CN116308012B - Method, system and equipment based on 5G intelligent paper reading and paper tracking - Google Patents

Method, system and equipment based on 5G intelligent paper reading and paper tracking Download PDF

Info

Publication number
CN116308012B
CN116308012B CN202310603251.8A CN202310603251A CN116308012B CN 116308012 B CN116308012 B CN 116308012B CN 202310603251 A CN202310603251 A CN 202310603251A CN 116308012 B CN116308012 B CN 116308012B
Authority
CN
China
Prior art keywords
result
paper
test paper
reply
review
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310603251.8A
Other languages
Chinese (zh)
Other versions
CN116308012A (en
Inventor
袁亚兴
贺媛婧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Cdb Online Education Technology Co ltd
Original Assignee
Cdb Online Education Technology Co ltd
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 Cdb Online Education Technology Co ltd filed Critical Cdb Online Education Technology Co ltd
Priority to CN202310603251.8A priority Critical patent/CN116308012B/en
Publication of CN116308012A publication Critical patent/CN116308012A/en
Application granted granted Critical
Publication of CN116308012B publication Critical patent/CN116308012B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0833—Tracking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0832—Special goods or special handling procedures, e.g. handling of hazardous or fragile goods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/20—Education
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00—Economic sectors
    • G16Y10/40—Transportation
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00—Economic sectors
    • G16Y10/55—Education
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Operations Research (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Quality & Reliability (AREA)
  • Educational Technology (AREA)
  • Accounting & Taxation (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a method, a system and equipment for intelligent examination paper and examination paper tracking based on 5G, which belong to the technical field of Internet of things, and the method comprises the following steps: step 1: dynamically tracking the target test paper; step 2: if the target test paper is qualified, the target test paper is transmitted to an on-line review platform built by a 5G network and is combined with an artificial intelligence technology, the target test paper is intelligently reviewed, and the fidelity review result and the to-be-confirmed review result of the same target test paper are obtained, and step 3: analyzing the reliable reply content and the unreliable reply content in each to-be-confirmed review result, and acquiring clear review results according to analysis indexes of the unreliable reply content; step 4: and obtaining a final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper, and outputting and displaying the final evaluation result. The intelligent paper reading is realized, the paper reading efficiency is improved, and the reliability of the intelligent paper reading result and the accuracy of the intelligent paper reading are further ensured.

Description

Method, system and equipment based on 5G intelligent paper reading and paper tracking
Technical Field
The invention relates to the technical field of the Internet of things, in particular to a method, a system and equipment based on 5G intelligent paper reading and paper tracking.
Background
At present, in the traditional history paper reading process, the paper is read purely by manpower, but huge manpower and material resources are undoubtedly spent, then intelligent paper reading and manual combination are started, namely the distinction between a question answering card and a subjective paper is realized, the subjective paper is read on a computer by manpower to obtain a scoring result, and in the process, the situations of manpower waste and unreasonable paper reading task allocation still exist.
Therefore, the invention provides a method, a system and equipment based on 5G intelligent paper reading and paper tracking.
Disclosure of Invention
The invention provides a method, a system and equipment for intelligent examination paper based on 5G and examination paper tracking, which are used for realizing intelligent examination paper reading, improving examination paper reading efficiency and further guaranteeing reliability of intelligent examination paper reading and accuracy of intelligent examination paper by intelligently examining examination paper and distinguishing reliable and unreliable answer contents of examination paper reading results.
The invention provides a method for intelligent examination paper and examination paper tracking based on 5G, which comprises the following steps:
step 1: dynamically tracking the target test paper;
step 2: if the target test paper is qualified, transmitting the target test paper to an online review platform built by a 5G network and combining with an artificial intelligence technology, and intelligently reviewing the target test paper to obtain a fidelity review result and a to-be-confirmed review result of the same target test paper, wherein each examination question in each target test paper corresponds to a question file, and the question file comprises examination questions and reply contents;
Step 3: analyzing reliable reply content and unreliable reply content in each to-be-confirmed review result, and acquiring clear review results according to analysis indexes of the unreliable reply content;
step 4: and obtaining a final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper, and outputting and displaying the final evaluation result.
