WO2020238555A1 - 动态调整时间阈值的方法、装置、计算机设备和存储介质 - Google Patents

动态调整时间阈值的方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2020238555A1
WO2020238555A1 PCT/CN2020/088041 CN2020088041W WO2020238555A1 WO 2020238555 A1 WO2020238555 A1 WO 2020238555A1 CN 2020088041 W CN2020088041 W CN 2020088041W WO 2020238555 A1 WO2020238555 A1 WO 2020238555A1
Authority
WO
WIPO (PCT)
Prior art keywords
text
client
preset
designated
recognized text
Prior art date
Application number
PCT/CN2020/088041
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 WO2020238555A1 publication Critical patent/WO2020238555A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/22Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/34Network arrangements or protocols for supporting network services or applications involving the movement of software or configuration parameters 
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/60Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/62Establishing a time schedule for servicing the requests

Definitions

  • This application relates to the computer field, and in particular to a method, device, computer equipment and storage medium for dynamically adjusting a time threshold.
  • the server is often used in current production and life to provide services to clients.
  • clients accessing the server at the same time, such as receiving a large number of clients to poll the interface in a short time, it will cause server pressure If it is too large, the performance will decrease, or even the server will crash.
  • server used for text recognition is visited too much, there are also problems of excessive pressure, performance degradation, and even server crash.
  • Traditional methods do not have a good solution to the above problems, and there is an urgent need for a technical solution that can solve the problem of server performance degradation caused by excessive traffic.
  • the main purpose of this application is to provide a method, device, computer equipment and storage medium for dynamically adjusting the time threshold, aiming to solve the problem of server performance degradation caused by excessive access.
  • this application proposes a method for dynamically adjusting the time threshold, which is applied to the server and includes the following steps: obtaining the request for calling a specified interface sent by the client, and judging the current time of the server and the client Whether the time interval of the time point at which the client last called the designated interface is greater than the preset time threshold; if the time interval is greater than the preset time threshold, the client is allowed to call the specified interface and receive the The specified text image sent by the client through the specified interface; the specified text image is recognized by the preset image text recognition technology to obtain the specified recognized text; the preset similarity calculation method is used to calculate the specified recognized text and pre-stored The similarity value of the last recognized text, the last recognized text is the recognition result of the server to the client's latest text image to be recognized; it is determined whether the similarity value is greater than the preset similarity Threshold; if the similarity degree value is greater than the preset similarity degree threshold, information that the recognized text
  • This application provides a device for dynamically adjusting the time threshold, applied to a server, and includes: invoking an application obtaining unit, configured to obtain an application for invoking a specified interface sent by a client, and determine the current time of the server and the client Whether the time interval of the time point of the last call of the designated interface is greater than a preset time threshold; the designated text and picture receiving unit is used to allow the client to call the client if the time interval is greater than the preset time threshold Specify the interface, and receive the specified text image sent by the client through the specified interface; the specified recognized text acquisition unit, used to recognize the specified text image using a preset image text recognition technology to obtain the specified recognized text; degree of similarity
  • the value calculation unit is configured to use a preset similarity calculation method to calculate the similarity value between the designated recognized text and the pre-stored previously recognized text, and the previously recognized text is the last time the server gave the client The sent recognition result of the text image to be recognized; a similarity value judgment unit for judging whether the similarity
  • 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 server pressure on the premise of ensuring the quality of service to the client.
  • FIG. 1 is a schematic flowchart of a method for dynamically adjusting a time threshold according to an embodiment of the application
  • FIG. 2 is a schematic block diagram of the structure of a device for dynamically adjusting a time threshold 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 method for dynamically adjusting a time threshold, applied to a server, and includes the following steps:
  • time interval is greater than a preset time threshold, allow the client to call the designated interface, and receive a designated text image sent by the client through the designated interface;
  • Servers are often used in current production and life to provide services to clients. When there are too many clients accessing the server at the same time, it will cause excessive server pressure, reduced performance, and even server crashes. When the server used for text recognition is visited too much, there are also problems of excessive pressure, performance degradation, and even server crash. Traditional methods do not have a good solution to the above problems, and there is an urgent need for a technical solution that can solve the problem of server performance degradation caused by excessive traffic.
  • This application dynamically adjusts the time threshold for different clients based on the result of judging whether the picture uploaded by the client to the server has been uploaded to the server, so as to reduce the pressure on the server while ensuring the quality of service to the client.
  • step S1 obtain the request for calling the designated interface sent by the client, and determine whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than the preset Time threshold.
  • the designated interface is used to interface the text picture of the client.
  • the client polls the specified interface at regular intervals.
  • this application adopts a preset time threshold to prevent the same client from calling the designated interface multiple times within the preset time threshold, thereby reducing server pressure.
  • the client is allowed to call the designated interface, and receive the designated text image sent by the client through the designated interface. If the time interval is greater than the preset time threshold, it indicates that the client does not frequently call the designated interface, so the client is allowed to call the designated interface and receive the designated text sent by the client through the designated interface image.
  • the text picture refers to a picture including text, for example, a picture including text obtained by a client taking a photo or scanning of a paper book.
  • the predetermined image character recognition technology is used to recognize the specified text image to obtain the specified recognized text.
  • the picture text recognition technology is used to recognize the picture, so as to propose the text information in the picture.
  • the picture text recognition technology can use the existing mature recognition technology, such as OCR technology (OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition), which can be used to convert image information into computer input technology that can be used). This will not be repeated here. According to this, the designated recognized character proposed by the designated character picture can be obtained.
  • recognizing the specified text picture by using a preset picture text recognition technology to obtain the specified recognized text includes: using the preset picture text recognition technology to recognize a predetermined area in the specified text picture to obtain the specified recognized text, wherein The area of the area is smaller than the area of the designated text picture, so that only by recognizing text in a smaller area, the pressure on the server can be further reduced.
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the pre-stored last recognized text, and the last recognized text is the most recent value of the server to the client.
  • the preset similarity calculation method, the calculation of the similarity value between the designated recognized text and the pre-stored previously recognized text can be any method, for example: using WMD algorithm (word move's distance), simhash algorithm, based on cosine similarity , The algorithm of word frequency vector to calculate the similarity value between the designated recognized text and the pre-stored previously recognized text.
  • using a preset similarity calculation method to calculate the similarity value between the designated recognized text and the previously stored previously recognized text includes: using a preset similarity calculation method to calculate the designated recognized text and the pre-stored previous recognition value.
  • the similarity value in the second recognized text, the previous recognized text is the recognition result of the predetermined area in the text image to be recognized last sent by the client by the server, so as to use the recognition in a smaller area You can judge whether the designated recognized text is similar to the previous recognized text, and can further reduce the pressure on the server (reduce the amount of calculation).
  • step S5 it is determined whether the similarity degree value is greater than a preset similarity degree threshold.
  • the similarity value reflects whether the specified text image uploaded by the client is the same as the text image uploaded last time. If the similarity value is greater than the preset similarity threshold, it indicates that the specified text image is the same as the most recently uploaded text image; if the similarity value is not greater than the preset similarity threshold, it indicates that the specified text image is the most recent upload The text picture is different.
  • the similarity degree value is greater than the preset similarity degree threshold, the information that does not need to update the recognized text is sent to the client, and the time is increased according to the preset time threshold adjustment rule Threshold. If the similarity value is greater than the preset similarity threshold, it indicates that the specified text picture is the same as the last uploaded text picture (for example, the client is in a standby state, and the image capture device is still collecting the same picture and uploading it to the server), Since the client already has the recognition result of the text image (the server has sent the recognition result of the last uploaded text image to the client), it is not necessary to send the scan result of the specified text image to the client, and accordingly The client sends information that does not need to update the recognized text (the client only needs to call the recognized text stored locally according to the information), thereby reducing the cost of sending information and further reducing the pressure on the server.
  • the preset time threshold adjustment rule is, for example, counting the total number of times and the number of successful times that the client invokes the specified interface within a specified time; using the quotient of the number of successful times/the total number of times, according to the forecast Set the corresponding relationship between the quotient and the time threshold adjustment parameter, and obtain the specified time threshold adjustment parameter corresponding to the quotient of the number of successes/the total number of times; add the specified time threshold adjustment on the basis of the time threshold parameter.
