字展示终端的流量值；判断所述流量值是否大于预设的流量阈值；若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
本申请的计算机可读存储介质，接收上传终端发送的文字识别申请；判断所述指定图片与预存图片是否相似；若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；接收所述指定文本识别终端发送的所述图片文字文本，计算所述图片文字文本与预存文字的相似度值；若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。从而减少了重复图片对应的文字信息的重复发送导致的额外网络开销，并且实现了实时展示识别文字。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一非易失性计算机可读取存储介质中，该计算机程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的和实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可以包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双速据率SDRAM(SSRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、装置、物品或者方法不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、装置、物品或者方法所固 有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、装置、物品或者方法中还存在另外的相同要素。
以上所述仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (21)

  1. 一种基于配置平台的数据传输方法，其中，包括：
    接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；
    利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；
    若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；
    接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；
    若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
  2. 根据权利要求1所述的基于配置平台的数据传输方法，其中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：
    分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；
    计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；
    根据公式：
    Figure PCTCN2020088049-appb-100001
    计算所述灰度图片的第m列或者第m行的总体方差
    Figure PCTCN2020088049-appb-100002
    其中N为所述灰度图片中的列或者行的总数量；
    根据公式:
    Figure PCTCN2020088049-appb-100003
    获得两张所述灰度图片的第m列或者第m行的总体方差之差
    Figure PCTCN2020088049-appb-100004
    其中，
    Figure PCTCN2020088049-appb-100005
    为第一张灰度图片的第m列或者第m行的总体方差，
    Figure PCTCN2020088049-appb-100006
    为第二张灰度图片的第m列或者第m行的总体方差；
    判断
    Figure PCTCN2020088049-appb-100007
    是否小于预设的方差误差阈值；
    若
    Figure PCTCN2020088049-appb-100008
    小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
  3. 根据权利要求1所述的基于配置平台的数据传输方法，其中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：
    依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；
    根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比；
    判断所述相同像素点占比是否大于预设的占比阈值；
    若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
  4. 根据权利要求1所述的基于配置平台的数据传输方法，其中，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤之后,包括：
    若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
    在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；
    将所述位置信息发送给所述文字展示终端。
  5. 根据权利要求1所述的基于配置平台的数据传输方法，其中，所述若所述指定图片 与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本的步骤,包括：
    若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；
    根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
    对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；
    若所述预存图片对应的文字文本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
  6. 根据权利要求1所述的基于配置平台的数据传输方法，其中，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤,包括：
    接收所述指定文本识别终端发送的所述图片文字文本；
    采用公式：
    Figure PCTCN2020088049-appb-100009
    计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；
    判断所述相似度值是否大于预设的相似度阈值。
  7. 根据权利要求1所述的基于配置平台的数据传输方法，其中，所述若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端的步骤之前，包括：
    获取预定时间内发送至所述文字展示终端的流量值；
    判断所述流量值是否大于预设的流量阈值；
    若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
  8. 一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，其中，所述处理器执行所述计算机程序时实现如下步骤：
    接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；
    利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；
    若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；
    接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；
    若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
  9. 根据权利要求8所述的计算机设备，其特征在于，所述利用预设的图片相似度判断 方法，判断所述指定图片与预存图片是否相似的步骤,包括：
    分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；
    计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；
    根据公式：
    Figure PCTCN2020088049-appb-100010
    计算所述灰度图片的第m列或者第m行的总体方差
    Figure PCTCN2020088049-appb-100011
    其中N为所述灰度图片中的列或者行的总数量；
    根据公式:
    Figure PCTCN2020088049-appb-100012
    获得两张所述灰度图片的第m列或者第m行的总体方差之差
    Figure PCTCN2020088049-appb-100013
    其中，
    Figure PCTCN2020088049-appb-100014
    为第一张灰度图片的第m列或者第m行的总体方差，
    Figure PCTCN2020088049-appb-100015
    为第二张灰度图片的第m列或者第m行的总体方差；
    判断
    Figure PCTCN2020088049-appb-100016
    是否小于预设的方差误差阈值；
    若
    Figure PCTCN2020088049-appb-100017
    小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