Preferably, the method for dynamically tracking the target test paper comprises the following steps:
acquiring the current position of each test paper file pocket in real time based on a positioner arranged on each test paper file pocket, and constructing a first moving track based on the test paper file pocket;
acquiring a unique identification code of each examination paper, and setting a response array for each examination paper, wherein the response array comprises: decoding information leaving different positions and code scanning information entering different positions;
determining a second moving track of each examination paper according to the response array;
simultaneously, capturing a first personnel video for driving the corresponding examination paper to leave the corresponding position and a second personnel video for driving the examination paper to enter the corresponding position, and obtaining a track auxiliary video;
and judging whether the corresponding examination paper is qualified or not according to the first moving track, the second moving track and the track auxiliary video.
Preferably, determining whether the corresponding examination paper is qualified according to the first moving track, the second moving track and the track auxiliary video includes:
locking an entering position point and an exiting position point in the first moving track, and determining a tracking pair;
acquiring a movement pair in a second movement track, carrying out position matching management on the movement pair tracking pair, and determining whether the movement pair is completely matched;
if the test paper is completely matched, judging that the corresponding test paper is qualified in tracking;
if the matching is incomplete, acquiring a mismatching management combination, wherein the mismatching management combination comprises a mismatching tracking pair and a mismatching moving pair;
analyzing a first location range of the unmatched tracking pair and a second location range of the unmatched moving pair;
extracting a first auxiliary track related to a first position range and a second auxiliary track related to a second position range according to the track auxiliary video;
and determining the association relation between the first auxiliary track and the second auxiliary track, and judging that the corresponding examination paper is unqualified when the association relation is irrelevant.
Preferably, the intelligent evaluation is performed on the target test paper, and the obtaining of the fidelity evaluation result and the evaluation result to be confirmed of the same target test paper includes:
Scanning reply contents corresponding to each test paper question in the target test paper, confirming the definition of each reply character in the reply contents, and constructing a definition list based on the reply contents;
locking the position of the low-level definition of the definition list according to the definition classification, and obtaining non-definition distribution of the corresponding reply content according to the locking position;
when the unclear distribution is of a focusing unclear type, acquiring a residual range outside a focusing clear range, and carrying out clear focusing division on the residual range according to a set focusing condition;
splicing clear focusing division results corresponding to the same reply content to obtain content to be reviewed;
when the unclear distribution is of a non-focusing unclear type, controlling idle equipment to move to the maximum connection area according to the maximum connection area of the unclear area when the occupied area of the unclear distribution is smaller than a preset area, and scanning to obtain a first scanning result of the maximum connection area;
when the first scanning result still contains the unclear character, adjusting the scanning resolution of the idle equipment, scanning the largest connecting area again, and splicing the scanning result corresponding to the focusing clear range and the fuzzy result of the rest of the non-connecting area to obtain the content to be reviewed;
When clear characters do not exist in the first scanning result, splicing the first scanning result, the scanning result corresponding to the focusing clear range and the fuzzy result of the rest non-connection area to obtain contents to be reviewed;
when the occupation area of the unclear distribution is larger than or equal to a preset area, scanning resolution of the scanning equipment in a working state is adjusted to be high, and the reply content is scanned again to obtain the content to be reviewed;
extracting key items from the contents to be reviewed according to the mapping relation between the same test paper questions and standard answers, and obtaining corresponding fidelity review results when the definition of each extraction result meets the preset definition;
otherwise, when the definition of the extraction result of at least one key item does not meet the preset definition, obtaining the to-be-confirmed review result.
Preferably, analyzing the reliable reply content and the unreliable reply content in each to-be-confirmed review result includes:
splitting according to the definition of the key item in the to-be-confirmed review result to obtain a reliable reply item and an unreliable reply item;
establishing a standard answer step list according to the mapping relation between the same test paper questions and standard answers, and carrying out first labeling on the standard answer step list based on the reliable answer items and second labeling on the standard answer step list based on the unreliable answer items;
And determining reliable reply content and unreliable reply content according to the first labeling result and the second labeling result.