  • the information that does not need to update the recognized text is also accompanied by information that requires the client to retrieve and display the locally stored previously recognized text. Therefore, the client can retrieve and display the same text as the recognition result of the designated text picture from the local storage.
  • the time threshold after step S1 includes:
  • the call request is rejected. If the time interval is not greater than the preset time threshold, it indicates that the client frequently calls the designated interface, but if the current pressure on the server is not great, the client can still be provided with services. Therefore, the current number of visitors of the server is obtained, and it is judged whether the current number of visitors is greater than the preset number threshold. If the current number of visitors is greater than the preset number threshold, it indicates that the server is under pressure, and the call request is rejected accordingly. . Further, if the current number of visitors is not greater than a preset number threshold, the client is allowed to call the designated interface, and receive a designated text image sent by the client through the designated interface.
  • the step S3 of recognizing the specified text picture by using a preset image text recognition technology to obtain the specified recognized text includes:
  • ST3 Recognizing a predetermined area in the designated text picture by using a preset picture text recognition technology to obtain a designated recognized text, wherein the area of the predetermined area is smaller than the area of the designated text picture;
  • the use of the preset similarity calculation method to calculate the similarity value between the designated recognized text and the pre-stored previously recognized text, and the previously recognized text is the last time to be recognized sent by the server to the client Step S4 of the recognition result of the text picture includes:
  • the use of a predetermined area is achieved to further reduce server pressure.
  • the specified recognized text obtained is the text in the predetermined area in the specified text picture, so the server does not need to recognize the entire specified text picture, but only needs to recognize the predetermined area, thereby reducing calculations.
  • the last recognized text is the recognition result of the predetermined area in the text image to be recognized last sent by the client by the server, so that when calculating the similarity value, the recognized smaller area is used
  • the text inside can determine whether the designated recognized text is similar to the previous recognized text, thereby further reducing the pressure on the server (reducing the amount of calculation).
  • the call request carries a user's unique identifier
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the prestored previously recognized text.
  • the text is the recognition result of the text image to be recognized last sent by the client by the server before step S4, including:
  • the previously recognized text is obtained.
  • the call request carries the user's unique identifier, so the server can uniquely determine the user corresponding to the client through the user's unique identifier, and then determine the client's correspondence based on the preset correspondence between the user's unique identifier and the identification text (The server pre-stores the recognition result of the text image uploaded by the client beforehand). And the most recently added recognition text of the recognized text is marked with a reading mark, that is, the content of the reading mark is the recognition result of the text image to be recognized last sent by the client, so by corresponding to the client By querying the reading identifier in the recognized text of, the last recognized text can be obtained.
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the pre-stored previous recognized text, and the previous recognized text is the comparison between the server and the client.
  • the step S4 of the recognition result of the text image to be recognized that is sent last time includes:
  • 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.
  • the method includes:
  • the similarity value is not greater than the preset similarity threshold, it indicates that the specified text picture is not the same as the text image to be recognized last time sent by the client, so the specified recognized text needs to be sent to the client end.
  • the client has a requirement to continuously recognize characters, so according to a preset time threshold adjustment rule, the time threshold is reduced.
  • the decrease value of the time threshold is proportional to the number of times the client successfully invokes the designated interface within a predetermined time, so as to meet the needs of the client for fast character recognition.
  • the client sends information that does not need to update the recognized text, and according to the preset time threshold adjustment rule, the Step S6 of the time threshold includes:
  • S602. Calculate the quotient of the number of successes/the total number of times, and obtain a specified value corresponding to the quotient of the number of successes/the total number of times according to the corresponding relationship between the preset quotient and the time threshold adjustment parameter.
  • a time threshold adjustment parameter wherein the specified time threshold adjustment parameter is greater than 0;
  • the time threshold is increased according to the preset time threshold adjustment rule.
  • This embodiment uses historical data: the total number of times and the number of successes of calling the specified interface by the client within a specified time, and the quotient of the number of successes/the total number of times is used to determine the client's response to text The urgency of the identified need.
  • the quotient of the number of successes/the total number of times is higher, the corresponding client has a higher urgency for character recognition, and the corresponding time threshold adjustment parameter is also higher. Accordingly, the corresponding relationship between the quotient value and the time threshold adjustment parameter is preset.
  • a designated time threshold corresponding to the quotient of the number of successes/the total number of times is obtained Adjustment parameters.
  • the specified time threshold adjustment parameter is added on the basis of the time threshold to obtain the adjusted time threshold, thereby realizing the dynamic adjustment of the time threshold, thereby reducing the pressure on the server.
  • the method for dynamically adjusting the time threshold of the present application obtains the request for calling a specified interface sent by the client, and determines whether the time interval between the current time of the server and the time point when the client last called the specified interface is greater than A preset time threshold; if the time interval is greater than the preset time threshold, receive the specified text picture sent by the client through the specified interface; recognize the specified text picture to obtain the specified recognized text; calculate the Specify the similarity value between the recognized text and the pre-stored previously recognized text; if the similarity value is greater than the preset similarity threshold, then send the client the information that the recognized text does not need to be updated, and according to the preset time threshold Adjust the rules to increase the time threshold.
  • the pressure on the server is reduced.
  • an embodiment of the present application provides a device for dynamically adjusting a time threshold, which is applied to a server, and includes:
  • the call application obtaining unit 10 is used to obtain the call application for the designated interface sent by the client, and determine whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than a preset Time threshold;
  • the designated text picture receiving unit 20 is configured to, if the time interval is greater than a preset time threshold, allow the client to call the designated interface, and receive the designated text picture sent by the client through the designated interface;
  • the designated recognized text obtaining unit 30 is configured to recognize the designated text picture by using a preset picture text recognition technology to obtain the designated recognized text;
  • the similarity value calculation unit 40 is configured to use a preset similarity calculation method to calculate the similarity value between the designated recognized text and the pre-stored previously recognized text, and the previously recognized text is the comparison between the server and the client.
  • the similarity value judgment unit 50 is used to judge whether the similarity value is greater than a preset similarity threshold
  • the time threshold increasing unit 60 is configured to, if the similarity degree value is greater than a preset similarity degree threshold, send to the client information that the recognized text does not need to be updated, and increase the value according to the preset time threshold adjustment rule.
  • the time threshold is configured to, if the similarity degree value is greater than a preset similarity degree threshold, send to the client information that the recognized text does not need to be updated, and increase the value according to the preset time threshold adjustment rule.
  • Servers are often used in current production and life to provide services to clients. When there are too many clients accessing the server at the same time, it will cause excessive server pressure, reduced performance, and even server crashes. When the server used for text recognition is visited too much, there are also problems of excessive pressure, performance degradation, and even server crash. Traditional methods do not have a good solution to the above problems, and there is an urgent need for a technical solution that can solve the problem of server performance degradation caused by excessive traffic.
  • This application dynamically adjusts the time threshold for different clients based on the result of judging whether the picture uploaded by the client to the server has been uploaded to the server, so as to reduce the pressure on the server while ensuring the quality of service to the client.
  • the client obtains the request for calling the designated interface sent by the client, and determine whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than the preset Time threshold.
  • the designated interface is used to interface the text picture of the client.
  • the client polls the specified interface at regular intervals.
  • this application adopts a preset time threshold to prevent the same client from calling the designated interface multiple times within the preset time threshold, thereby reducing server pressure.
  • the client is allowed to call the designated interface, and receive the designated text image sent by the client through the designated interface. If the time interval is greater than the preset time threshold, it indicates that the client does not frequently call the designated interface, so the client is allowed to call the designated interface and receive the designated text sent by the client through the designated interface image.
  • the text picture refers to a picture including text, for example, a picture including text obtained by a client taking a photo or scanning of a paper book.
  • the predetermined image character recognition technology is used to recognize the specified text image to obtain the specified recognized text.
  • the picture text recognition technology is used to recognize the picture, so as to propose the text information in the picture.
  • the picture text recognition technology can use the existing mature recognition technology, such as OCR technology (OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition), which can be used to convert image information into computer input technology that can be used). This will not be repeated here. According to this, the designated recognized character proposed by the designated character picture can be obtained.
  • recognizing the specified text picture by using a preset picture text recognition technology to obtain the specified recognized text includes: using the preset picture text recognition technology to recognize a predetermined area in the specified text picture to obtain the specified recognized text, wherein The area of the area is smaller than the area of the designated text picture, so that only by recognizing text in a smaller area, the pressure on the server can be further reduced.