  10. 根据权利要求8所述的计算机设备，其特征在于，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：
    依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；
    根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比；
    判断所述相同像素点占比是否大于预设的占比阈值；
    若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
  11. 根据权利要求8所述的计算机设备，其特征在于，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤之后,包括：
    若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
    在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；
    将所述位置信息发送给所述文字展示终端。
  12. 根据权利要求8所述的计算机设备，其特征在于，所述若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本的步骤,包括：
    若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；
    根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
    对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；
    若所述预存图片对应的文字文本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
  13. 根据权利要求8所述的计算机设备，其特征在于，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文 字指通过对所述预存图片进行文字识别而获得的文字的步骤,包括：
    接收所述指定文本识别终端发送的所述图片文字文本；
    采用公式：
    Figure PCTCN2020088049-appb-100018
    计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；
    判断所述相似度值是否大于预设的相似度阈值。
  14. 根据权利要求8所述的计算机设备，其特征在于，所述若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端的步骤之前，包括：
    获取预定时间内发送至所述文字展示终端的流量值；
    判断所述流量值是否大于预设的流量阈值；
    若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
  15. 一种计算机可读存储介质，其上存储有计算机程序，其中，所述计算机程序被处理器执行时实现如下步骤：
    接收上传终端发送的文字识别申请，所述文字识别申请指定了文字展示终端，并且所述文字识别申请携带有指定图片；
    利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似；
    若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本；
    接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；
    若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端。
  16. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：
    分别对所述指定图片与所述预存图片进行灰度化处理，得到第一灰度图片和第二灰度图片；
    计算所述灰度图片的第m列或者第m行的所有像素点的灰度值的平均值Am，以及计算所述灰度图片中所有像素点的灰度值的平均值B；
    根据公式：
    Figure PCTCN2020088049-appb-100019
    计算所述灰度图片的第m列或者第m行的总体方差
    Figure PCTCN2020088049-appb-100020
    其中N为所述灰度图片中的列或者行的总数量；
    根据公式:
    Figure PCTCN2020088049-appb-100021
    获得两张所述灰度图片的第m列或者第m行的总体方差之差
    Figure PCTCN2020088049-appb-100022
    其中，
    Figure PCTCN2020088049-appb-100023
    为第一张灰度图片的第m列或者第m行的总体方差，
    Figure PCTCN2020088049-appb-100024
    为第二张灰度图片的第m列或者第m行的总体方差；
    判断
    Figure PCTCN2020088049-appb-100025
    是否小于预设的方差误差阈值；
    若
    Figure PCTCN2020088049-appb-100026
    小于预设的方差误差阈值，则判定所述指定图片与预存图片相似。
  17. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤,包括：
    依次对比所述指定图片与预存图片中对应的像素点，并统计相同像素点的数量；
    根据公式：相同像素点占比＝所述相同像素点的数量/所述指定图片中所有像素点的数量，计算出所述相同像素点占比；
    判断所述相同像素点占比是否大于预设的占比阈值；
    若所述相同像素点占比大于预设的占比阈值，则判定所述指定图片与预存图片相似。
  18. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述利用预设的图片相似度判断方法，判断所述指定图片与预存图片是否相似的步骤之后,包括：
    若所述指定图片与预存图片相似，则获取所述上传终端对应的所有文字，并根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
    在所述上传终端对应的所有文字中搜索所述预存图片对应的文字文本，从而获取所述预存图片对应的文字文本在所述上传终端对应的所有文字中的位置信息；
    将所述位置信息发送给所述文字展示终端。
  19. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述若所述指定图片与预存图片不相似，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本的步骤,包括：
    若所述指定图片与预存图片不相似，则利用预设的初步文字识别技术识别所述图片文字文本图片中的预定区域，得到初步识别文字，其中预定区域的面积小于所述指定图片的面积；
    根据预设的图片与预存的文字文本的对应关系，获取所述预存图片对应的文字文本；
    对比所述初步识别文字与所述预存图片对应的文字文本，从而判断所述预存图片对应的文字文本是否包含所述初步识别文字；
    若所述预存图片对应的文字文本不包含所述初步识别文字，则将所述指定图片发送给指定文本识别终端，所述指定文本识别终端用于将所述指定图片中的文字内容识别为图片文字文本。
  20. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述接收所述指定文本识别终端发送的所述图片文字文本，并利用预设的文本相似度计算方法，计算所述图片文字文本与预存文字的相似度值，并判断所述相似度值是否大于预设的相似度阈值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字的步骤,包括：
    接收所述指定文本识别终端发送的所述图片文字文本；
    采用公式：
    Figure PCTCN2020088049-appb-100027
    计算所述图片文字文本与预存文字的相似度值，其中所述预存文字指通过对所述预存图片进行文字识别而获得的文字；其中similarity为相似度值，A为所述图片文字文本的词频向量，B为所述预存文字的词频向量，Ai为所述图片文字文本的第i个单词出现的次数，Bi为所述预存文字的第i个单词出现的次数；
    判断所述相似度值是否大于预设的相似度阈值。
  21. 根据权利要求15所述的计算机可读存储介质，其特征在于，所述若所述相似度值不大于预设的相似度阈值，则将所述图片文字文本发送给所述文字展示终端的步骤之前， 包括：
    获取预定时间内发送至所述文字展示终端的流量值；
    判断所述流量值是否大于预设的流量阈值；
    若所述流量值大于预设的流量阈值，则通过降低预设的标准阈值的数值的方式获得所述相似度阈值。
PCT/CN2020/088049 2019-05-30 2020-04-30 基于配置平台的数据传输方法、装置和计算机设备 WO2020238556A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910463920.X 2019-05-30
CN201910463920.XA CN110334702B (zh) 2019-05-30 2019-05-30 基于配置平台的数据传输方法、装置和计算机设备

Publications (1)

Publication Number Publication Date
WO2020238556A1 true WO2020238556A1 (zh) 2020-12-03

Family

ID=68140683

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/088049 WO2020238556A1 (zh) 2019-05-30 2020-04-30 基于配置平台的数据传输方法、装置和计算机设备

Country Status (2)