Preferably, determining reliable reply content and unreliable reply content according to the first labeling result and the second labeling result includes:
determining a single item to be analyzed and a continuous item to be analyzed based on the first labeling result and the second labeling result;
determining a left standard answer definition of a left adjacent item of the single item to be analyzed and a right standard answer definition of a right adjacent item of the single item to be analyzed;
calculating the marked adjacent correlation value of the single item to be analyzed, and carrying out first attribution on the single item to be analyzed;
,
wherein ,Z(gi-1 , g i+1 ) Representing a left answer definition g corresponding to a single item to be analyzed i-1 And right answer definition g i+1 A consistency function with the definition of the left standard answer and the definition of the right standard answer; g1 represents the step tightness coefficients corresponding to a single item to be analyzed and the left adjacent item;an entry weight representing a left adjacent entry; />An entry weight representing a right adjacent entry; g2 represents that the corresponding single item to be analyzed is adjacent to the rightStep tightness coefficients of the entries;
if the value of the consistent function is 0, the first attribution of the corresponding single item to be analyzed is unreliable reply content;
If the value of the consistent function is 1, the first attribution corresponding to the single item to be analyzed is reliable reply content;
if the value of the coincidence function is 0.5, whenWhen the calculation result of the corresponding single item to be analyzed is smaller than the preset result and the item weight of the single item to be analyzed is smaller than the preset weight, judging that the first attribution of the corresponding single item to be analyzed is reliable reply content, otherwise, judging that the first attribution of the corresponding single item to be analyzed is unreliable reply content;
performing second attribution on the continuous items to be analyzed;
when the reply results of the leftmost adjacent item and the rightmost adjacent item of the continuous items to be analyzed are correct, attributing each item in the continuous items to be analyzed as reliable reply content;
when the reply result corresponding to the leftmost adjacent item is correct and the reply result corresponding to the rightmost adjacent item is wrong, screening the item number according to the item number in the continuous items to be analyzed and the total weight of the continuous items, belonging to reliable reply content, and belonging to unreliable reply content of the rest items in the continuous items to be analyzed;
,
wherein ,qmin Representing the minimum entry weight corresponding to the continuous entry to be analyzed; sum (q) represents the accumulated sum of all the entry weights in the successive entries to be analyzed; [ ]Representing a rounding symbol; n1 represents the number of entries in the continuous entries to be analyzed;
determining reliable reply content and unreliable reply content according to the first attribution result and the second attribution result;
the first labeling results are all reliable reply contents, and the second labeling results are all unreliable reply contents or part of unreliable reply contents and part of reliable reply contents or all reliable reply contents.
Preferably, the obtaining the clear reading result according to the analysis index of the unreliable reply content includes:
extracting clear content and non-clear content in the unreliable reply content;
and according to the first review of the clear content and the second review of the clear content and the non-clear content, a clear review result is obtained.
Preferably, according to all the fidelity review results and the definitely review results of the same target test paper, obtaining the final review result and outputting and displaying the final review result, including:
according to the scoring rule frame of the same target test paper, obtaining a first scoring value of each fidelity scoring result and a second scoring value of each clear scoring result;
and accumulating all the first evaluation values and the second evaluation values to obtain a final evaluation result and outputting and displaying the final evaluation result.
The invention provides a system based on 5G intelligent paper reading and paper tracking, which comprises:
the dynamic tracking module is used for dynamically tracking the test paper of the target test paper;
the system comprises a review result acquisition module, a test paper identification module and a test paper identification module, wherein the review result acquisition module is used for transmitting a target test paper to an online review platform constructed by a 5G network and combining with an artificial intelligence technology to intelligently review the target test paper to acquire a fidelity review result and a to-be-confirmed review result of the same target test paper, each test question in each target test paper corresponds to a question file, and the question file comprises test questions and reply contents;
the analysis module is used for analyzing the reliable reply content and the unreliable reply content in each review result to be confirmed and acquiring clear review results according to the analysis indexes of the unreliable reply content;
and the output display module is used for obtaining the final evaluation result and outputting and displaying the final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper.