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the pre-stored last recognized text, and the last recognized text is the most recent value of the server to the client.
  • the preset similarity calculation method, the calculation of the similarity value between the designated recognized text and the pre-stored previously recognized text can be any method, for example: using WMD algorithm (word move's distance), simhash algorithm, based on cosine similarity , The algorithm of word frequency vector to calculate the similarity value between the designated recognized text and the pre-stored previously recognized text.
  • using a preset similarity calculation method to calculate the similarity value between the designated recognized text and the previously stored previously recognized text includes: using a preset similarity calculation method to calculate the designated recognized text and the pre-stored previous recognition value.
  • the similarity value in the second recognized text, the previous recognized text is the recognition result of the predetermined area in the text image to be recognized last sent by the client by the server, so as to use the recognition in a smaller area You can judge whether the designated recognized text is similar to the previous recognized text, and can further reduce the pressure on the server (reduce the amount of calculation).
  • the similarity degree value is greater than a preset similarity degree threshold.
  • the similarity value reflects whether the specified text image uploaded by the client is the same as the text image uploaded last time. If the similarity value is greater than the preset similarity threshold, it indicates that the specified text image is the same as the most recently uploaded text image; if the similarity value is not greater than the preset similarity threshold, it indicates that the specified text image is the most recent upload The text picture is different.
  • the similarity degree value is greater than the preset similarity degree threshold, the information that does not need to update the recognized text is sent to the client, and the time is increased according to the preset time threshold adjustment rule. Threshold. If the similarity value is greater than the preset similarity threshold, it indicates that the specified text picture is the same as the last uploaded text picture (for example, the client is in a standby state, and the image capture device is still collecting the same picture and uploading it to the server), Since the client already has the recognition result of the text image (the server has sent the recognition result of the last uploaded text image to the client), it is not necessary to send the scan result of the specified text image to the client, and accordingly The client sends information that does not need to update the recognized text (the client only needs to call the recognized text stored locally according to the information), thereby reducing the cost of sending information and further reducing the pressure on the server.
  • the preset time threshold adjustment rule is, for example, counting the total number of times and the number of successful times that the client invokes the specified interface within a specified time; using the quotient of the number of successful times/the total number of times, according to the forecast Set the corresponding relationship between the quotient and the time threshold adjustment parameter, and obtain the specified time threshold adjustment parameter corresponding to the quotient of the number of successes/the total number of times; add the specified time threshold adjustment on the basis of the time threshold parameter.
  • the information that does not need to update the recognized text is also accompanied by information that requires the client to retrieve and display the locally stored previously recognized text. Therefore, the client can retrieve and display the same text as the recognition result of the designated text picture from the local storage.
  • the device includes:
  • the current visitor number obtaining unit is configured to obtain the current visitor number of the server if the time interval is not greater than a preset time threshold;
  • the number threshold judging unit is used to determine whether the current number of visitors is greater than a preset number threshold
  • the calling rejection unit is configured to reject the calling application if the current number of visitors is greater than a preset number threshold.
  • the call request is rejected. If the time interval is not greater than the preset time threshold, it indicates that the client frequently calls the designated interface, but if the current pressure on the server is not high, then the client can still be provided with services. Therefore, the current number of visitors of the server is obtained, and it is judged whether the current number of visitors is greater than the preset number threshold. If the current number of visitors is greater than the preset number threshold, it indicates that the server is under pressure, and the call request is rejected accordingly. . Further, if the current number of visitors is not greater than a preset number threshold, the client is allowed to call the designated interface, and receive a designated text image sent by the client through the designated interface.
  • the designated recognized text obtaining unit 30 includes:
  • the designated recognized text acquisition subunit is configured to recognize a predetermined area in the designated text picture by using a preset picture text recognition technology to obtain a designated recognized text, wherein the area of the predetermined area is smaller than the area of the designated text picture;
  • the similarity value calculation unit 40 includes:
  • the similarity value calculation subunit is used to calculate the similarity value between the designated recognized text and the pre-stored previously recognized text by using a preset similarity calculation method, and the previously recognized text is the response of the server to the The recognition result of the predetermined area in the text image to be recognized last sent by the client.
  • the use of a predetermined area is achieved to further reduce server pressure.
  • the specified recognized text obtained is the text in the predetermined area in the specified text picture, so the server does not need to recognize the entire specified text picture, but only needs to recognize the predetermined area, thereby reducing calculations.
  • the last recognized text is the recognition result of the predetermined area in the text image to be recognized last sent by the client by the server, so that when calculating the similarity value, the recognized smaller area is used
  • the text inside can determine whether the designated recognized text is similar to the previous recognized text, thereby further reducing the pressure on the server (reducing the amount of calculation).
  • the call request carries a unique identifier of the user
  • the device includes:
  • the recognition text determining unit is configured to use the unique user identifier to determine the recognized text corresponding to the client according to the preset correspondence between the unique user identifier and the recognized text, wherein the recognized text added most recently in the recognized text Marked with a reading mark;
  • the previously recognized text obtaining unit is configured to obtain the previously recognized text by querying the reading identifier in the recognized text corresponding to the client.
  • the previously recognized text is obtained.
  • the call request carries the user's unique identifier, so the server can uniquely determine the user corresponding to the client through the user's unique identifier, and then determine the client's correspondence based on the preset correspondence between the user's unique identifier and the identification text (The server pre-stores the recognition result of the text image uploaded by the client beforehand). And the most recently added recognition text of the recognized text is marked with a reading mark, that is, the content of the reading mark is the recognition result of the text image to be recognized last sent by the client, so by corresponding to the client By querying the reading identifier in the recognized text of, the last recognized text can be obtained.
  • the similarity value calculation unit 40 includes:
  • the similarity value calculation subunit is used to adopt the formula:
  • 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.
  • the device includes:
  • a designated recognized text sending unit configured to send the designated recognized text to the client if the similarity degree value is not greater than a preset similarity degree threshold
  • the time threshold reducing unit is used to reduce the time threshold according to a preset time threshold adjustment rule.
  • the similarity value is not greater than the preset similarity threshold, it indicates that the specified text picture is not the same as the text image to be recognized last time sent by the client, so the specified recognized text needs to be sent to the client end.
  • the client has a requirement to continuously recognize characters, so according to a preset time threshold adjustment rule, the time threshold is reduced.
  • the decrease value of the time threshold is proportional to the number of times the client successfully invokes the designated interface within a predetermined time, so as to meet the needs of the client for fast character recognition.
  • the time threshold increasing unit 60 includes:
  • the number of calls counting subunit is used to count the total number of times and the number of successful calls of the specified interface by the client within a specified time;
  • the specified time threshold adjustment parameter acquisition subunit is used to calculate the quotient of the number of successes/the total number of times, and obtain the corresponding relationship between the preset quotient and the time threshold adjustment parameter and the number of successes/the total number of times.
  • the specified time threshold adjustment parameter corresponding to the quotient of the total number of times, wherein the specified time threshold adjustment parameter is greater than 0;
  • the specified time threshold adjustment parameter increase subunit is used to increase the specified time threshold adjustment parameter on the basis of the time threshold.
  • the time threshold is increased according to the preset time threshold adjustment rule.
  • This embodiment uses historical data: the total number of times and the number of successes of calling the specified interface by the client within a specified time, and the quotient of the number of successes/the total number of times is used to determine the client's response to text The urgency of the identified need.
  • the quotient of the number of successes/the total number of times is higher, the corresponding client has a higher urgency for character recognition, and the corresponding time threshold adjustment parameter is also higher. Accordingly, the corresponding relationship between the quotient value and the time threshold adjustment parameter is preset.
  • a designated time threshold corresponding to the quotient of the number of successes/the total number of times is obtained Adjustment parameters.
  • the specified time threshold adjustment parameter is added on the basis of the time threshold to obtain the adjusted time threshold, thereby realizing the dynamic adjustment of the time threshold, thereby reducing the pressure on the server.
  • the device for dynamically adjusting the time threshold of the present application obtains the request for calling a specified interface sent by the client, and determines whether the time interval between the current time of the server and the time point when the client last called the specified interface is greater than A preset time threshold; if the time interval is greater than the preset time threshold, receive the specified text picture sent by the client through the specified interface; recognize the specified text picture to obtain the specified recognized text; calculate the Specify the similarity value between the recognized text and the pre-stored previously recognized text; if the similarity value is greater than the preset similarity threshold, then send the client the information that the recognized text does not need to be updated, and according to the preset time threshold Adjust the rules to increase the time threshold.