Country Link
CN (1) CN110334702B (zh)
WO (1) WO2020238556A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116389227A (zh) * 2023-03-23 2023-07-04 江苏锦花电子股份有限公司 一种基于物联网的智能预警系统及方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110334702B (zh) * 2019-05-30 2023-04-18 深圳壹账通智能科技有限公司 基于配置平台的数据传输方法、装置和计算机设备
CN111191552A (zh) * 2019-12-23 2020-05-22 合肥美的智能科技有限公司 基于视觉终端的图像识别方法以及视觉终端

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106469310A (zh) * 2015-08-17 2017-03-01 腾讯科技（深圳）有限公司 图片中字符提取方法及装置
CN108805115A (zh) * 2018-04-24 2018-11-13 五八同城信息技术有限公司 一种图片中文字的识别方法、装置、设备及存储介质
CN109325159A (zh) * 2018-08-01 2019-02-12 昆明理工大学 一种微博热点事件挖掘方法
CN109583438A (zh) * 2018-10-17 2019-04-05 龙马智芯（珠海横琴）科技有限公司 电子图像的文字的识别方法及图像处理装置
CN110334702A (zh) * 2019-05-30 2019-10-15 深圳壹账通智能科技有限公司 基于配置平台的数据传输方法、装置和计算机设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106325750A (zh) * 2016-08-26 2017-01-11 曹蕊 一种应用于终端设备的文字识别方法及系统
CN107992872B (zh) * 2017-12-25 2020-04-28 广东小天才科技有限公司 一种对图片进行文本识别的方法及移动终端
CN108256523B (zh) * 2018-01-11 2022-06-10 上海展扬通信技术有限公司 基于移动终端的识别方法、装置及计算机可读存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106469310A (zh) * 2015-08-17 2017-03-01 腾讯科技（深圳）有限公司 图片中字符提取方法及装置
CN108805115A (zh) * 2018-04-24 2018-11-13 五八同城信息技术有限公司 一种图片中文字的识别方法、装置、设备及存储介质
CN109325159A (zh) * 2018-08-01 2019-02-12 昆明理工大学 一种微博热点事件挖掘方法
CN109583438A (zh) * 2018-10-17 2019-04-05 龙马智芯（珠海横琴）科技有限公司 电子图像的文字的识别方法及图像处理装置
CN110334702A (zh) * 2019-05-30 2019-10-15 深圳壹账通智能科技有限公司 基于配置平台的数据传输方法、装置和计算机设备

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116389227A (zh) * 2023-03-23 2023-07-04 江苏锦花电子股份有限公司 一种基于物联网的智能预警系统及方法
CN116389227B (zh) * 2023-03-23 2023-11-17 江苏锦花电子股份有限公司 一种基于物联网的智能预警系统及方法

Also Published As

Publication number Publication date
CN110334702A (zh) 2019-10-15
CN110334702B (zh) 2023-04-18

Similar Documents

Publication Publication Date Title
WO2020238556A1 (zh) 基于配置平台的数据传输方法、装置和计算机设备
CN109146892B (zh) 一种基于美学的图像裁剪方法及装置
WO2020232872A1 (zh) 表格识别方法、装置、计算机设备和存储介质
WO2020238054A1 (zh) Pdf文档中图表的定位方法、装置及计算机设备
WO2020037898A1 (zh) 人脸特征点检测方法、装置、计算机设备及存储介质
WO2019056503A1 (zh) 门店监控评价方法、装置及存储介质
CN111666907B (zh) 一种视频中对象信息的识别方法、装置及服务器
CN110049310B (zh) 视频图像获取、视频质量检测方法和装置
US11714921B2 (en) Image processing method with ash code on local feature vectors, image processing device and storage medium
CN112149733B (zh) 模型训练、质量确定方法、装置、电子设备及存储介质
CN113222942A (zh) 多标签分类模型的训练方法和预测标签的方法
CN111898735A (zh) 蒸馏学习方法、装置、计算机设备和存储介质
WO2022042609A1 (zh) 提取热词的方法、装置、电子设备及介质
WO2021169642A1 (zh) 基于视频的眼球转向确定方法与系统
WO2023061082A1 (zh) 图像安全处理方法和装置、电子设备和存储介质
CN115331150A (zh) 图像识别方法、装置、电子设备、存储介质
CN111104917A (zh) 基于人脸的活体检测方法、装置、电子设备及介质
CN110942456B (zh) 篡改图像检测方法、装置、设备及存储介质
CN114463567A (zh) 一种基于区块链的智慧教育作业大数据防抄袭方法与系统
CN111078924B (zh) 图像检索方法、装置、终端及存储介质
CN105162931A (zh) 一种通信号码的分类方法及装置
CN115527106A (zh) 基于量化鱼类识别神经网络模型的成像识别方法及装置
CN110929118B (zh) 网络数据处理方法、设备、装置、介质
GB2608063A (en) Device and method for providing missing child search service based on face recognition using deep learning
CN111275183B (zh) 视觉任务的处理方法、装置和电子系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20812516

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20812516

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 22.03.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20812516

Country of ref document: EP

Kind code of ref document: A1