The present invention provides an electronic device including: a processor and a memory for storing one or more program instructions; the processor is configured to execute one or more program instructions to perform any of the methods recited in any of the claims.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of a method based on 5G intelligent paper reading and paper tracking in an embodiment of the invention;
fig. 2 is a block diagram of a system based on 5G intelligent paper reading and paper tracking in an embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
The invention provides a method for intelligent examination paper and examination paper tracking based on 5G, as shown in figure 1, comprising the following steps:
Step 1: dynamically tracking the target test paper;
step 2: if the target test paper is qualified, transmitting the target test paper to an online review platform built by a 5G network and combining with an artificial intelligence technology, and intelligently reviewing the target test paper to obtain a fidelity review result and a to-be-confirmed review result of the same target test paper, wherein each examination question in each target test paper corresponds to a question file, and the question file comprises examination questions and reply contents;
step 3: analyzing reliable reply content and unreliable reply content in each to-be-confirmed review result, and acquiring clear review results according to analysis indexes of the unreliable reply content;
step 4: and obtaining a final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper, and outputting and displaying the final evaluation result.
In the embodiment, the system design has sufficient prospective, supports the development of test paper management service and the extension and expansion of service range, and adopts a modularized design thought.
The test paper tracking comprises a test paper tracking system platform, a test paper tracking APP and a mobile smart phone, the test paper tracking system platform can remotely and visually command and manage test paper escort vehicles, the travel track of the on-site test paper escort vehicles can be uniformly managed in a video imaging mode, the visual position of the escort vehicles can be displayed on an electronic map of the system, and escort personnel can be monitored through wireless networks. The test paper tracking APP has the functions of travel track tracking, video monitoring, information input and the like. Through examination paper tracking APP, escort and vehicle are convenient for the system to establish a rapid, accurate and effective information transmission channel. The test paper tracking APP is combined with the mobile smart phone 5G mobile internet technology to upload information such as travel track and monitoring video of the escort car to the test paper tracking management system platform, so that all levels of inspection command centers can master the whole process and geographic position of test paper escort at any time, and dispatch commands can be issued rapidly through text voice communication and audio video intercommunication. The terminal equipment comprises a vehicle-mounted monitoring terminal. The test paper escort management platform realizes comprehensive tracking management of the test paper in the escort process through the vehicle-mounted monitoring equipment. The test paper escort personnel can inquire about the current test paper escort task through the management platform, and the entry and confirmation of escort information are realized. And positioning longitude and latitude of the travelling crane through the vehicle-mounted monitoring equipment, recording positioning time, and drawing a travelling crane track of the transportation task according to the position information. When the vehicle deviates from a preset line, the test paper escort personnel changes, the vehicle stops for a long time, the running state of the vehicle is abnormal (overspeed alarm, detention alarm, fatigue driving alarm) and the like, the command center is automatically alerted. The command center manager or staff can command the transportation staff to work through various communication modes such as short messages, voice and video calls.
In the embodiment, based on the high-bandwidth transmission characteristic of the 5G network, the on-line review platform is built for teachers to realize the pipelining review, the review tasks are reasonably distributed, the management of the review process is enhanced, the artificial intelligent core technologies such as image-text recognition, deep learning, natural language processing and knowledge graph are integrated, the functions of intelligent review of English and thinking-political machines, batch automatic scoring of subjective questions, review quality inspection and the like are realized, the manual review is assisted, the review pressure is relieved for teachers, the strict scoring standard is effectively improved, the online, intelligent and convenient examination review links are realized, and fairness are ensured.