  • the pressure on the server is reduced.
  • an embodiment of the present application 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 method of dynamically adjusting the time threshold.
  • 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 method of dynamically adjusting the time threshold.
  • the above-mentioned processor executes the above-mentioned method for dynamically adjusting the time threshold, including the following steps: obtaining a request for calling a designated interface sent by a client, and judging the current time of the server and the time when the client last called the designated interface Whether the time interval between points is greater than the preset time threshold; if the time interval is greater than the preset time threshold, allow the client to call the designated interface, and receive the designated sent by the client through the designated interface Text picture; using a preset picture text recognition technology to recognize the specified text picture to obtain a specified recognized text; using a preset similarity calculation method to calculate the similarity value between the specified recognized text and the prestored previous recognized text, The previously recognized text is the recognition result of the text image to be recognized last sent by the client by the server; it is determined whether the similarity value is greater than a preset similarity threshold; if the similarity value is greater than the preset similarity If the similarity threshold is set, the information that does not need to update the recognized text is sent to
  • the acquiring application for calling the designated interface sent by the client and determining whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than a preset After the step of the time threshold, it includes: if the time interval is not greater than a preset time threshold, obtaining the current number of visitors to the server; judging whether the current number of visitors is greater than the preset threshold; if the current If the number of visitors is greater than the preset threshold, the call application is rejected.
  • the step of recognizing the designated text picture by using a preset picture text recognition technology to obtain a designated recognized text includes: using a preset picture text recognition technology to recognize a predetermined area in the designated text picture , Obtain the designated recognized text, wherein the area of the predetermined area is smaller than the area of the designated text picture; said using the preset similarity calculation method to calculate the similarity value between the designated recognized text and the pre-stored previous recognized text, so
  • the previous recognized text is the step of the server's recognition result of the text image to be recognized last sent by the client, including: using a preset similarity calculation method to calculate the designated recognized text and the pre-stored previous
  • the similarity value in the recognized text, and the previously recognized text is the recognition result of the predetermined area in the text image to be recognized last sent by the client by the server.
  • the call request carries a user's unique identifier
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the prestored previously recognized text.
  • the text is before the step of the server's recognition result of the text image to be recognized last sent by the client, including: using the user unique identifier to determine the corresponding relationship between the preset user unique identifier and the recognized text
  • the recognized text corresponding to the client wherein the most recently added recognized text of the recognized text is marked with a reading identifier; the previously recognized text is obtained by querying the reading identifier in the recognized text corresponding to the client .
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the pre-stored previous recognized text, and the previous recognized text is the comparison between the server and the client.
  • the steps of the recognition result of the last sent text image to be recognized include:
  • the method includes: if the similarity degree value is not greater than the preset similarity degree threshold, then the designated recognized text Send to the client; reduce the time threshold according to a preset time threshold adjustment rule.
  • the client sends information that does not need to update the recognized text, and according to the preset time threshold adjustment rule, the The step of time threshold includes: counting the total number of times and the number of successful times of calling the specified interface by the client within a specified time; calculating the quotient of the number of successful times/the total number of times, and according to a preset quotient The corresponding relationship between the value and the time threshold adjustment parameter, and obtain the specified time threshold adjustment parameter corresponding to the quotient of the number of successes/the total number of times, wherein the specified time threshold adjustment parameter is greater than 0; on the basis of the time threshold Add the specified time threshold adjustment parameter.
  • the computer device of this application obtains the request for calling the designated interface sent by the client, and determines whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than the preset time Threshold; if the time interval is greater than the preset time threshold, receive the specified text picture sent by the client through the specified interface; recognize the specified text picture to obtain the specified recognized text; calculate the specified recognized text and The pre-stored similarity value of the last recognized text; if the similarity value is greater than the preset similarity threshold, the message that the recognized text does not need to be updated is sent to the client, and the rule is adjusted according to the preset time threshold to increase Larger the time threshold.
  • the pressure on the server is reduced.
  • 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 method for dynamically adjusting the time threshold includes the following steps: obtaining a request for calling a designated interface sent by a client, and determining whether the time interval between the current time of the server and the time point when the client last called the designated interface Is greater than the preset time threshold; if the time interval is greater than the preset time threshold, the client is allowed to call the designated interface, and receive the designated text image sent by the client through the designated interface;
  • the image text recognition technology is set to recognize the specified text image to obtain the specified recognized text; use a preset similarity calculation method to calculate the similarity value between the specified recognized text and the pre-stored previously recognized text, the previous recognition
  • the text is the recognition result of the text image to be recognized last sent by the client by the server; it is judged whether the similarity
  • the acquiring application for calling the designated interface sent by the client and determining whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than a preset After the step of the time threshold, it includes: if the time interval is not greater than a preset time threshold, obtaining the current number of visitors to the server; judging whether the current number of visitors is greater than the preset threshold; if the current If the number of visitors is greater than the preset threshold, the call application is rejected.
  • the step of recognizing the designated text picture by using a preset picture text recognition technology to obtain a designated recognized text includes: using a preset picture text recognition technology to recognize a predetermined area in the designated text picture , Obtain the designated recognized text, wherein the area of the predetermined area is smaller than the area of the designated text picture; said using the preset similarity calculation method to calculate the similarity value between the designated recognized text and the pre-stored previous recognized text, so
  • the previous recognized text is the step of the server's recognition result of the text image to be recognized last sent by the client, including: using a preset similarity calculation method to calculate the designated recognized text and the pre-stored previous
  • the similarity value in the recognized text, and the previously recognized text is the recognition result of the predetermined area in the text image to be recognized last sent by the client by the server.
  • the call request carries a user's unique identifier
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the prestored previously recognized text.
  • the text is before the step of the server's recognition result of the text image to be recognized last sent by the client, including: using the user unique identifier to determine the corresponding relationship between the preset user unique identifier and the recognized text
  • the recognized text corresponding to the client wherein the most recently added recognized text of the recognized text is marked with a reading identifier; the previously recognized text is obtained by querying the reading identifier in the recognized text corresponding to the client .
  • the preset similarity calculation method is used to calculate the similarity value between the designated recognized text and the pre-stored previous recognized text, and the previous recognized text is the comparison between the server and the client.
  • the steps of the recognition result of the last sent text image to be recognized include:
  • the method includes: if the similarity degree value is not greater than the preset similarity degree threshold, then the designated recognized text Send to the client; reduce the time threshold according to a preset time threshold adjustment rule.
  • the client sends information that does not need to update the recognized text, and according to the preset time threshold adjustment rule, the The step of time threshold includes: counting the total number of times and the number of successful times of calling the specified interface by the client within a specified time; calculating the quotient of the number of successful times/the total number of times, and according to a preset quotient The corresponding relationship between the value and the time threshold adjustment parameter, and obtain the specified time threshold adjustment parameter corresponding to the quotient of the number of successes/the total number of times, wherein the specified time threshold adjustment parameter is greater than 0; on the basis of the time threshold Add the specified time threshold adjustment parameter.
  • the computer-readable storage medium of the present application obtains the request for calling a designated interface sent by the client, and determines whether the time interval between the current time of the server and the time point when the client last called the designated interface is greater than the expected time. If the time interval is greater than the preset time threshold, receive the specified text picture sent by the client through the specified interface; recognize the specified text picture to obtain the specified recognized text; calculate the specified The similarity value between the recognized text and the pre-stored previously recognized text; if the similarity value is greater than the preset similarity threshold, the client is sent a message that the recognized text does not need to be updated, and adjusted according to the preset time threshold Rule, increase the time threshold.
  • the pressure on the server is reduced.
  • 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.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Character Discrimination (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本申请涉及运维技术,揭示了一种动态调整时间阈值的方法,包括:获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;得到指定识别文字;计算所述指定识别文字与预存的前次识别文字的相似程度值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大时间阈值。从而在保证对客户端的服务质量的前提下,降低服务器压力。

Description

动态调整时间阈值的方法、装置、计算机设备和存储介质
本申请要求于2019年5月30日提交中国专利局,申请号为201910463109.1、发明名称为“动态调整时间阈值的方法、装置、计算机设备和存储介质”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,特别是涉及到一种动态调整时间阈值的方法、装置、计算机设备和存储介质。
背景技术
服务器在目前生产、生活中经常使用,用于对客户端提供服务,当同时对服务器进行访问的客户端过多时,例如接收大量客户端在短时间内进行轮询调用接口时,将导致服务器压力过大,性能降低,甚至于服务器崩溃。用于文字识别的服务器在访问量过多时,同样存在压力过大,性能降低,甚至于服务器崩溃的问题。传统方法对于上述问题没有很好的解决方案,亟需能够解决由访问量过多带来的服务器性能降低的问题的技术方案。
技术问题
本申请的主要目的为提供一种动态调整时间阈值的方法、装置、计算机设备和存储介质,旨在解决由访问量过多带来的服务器性能降低的问题。
技术解决方案
为了实现上述发明目的,本申请提出一种动态调整时间阈值的方法,应用于服务器,包括以下步骤:获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;判断所述相似程度值是否大于预设的相似程度阈值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
本申请提供一种动态调整时间阈值的装置,应用于服务器,包括:调用申请获取单元,用于获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;指定文字图片接收单元,用于若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;指定识别文字获取单元,用于利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;相似程度值计算单元,用于利用预设的相似度计算方法,计算所述指定识别 文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;相似程度值判断单元,用于判断所述相似程度值是否大于预设的相似程度阈值;时间阈值增大单元,用于若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
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
本申请在保证对客户端的服务质量的前提下,降低服务器压力。
附图说明
图1为本申请一实施例的动态调整时间阈值的方法的流程示意图;
图2为本申请一实施例的动态调整时间阈值的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动态调整时间阈值的方法,应用于服务器,包括以下步骤:
S1、获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;
S2、若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;
S3、利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;
S4、利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;
S5、判断所述相似程度值是否大于预设的相似程度阈值;
S6、若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间 阈值。
服务器在目前生产、生活中经常使用,用于对客户端提供服务,当同时对服务器进行访问的客户端过多时,将导致服务器压力过大,性能降低,甚至于服务器崩溃。用于文字识别的服务器在访问量过多时,同样存在压力过大,性能降低,甚至于服务器崩溃的问题。传统方法对于上述问题没有很好的解决方案,亟需能够解决由访问量过多带来的服务器性能降低的问题的技术方案。本申请根据判断客户端上传服务器的图片是否曾经上传过服务器的结果,从而针对不同的客户端动态调节时间阈值,以在保证对客户端的服务质量的前提下,减轻服务器的压力。
如上述步骤S1所述,获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值。其中所述指定接口用于接口客户端的文字图片。当调用指定接口的客户端过多时,会使服务器的压力过大,例如客户端每隔一段时间轮询指定接口,当高并发的情况存在时,同时调用指定接口的客户端过多,使服务器的压力过大。因此本申请采用预设的时间阈值的方式,避免同一客户端在预设的时间阈值之内多次调用指定接口,从而降低服务器压力。
如上述步骤S2所述,若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片。若所述时间间隔大于预设的时间阈值,表明所述客户端并非频繁调用指定接口,因此允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片。其中,所述文字图片指包括文字的图片,例如客户端对纸质书籍进行拍照、扫描而获得的包括文字的图片。
如上述步骤S3所述,利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字。所述图片文字识别技术用于对图片进去识别,从而提出图片中的文字信息。所述图片文字识别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。据此,可获得由指定文字图片中提出得到的指定识别文字。进一步地,利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字包括:利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积,从而通过识别更小的区域内的文字即可,能够进一步减小服务器的压力。
如上述步骤S4所述,利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果。其中,预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值可为任意方法,例如:采用WMD算法(word mover’s distance)、simhash算法、基于余弦相似度、词频向量的算法以计算所述指定识别文字与预存的前次识 别文字的相似程度值。进一步地,利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值包括:利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果,从而利用识别得到的更小的区域内的文字即可判断指定识别文字与前次识别文字是否相似,还能进一步减小服务器的压力(减少计算量)。
如上述步骤S5所述,判断所述相似程度值是否大于预设的相似程度阈值。所述相似程度值反应了客户端上传的指定文字图片与最近一次上传的文字图片是否相同。若所述相似程度值大于预设的相似程度阈值,表明指定文字图片与最近一次上传的文字图片相同;若所述相似程度值不大于预设的相似程度阈值,表明指定文字图片与最近一次上传的文字图片不同。
如上述步骤S6所述,若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。若所述相似程度值大于预设的相似程度阈值,表明指定文字图片与最近一次上传的文字图片相同(例如客户端处于待机状态,并且图像采集设备仍在采集相同的图片并上传至服务器),由于客户端已经具有文字图片的识别结果(服务器已将对最近一次上传的文字图片的识别结果发送给了客户端),因此没有必要将对指定文字图片的扫描结果发送给客户端,据此向所述客户端发送无需更新识别文字的信息(客户端根据所述信息,调用本地存储的识别文字即可),从而减少了发送信息的开销,进一步减小服务器压力。并且,由于客户端本次上传的文字图片与最近一次上传的文字图片相同,表明所述客户端实际调用指定接口的需求较低,据此增大所述时间阈值,从而进一步减少服务器的压力。其中,预设的时间阈值调整规则例如为:统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;利用所述成功次数/所述总次数的商值,根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数;在所述时间阈值的基础上增加所述指定时间阈值调整参数。进一步地,所述无需更新识别文字的信息还附带有要求所述客户端调取本地存储的前次识别文字并展示的信息。从而客户端从本地存储即可调取并展示与所述指定文字图片的识别结果相同的文字。
在一个实施方式中,所述获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值的步骤S1之后,包括:
S11、若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;
S12、判断所述当前访问人数是否大于预设的人数阈值;
S13、若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
如上所述,实现了拒绝所述调用申请。若所述时间间隔不大于预设的时 间阈值,表明所述客户端频繁调用指定接口,但是若服务器当前的压力不大,那么仍旧可以给所述客户端提供服务。因此获取所述服务器的当前访问人数,判断所述当前访问人数是否大于预设的人数阈值,若所述当前访问人数大于预设的人数阈值,表明服务器压力过大,据此拒绝所述调用申请。进一步地,若所述当前访问人数不大于预设的人数阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片。
在一个实施方式中,所述利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字的步骤S3,包括:
ST3、利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;
所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤S4,包括:
ST4、利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
如上所述,实现了通过采用预定区域的方式,进一步减少服务器压力。本实施方式中,获得的指定识别文字为所述指定文字图片中的预定区域中的文字,因此服务器并不需要对整个指定文字图中进行识别,而仅需识别预定区域即可,从而减少计算量以减少服务器压力。相应地,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果,从而在计算相似程度值时,利用识别得到的更小的区域内的文字即可判断指定识别文字与前次识别文字是否相似,从而进一步减小服务器的压力(减少计算量)。
在一个实施方式中,所述调用申请携带有用户唯一标识,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤S4之前,包括:
S31、利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文字标记有阅读标识;
S32、通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
如上所述,实现了获取所述前次识别文字。所述调用申请携带有用户唯一标识,因此服务器通过用户唯一标识,即可唯一确定所述客户端对应的用户,再根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本(所述服务器预先保存有所述客户端之前上传的文字图片的识别结果)。而所述识别文本的最近一次添加的识别文字标记有阅读标识,即所述阅读标识标记的内容为所述客户端最近一次发送的待识别文字图片的识别 结果,因此通过在所述客户端对应的识别文本中查询所述阅读标识,即可得到所述前次识别文字。
在一个实施方式中,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤S4,包括:
S401、采用公式:
Figure PCTCN2020088041-appb-000001
计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
如上所述,实现了计算所述指定识别文字与预存的前次识别文字的相似程度值。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。
在一个实施方式中,所述判断所述相似程度值是否大于预设的相似程度阈值的步骤S5之后,包括:
S51、若所述相似程度值不大于预设的相似程度阈值,则将所述指定识别文字发送给所述客户端;
S52、根据预设的时间阈值调整规则,减小所述时间阈值。
如上所述,实现了动态调整所述时间阈值。若所述相似程度值不大于预设的相似程度阈值,表明所述指定文字图片与所述客户端最近一次发送的待识别文字图片不相同,因此需要将所述指定识别文字发送给所述客户端。并且所述客户端具有连续识别文字的需求,因此根据预设的时间阈值调整规则,减小所述时间阈值。进一步地,所述时间阈值的减小数值与预定时间内所述客户端成功调用所述指定接口的次数成正比,从而满足所述客户端对于快速文字识别的需要。
在一个实施方式中,所述若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值的步骤S6,包括:
S601、统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;
S602、计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值调整参数大于0;
S603、在所述时间阈值的基础上增加所述指定时间阈值调整参数。
如上所述,实现了根据预设的时间阈值调整规则,增大所述时间阈值。本实施方式利用历史数据:在指定时间内,所述客户端调用所述指定接口的总次数与成功次数,并利用所述成功次数/所述总次数的商值以确定所述客户端对于文字识别的需求的迫切程度。当所述成功次数/所述总次数的商值较高时,相应的所述客户端对于文字识别的需求的迫切程度较高,此时对应的时间阈值调整参数也较大。据此,预设商值与时间阈值调整参数的对应关系。从而,利用所述成功次数/所述总次数的商值,根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数。并在所述时间阈值的基础上增加所述指定时间阈值调整参数,即可得到调整后的时间阈值,从而实现了动态调整时间阈值,进而减少了服务器的压力。
本申请的动态调整时间阈值的方法,获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则接收所述客户端通过所述指定接口发送的指定文字图片;识别所述指定文字图片,得到指定识别文字;计算所述指定识别文字与预存的前次识别文字的相似程度值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。从而在保证对客户端的服务质量的前提下,降低服务器压力。
参照图2,本申请实施例提供一种动态调整时间阈值的装置,应用于服务器,包括:
调用申请获取单元10,用于获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;
指定文字图片接收单元20,用于若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;
指定识别文字获取单元30,用于利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;
相似程度值计算单元40,用于利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;
相似程度值判断单元50,用于判断所述相似程度值是否大于预设的相似程度阈值;
时间阈值增大单元60,用于若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
服务器在目前生产、生活中经常使用,用于对客户端提供服务,当同时对服务器进行访问的客户端过多时,将导致服务器压力过大,性能降低,甚至于服务器崩溃。用于文字识别的服务器在访问量过多时,同样存在压力过大,性能降低,甚至于服务器崩溃的问题。传统方法对于上述问题没有很好的解决方案,亟需能够解决由访问量过多带来的服务器性能降低的问题的技术方案。本申请根据判断客户端上传服务器的图片是否曾经上传过服务器的结果,从而针对不同的客户端动态调节时间阈值,以在保证对客户端的服务质量的前提下,减轻服务器的压力。
如上述单元10所述,获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值。其中所述指定接口用于接口客户端的文字图片。当调用指定接口的客户端过多时,会使服务器的压力过大,例如客户端每隔一段时间轮询指定接口,当高并发的情况存在时,同时调用指定接口的客户端过多,使服务器的压力过大。因此本申请采用预设的时间阈值的方式,避免同一客户端在预设的时间阈值之内多次调用指定接口,从而降低服务器压力。
如上述单元20所述,若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片。若所述时间间隔大于预设的时间阈值,表明所述客户端并非频繁调用指定接口,因此允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片。其中,所述文字图片指包括文字的图片,例如客户端对纸质书籍进行拍照、扫描而获得的包括文字的图片。
如上述单元30所述,利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字。所述图片文字识别技术用于对图片进去识别,从而提出图片中的文字信息。所述图片文字识别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。据此,可获得由指定文字图片中提出得到的指定识别文字。进一步地,利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字包括:利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积,从而通过识别更小的区域内的文字即可,能够进一步减小服务器的压力。
如上述单元40所述,利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对 所述客户端最近一次发送的待识别文字图片的识别结果。其中,预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值可为任意方法,例如:采用WMD算法(word mover’s distance)、simhash算法、基于余弦相似度、词频向量的算法以计算所述指定识别文字与预存的前次识别文字的相似程度值。进一步地,利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值包括:利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果,从而利用识别得到的更小的区域内的文字即可判断指定识别文字与前次识别文字是否相似,还能进一步减小服务器的压力(减少计算量)。
如上述单元50所述,判断所述相似程度值是否大于预设的相似程度阈值。所述相似程度值反应了客户端上传的指定文字图片与最近一次上传的文字图片是否相同。若所述相似程度值大于预设的相似程度阈值,表明指定文字图片与最近一次上传的文字图片相同;若所述相似程度值不大于预设的相似程度阈值,表明指定文字图片与最近一次上传的文字图片不同。
如上述单元60所述,若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。若所述相似程度值大于预设的相似程度阈值,表明指定文字图片与最近一次上传的文字图片相同(例如客户端处于待机状态,并且图像采集设备仍在采集相同的图片并上传至服务器),由于客户端已经具有文字图片的识别结果(服务器已将对最近一次上传的文字图片的识别结果发送给了客户端),因此没有必要将对指定文字图片的扫描结果发送给客户端,据此向所述客户端发送无需更新识别文字的信息(客户端根据所述信息,调用本地存储的识别文字即可),从而减少了发送信息的开销,进一步减小服务器压力。并且,由于客户端本次上传的文字图片与最近一次上传的文字图片相同,表明所述客户端实际调用指定接口的需求较低,据此增大所述时间阈值,从而进一步减少服务器的压力。其中,预设的时间阈值调整规则例如为:统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;利用所述成功次数/所述总次数的商值,根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数;在所述时间阈值的基础上增加所述指定时间阈值调整参数。进一步地,所述无需更新识别文字的信息还附带有要求所述客户端调取本地存储的前次识别文字并展示的信息。从而客户端从本地存储即可调取并展示与所述指定文字图片的识别结果相同的文字。
在一个实施方式中,所述装置,包括:
当前访问人数获取单元,用于若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;
人数阈值判断单元,用于判断所述当前访问人数是否大于预设的人数阈值;
拒绝调用单元,用于若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
如上所述,实现了拒绝所述调用申请。若所述时间间隔不大于预设的时间阈值,表明所述客户端频繁调用指定接口,但是若服务器当前的压力不大,那么仍旧可以给所述客户端提供服务。因此获取所述服务器的当前访问人数,判断所述当前访问人数是否大于预设的人数阈值,若所述当前访问人数大于预设的人数阈值,表明服务器压力过大,据此拒绝所述调用申请。进一步地,若所述当前访问人数不大于预设的人数阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片。
在一个实施方式中,所述指定识别文字获取单元30,包括:
指定识别文字获取子单元,用于利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;
所述相似程度值计算单元40,包括:
相似程度值计算子单元,用于利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
如上所述,实现了通过采用预定区域的方式,进一步减少服务器压力。本实施方式中,获得的指定识别文字为所述指定文字图片中的预定区域中的文字,因此服务器并不需要对整个指定文字图中进行识别,而仅需识别预定区域即可,从而减少计算量以减少服务器压力。相应地,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果,从而在计算相似程度值时,利用识别得到的更小的区域内的文字即可判断指定识别文字与前次识别文字是否相似,从而进一步减小服务器的压力(减少计算量)。
在一个实施方式中,所述调用申请携带有用户唯一标识,所述装置,包括:
识别文本确定单元,用于利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文字标记有阅读标识;
前次识别文字获取单元,用于通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
如上所述,实现了获取所述前次识别文字。所述调用申请携带有用户唯一标识,因此服务器通过用户唯一标识,即可唯一确定所述客户端对应的用户,再根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本(所述服务器预先保存有所述客户端之前上传的文字图片的识别结果)。而所述识别文本的最近一次添加的识别文字标记有阅读标识,即所述阅读标识标记的内容为所述客户端最近一次发送的待识别文字图片的识别 结果,因此通过在所述客户端对应的识别文本中查询所述阅读标识,即可得到所述前次识别文字。
在一个实施方式中,所述相似程度值计算单元40,包括:
相似程度值计算子单元,用于采用公式:
Figure PCTCN2020088041-appb-000002
计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
如上所述,实现了计算所述指定识别文字与预存的前次识别文字的相似程度值。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。
在一个实施方式中,所述装置,包括:
指定识别文字发送单元,用于若所述相似程度值不大于预设的相似程度阈值,则将所述指定识别文字发送给所述客户端;
时间阈值减小单元,用于根据预设的时间阈值调整规则,减小所述时间阈值。
如上所述,实现了动态调整所述时间阈值。若所述相似程度值不大于预设的相似程度阈值,表明所述指定文字图片与所述客户端最近一次发送的待识别文字图片不相同,因此需要将所述指定识别文字发送给所述客户端。并且所述客户端具有连续识别文字的需求,因此根据预设的时间阈值调整规则,减小所述时间阈值。进一步地,所述时间阈值的减小数值与预定时间内所述客户端成功调用所述指定接口的次数成正比,从而满足所述客户端对于快速文字识别的需要。
在一个实施方式中,所述时间阈值增大单元60,包括:
调用次数统计子单元,用于统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;
指定时间阈值调整参数获取子单元,用于计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值调整参数大于0;
指定时间阈值调整参数增加子单元,用于在所述时间阈值的基础上增加所 述指定时间阈值调整参数。
如上所述,实现了根据预设的时间阈值调整规则,增大所述时间阈值。本实施方式利用历史数据:在指定时间内,所述客户端调用所述指定接口的总次数与成功次数,并利用所述成功次数/所述总次数的商值以确定所述客户端对于文字识别的需求的迫切程度。当所述成功次数/所述总次数的商值较高时,相应的所述客户端对于文字识别的需求的迫切程度较高,此时对应的时间阈值调整参数也较大。据此,预设商值与时间阈值调整参数的对应关系。从而,利用所述成功次数/所述总次数的商值,根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数。并在所述时间阈值的基础上增加所述指定时间阈值调整参数,即可得到调整后的时间阈值,从而实现了动态调整时间阈值,进而减少了服务器的压力。
本申请的动态调整时间阈值的装置,获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则接收所述客户端通过所述指定接口发送的指定文字图片;识别所述指定文字图片,得到指定识别文字;计算所述指定识别文字与预存的前次识别文字的相似程度值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。从而在保证对客户端的服务质量的前提下,降低服务器压力。
参照图3,本申请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动态调整时间阈值的方法所用数据。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动态调整时间阈值的方法。
上述处理器执行上述动态调整时间阈值的方法,包括以下步骤:获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;判断所述相似程度值是否大于预设的相似程度阈值; 若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
在一个实施方式中,所述获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值的步骤之后,包括:若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;判断所述当前访问人数是否大于预设的人数阈值;若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
在一个实施方式中,所述利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字的步骤,包括:利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
在一个实施方式中,所述调用申请携带有用户唯一标识,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤之前,包括:利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文字标记有阅读标识;通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
在一个实施方式中,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
采用公式:
Figure PCTCN2020088041-appb-000003
计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
在一个实施方式中,所述判断所述相似程度值是否大于预设的相似程度阈值的步骤之后,包括:若所述相似程度值不大于预设的相似程度阈值,则 将所述指定识别文字发送给所述客户端;根据预设的时间阈值调整规则,减小所述时间阈值。
在一个实施方式中,所述若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值的步骤,包括:统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值调整参数大于0;在所述时间阈值的基础上增加所述指定时间阈值调整参数。
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,获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则接收所述客户端通过所述指定接口发送的指定文字图片;识别所述指定文字图片,得到指定识别文字;计算所述指定识别文字与预存的前次识别文字的相似程度值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。从而在保证对客户端的服务质量的前提下,降低服务器压力。
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动态调整时间阈值的方法,包括以下步骤:获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;判断所述相似程度值是否大于预设的相似程度阈值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
在一个实施方式中,所述获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值的步骤之后,包括:若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;判断所述当前访问人数是否大于预设的人数阈值;若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
在一个实施方式中,所述利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字的步骤,包括:利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
在一个实施方式中,所述调用申请携带有用户唯一标识,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤之前,包括:利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文字标记有阅读标识;通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
在一个实施方式中,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
采用公式:
Figure PCTCN2020088041-appb-000004
计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
在一个实施方式中,所述判断所述相似程度值是否大于预设的相似程度阈值的步骤之后,包括:若所述相似程度值不大于预设的相似程度阈值,则将所述指定识别文字发送给所述客户端;根据预设的时间阈值调整规则,减小所述时间阈值。
在一个实施方式中,所述若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值的步骤,包括:统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值 调整参数大于0;在所述时间阈值的基础上增加所述指定时间阈值调整参数。
本申请的计算机可读存储介质,获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;若所述时间间隔大于预设的时间阈值,则接收所述客户端通过所述指定接口发送的指定文字图片;识别所述指定文字图片,得到指定识别文字;计算所述指定识别文字与预存的前次识别文字的相似程度值;若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。从而在保证对客户端的服务质量的前提下,降低服务器压力。
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. 一种动态调整时间阈值的方法,应用于服务器,其中,包括:
    获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;
    若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;
    利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;
    利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;
    判断所述相似程度值是否大于预设的相似程度阈值;
    若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
  2. 根据权利要求1所述的动态调整时间阈值的方法,其中,所述获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值的步骤之后,包括:
    若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;
    判断所述当前访问人数是否大于预设的人数阈值;
    若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
  3. 根据权利要求1所述的动态调整时间阈值的方法,其中,所述利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字的步骤,包括:
    利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;
    所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
    利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
  4. 根据权利要求1所述的动态调整时间阈值的方法,其中,所述调用申请携带有用户唯一标识,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤之前,包括:
    利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文 字标记有阅读标识;
    通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
  5. 根据权利要求1所述的动态调整时间阈值的方法,其中,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
    采用公式:
    Figure PCTCN2020088041-appb-100001
    计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
  6. 根据权利要求1所述的动态调整时间阈值的方法,其中,所述判断所述相似程度值是否大于预设的相似程度阈值的步骤之后,包括:
    若所述相似程度值不大于预设的相似程度阈值,则将所述指定识别文字发送给所述客户端;
    根据预设的时间阈值调整规则,减小所述时间阈值。
  7. 根据权利要求1所述的动态调整时间阈值的方法,其中,所述若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值的步骤,包括:
    统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;
    计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值调整参数大于0;
    在所述时间阈值的基础上增加所述指定时间阈值调整参数。
  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:
    获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;
    若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;
    利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;
    利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送 的待识别文字图片的识别结果;
    判断所述相似程度值是否大于预设的相似程度阈值;
    若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
  9. 根据权利要求8所述的计算机设备,其中,所述获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值的步骤之后,包括:
    若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;
    判断所述当前访问人数是否大于预设的人数阈值;
    若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
  10. 根据权利要求8所述的计算机设备,其中,所述利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字的步骤,包括:
    利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;
    所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
    利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
  11. 根据权利要求8所述的计算机设备,其中,所述调用申请携带有用户唯一标识,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤之前,包括:
    利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文字标记有阅读标识;
    通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
  12. 根据权利要求8所述的计算机设备,其中,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
    采用公式:
    Figure PCTCN2020088041-appb-100002
    计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
  13. 根据权利要求8所述的计算机设备,其中,所述判断所述相似程度值是否大于预设的相似程度阈值的步骤之后,包括:
    若所述相似程度值不大于预设的相似程度阈值,则将所述指定识别文字发送给所述客户端;
    根据预设的时间阈值调整规则,减小所述时间阈值。
  14. 根据权利要求8所述的计算机设备,其中,所述若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值的步骤,包括:
    统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;
    计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值调整参数大于0;
    在所述时间阈值的基础上增加所述指定时间阈值调整参数。
  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:
    获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值;
    若所述时间间隔大于预设的时间阈值,则允许所述客户端调用所述指定接口,并接收所述客户端通过所述指定接口发送的指定文字图片;
    利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字;
    利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;
    判断所述相似程度值是否大于预设的相似程度阈值;
    若所述相似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值。
  16. 根据权利要求15所述的计算机可读存储介质,其中,所述获取客户端发送的对指定接口的调用申请,并判断所述服务器的当前时间与所述客户端最近一次调用所述指定接口的时间点的时间间隔是否大于预设的时间阈值的步骤之后,包括:
    若所述时间间隔不大于预设的时间阈值,则获取所述服务器的当前访问人数;
    判断所述当前访问人数是否大于预设的人数阈值;
    若所述当前访问人数大于预设的人数阈值,则拒绝所述调用申请。
  17. 根据权利要求15所述的计算机可读存储介质,其中,所述利用预设的图片文字识别技术识别所述指定文字图片,得到指定识别文字的步骤,包括:
    利用预设的图片文字识别技术识别所述指定文字图片中的预定区域,得到指定识别文字,其中预定区域的面积小于所述指定文字图片的面积;
    所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
    利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字中的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片中的预定区域的识别结果。
  18. 根据权利要求15所述的计算机可读存储介质,其中,所述调用申请携带有用户唯一标识,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤之前,包括:
    利用所述用户唯一标识,根据预设的用户唯一标识与识别文本的对应关系,确定所述客户端对应的识别文本,其中所述识别文本的最近一次添加的识别文字标记有阅读标识;
    通过在所述客户端对应的识别文本中查询所述阅读标识,从而得到所述前次识别文字。
  19. 根据权利要求15所述的计算机可读存储介质,其中,所述利用预设的相似度计算方法,计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果的步骤,包括:
    采用公式:
    Figure PCTCN2020088041-appb-100003
    计算所述指定识别文字与预存的前次识别文字的相似程度值,所述前次识别文字是所述服务器对所述客户端最近一次发送的待识别文字图片的识别结果;其中similarity为相似度,A为所述指定识别文字的词频向量,B为所述前次识别文字的词频向量,Ai为所述指定识别文字的第i个单词出现的次数,Bi为所述前次识别文字的第i个单词出现的次数。
  20. 根据权利要求15所述的计算机可读存储介质,其中,所述判断所述相似程度值是否大于预设的相似程度阈值的步骤之后,包括:
    若所述相似程度值不大于预设的相似程度阈值,则将所述指定识别文字发送给所述客户端;
    根据预设的时间阈值调整规则,减小所述时间阈值。
  21. 根据权利要求15所述的计算机可读存储介质,其中,所述若所述相 似程度值大于预设的相似程度阈值,则向所述客户端发送无需更新识别文字的信息,并根据预设的时间阈值调整规则,增大所述时间阈值的步骤,包括:
    统计在指定时间内,所述客户端调用所述指定接口的总次数与成功次数;
    计算出所述成功次数/所述总次数的商值,并根据预设的商值与时间阈值调整参数的对应关系,获得与所述成功次数/所述总次数的商值对应的指定时间阈值调整参数,其中所述指定时间阈值调整参数大于0;
    在所述时间阈值的基础上增加所述指定时间阈值调整参数。
PCT/CN2020/088041 2019-05-30 2020-04-30 动态调整时间阈值的方法、装置、计算机设备和存储介质 WO2020238555A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910463109.1 2019-05-30
CN201910463109.1A CN110351332B (zh) 2019-05-30 2019-05-30 动态调整时间阈值的方法、装置、计算机设备和存储介质

Publications (1)

Publication Number Publication Date
WO2020238555A1 true WO2020238555A1 (zh) 2020-12-03

Family

ID=68174512

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/088041 WO2020238555A1 (zh) 2019-05-30 2020-04-30 动态调整时间阈值的方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN110351332B (zh)
WO (1) WO2020238555A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112613020A (zh) * 2020-12-31 2021-04-06 中国农业银行股份有限公司 一种身份验证方法及装置
CN114359025A (zh) * 2022-01-10 2022-04-15 成都智元汇信息技术股份有限公司 一种图片分流调度到中心的方法、边缘识图盒子及系统

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110351332B (zh) * 2019-05-30 2022-04-15 深圳壹账通智能科技有限公司 动态调整时间阈值的方法、装置、计算机设备和存储介质
CN111245913A (zh) * 2020-01-06 2020-06-05 深圳壹账通智能科技有限公司 基于网关的报文发送方法、装置、计算机设备和存储介质
CN111367520B (zh) * 2020-02-14 2024-05-14 中国平安人寿保险股份有限公司 基于半集成框架的组件渲染方法、装置和计算机设备
CN111565321B (zh) * 2020-04-28 2022-03-15 聚好看科技股份有限公司 一种用于录屏的终端设备、服务器及方法
CN111832547A (zh) * 2020-06-24 2020-10-27 平安普惠企业管理有限公司 文字识别模型的动态部署方法、装置和计算机设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9118722B1 (en) * 2008-07-29 2015-08-25 Amazon Technologies, Inc. Peak traffic management
CN106557768A (zh) * 2016-11-25 2017-04-05 北京小米移动软件有限公司 对图片中的文字进行识别的方法及装置
CN106657345A (zh) * 2016-12-28 2017-05-10 深圳天珑无线科技有限公司 服务器访问方法、装置及终端
CN107017978A (zh) * 2016-01-28 2017-08-04 阿里巴巴集团控股有限公司 终端设备间的信息同步方法及装置
CN109743295A (zh) * 2018-12-13 2019-05-10 平安科技(深圳)有限公司 访问阈值调整方法、装置、计算机设备及存储介质
CN110351332A (zh) * 2019-05-30 2019-10-18 深圳壹账通智能科技有限公司 动态调整时间阈值的方法、装置、计算机设备和存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108550057B (zh) * 2018-04-18 2023-04-18 平安科技(深圳)有限公司 坐席答疑请求处理方法、电子装置、计算机可读存储介质
CN109146517A (zh) * 2018-08-02 2019-01-04 惠龙 一种基于在线字典学习数据匹配模型的假货数据识别方法
CN109614510B (zh) * 2018-11-23 2021-05-07 腾讯科技(深圳)有限公司 一种图像检索方法、装置、图形处理器和存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9118722B1 (en) * 2008-07-29 2015-08-25 Amazon Technologies, Inc. Peak traffic management
CN107017978A (zh) * 2016-01-28 2017-08-04 阿里巴巴集团控股有限公司 终端设备间的信息同步方法及装置
CN106557768A (zh) * 2016-11-25 2017-04-05 北京小米移动软件有限公司 对图片中的文字进行识别的方法及装置
CN106657345A (zh) * 2016-12-28 2017-05-10 深圳天珑无线科技有限公司 服务器访问方法、装置及终端
CN109743295A (zh) * 2018-12-13 2019-05-10 平安科技(深圳)有限公司 访问阈值调整方法、装置、计算机设备及存储介质
CN110351332A (zh) * 2019-05-30 2019-10-18 深圳壹账通智能科技有限公司 动态调整时间阈值的方法、装置、计算机设备和存储介质

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112613020A (zh) * 2020-12-31 2021-04-06 中国农业银行股份有限公司 一种身份验证方法及装置
CN112613020B (zh) * 2020-12-31 2024-05-28 中国农业银行股份有限公司 一种身份验证方法及装置
CN114359025A (zh) * 2022-01-10 2022-04-15 成都智元汇信息技术股份有限公司 一种图片分流调度到中心的方法、边缘识图盒子及系统
CN114359025B (zh) * 2022-01-10 2023-05-02 成都智元汇信息技术股份有限公司 一种图片分流调度到中心的方法、边缘识图盒子及系统

Also Published As

Publication number Publication date
CN110351332B (zh) 2022-04-15
CN110351332A (zh) 2019-10-18

Similar Documents

Publication Publication Date Title
WO2020238555A1 (zh) 动态调整时间阈值的方法、装置、计算机设备和存储介质
CN110334688B (zh) 基于人脸照片库的图像识别方法、装置和计算机可读存储介质
US6775660B2 (en) User-oriented method and system for database query
WO2020238556A1 (zh) 基于配置平台的数据传输方法、装置和计算机设备
WO2019154353A1 (zh) 系统运行参数查询方法、匹配方法、装置及节点设备
JP5372254B2 (ja) サーバ装置、投稿情報処理方法、投稿情報処理プログラム及び記録媒体
US9253409B2 (en) Imaging processing system and method and management apparatus
WO2019128930A1 (zh) 操作的处理方法、账号信息的处理方法、装置及存储介质
US20200311451A1 (en) Method for registering fingerprint, terminal and computer-readable storage medium
US11026076B2 (en) Mobile terminal, information processing method and program
WO2017125092A1 (zh) 号码标记方法和系统
US8339316B1 (en) Smart GPS use
CN110333957B (zh) 远程过程调用rpc服务调用方法、装置和计算机设备
JP6570978B2 (ja) クラスタ選択装置
CN107277624B (zh) 时长计算方法及其装置
CN109255023B (zh) 线索信息处理方法及装置
CN110891090B (zh) 一种请求方法、装置、服务端、系统及存储介质
CN106815334A (zh) 一种用于终端的数据查询方法及装置
EP2007118B1 (en) Communication terminal, management server, change recognition system, and change recognition method
CN112351424B (zh) 无线连网管理方法、系统、设备及计算机可读存储介质
CN111405107B (zh) 呼叫控制方法、装置、设备及存储介质
US10122705B2 (en) Intelligent terminal equipment and information transmission method and system using the same
CN109246686B (zh) 基于云端技术的多卡用户通讯录更新方法及装置
CN114095936A (zh) 短信验证码请求方法、攻击防御方法、装置、介质及设备
CN111555973A (zh) 一种基于5g数据转发平面的数据包转发方法及装置

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: 20813423

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: 20813423

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 18/03/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20813423

Country of ref document: EP

Kind code of ref document: A1