In the embodiment, the paper tracking management of the full life cycle of the paper circulation is realized through 5G around the off-line paper examination, and the method comprises the paper escort process and the paper recovery process. When the test paper escort, the escort task is taken as a unit, and personnel, vehicles and test paper in the escort process are supervised in the whole course through technologies such as 5G, beidou, AI and the like. When the test paper arrives at the security room, the identification terminal performs personnel verification on the personnel entering and exiting the warehouse, and records and reports the information of the test paper entering and exiting the warehouse; and the intelligent analysis of personnel entering the security room can automatically prompt and alarm if foreign personnel invade. And the full flow data visualization of the test paper escort is realized, and an auxiliary decision support is provided for zero error of test paper safety management.
In the embodiment, a plurality of examination questions exist in the target examination paper, and each examination question has a matched file to store examination questions and reply contents, so that a basis is provided for obtaining final review results later.
In this embodiment, the guaranteed review result refers to whether the content of the target test paper, which can be determined by hundreds of percent, is correct, the review result to be confirmed refers to whether the content of the target test paper cannot be determined by hundreds of percent, for example, the answer 1, the answer 2 and the answer 3 in the answer content of the question 1 are displayed clearly on the platform, at this time, the review results of the answer 1 and the answer 2 are fidelity review results, the display results of the answer 3 on the platform are unclear, at this time, the review result of the answer 3 is to be confirmed, because in the examination process, the problem of unclean paper surface, the misjudgment of the review result due to the unclean paper surface, the unclean handwriting and the like exists, and therefore, reliable and unreliable analysis is needed to be performed on the review result to obtain the clear review result of the answer 3, so as to provide a basis for obtaining the final review result.
In this embodiment, for example, there are 3 key points in reply 3, the first key point is reliable reply content, the second key point is unreliable reply content, and the third key point is reliable reply content, which is mainly determined according to the clarity of displaying the key points on the platform, because there are multiple steps for the answer of each question, and there are multiple key points in each step.
In this embodiment, the analysis index refers to analysis of unreliable reply content to get an explicit review.
In this embodiment, the final review results are accumulated from the value of the fidelity review results and the value of the definitive review results.
The beneficial effects of the technical scheme are as follows: by intelligently evaluating the examination paper and distinguishing reliable and unreliable reply contents from the evaluation result, the intelligent examination paper is realized, the examination paper evaluation efficiency is improved, and the reliability of the intelligent evaluation result and the accuracy of the intelligent examination paper are further ensured.
The invention provides a method for dynamically tracking a target test paper based on 5G intelligent test paper and test paper tracking, which comprises the following steps:
acquiring the current position of each test paper file pocket in real time based on a positioner arranged on each test paper file pocket, and constructing a first moving track based on the test paper file pocket;
Acquiring a unique identification code of each examination paper, and setting a response array for each examination paper, wherein the response array comprises: decoding information leaving different positions and code scanning information entering different positions;
determining a second moving track of each examination paper according to the response array;
simultaneously, capturing a first personnel video for driving the corresponding examination paper to leave the corresponding position and a second personnel video for driving the examination paper to enter the corresponding position, and obtaining a track auxiliary video;
and judging whether the corresponding examination paper is qualified or not according to the first moving track, the second moving track and the track auxiliary video.
In this embodiment, the positioner performs steps on the moving track of the file pocket to obtain a first moving track, including a position of entering and exiting the examination room, a position of moving to the examination room, and the like.
In the embodiment, the unique identification code is similar to the two-dimensional code placed on the shopping article, when entering the examination room, the examination paper can be scanned in the entrance based on the two-dimensional code, and the exit of the examination paper is monitored through the monitoring instrument, so that the examination paper is prevented from being lost without any accident.
In this embodiment, the response array refers to the input response and the output response of the test paper at different locations, and there is one response array at each location.
In this embodiment, each response array has a corresponding position, so that a movement track of the test paper can be obtained according to the position, and only the movement track is formed by the positions corresponding to the response arrays.
In this embodiment, the first person video and the second person video are mainly acquired to assist the track, so as to avoid the test paper from being lost.
In this embodiment, the first moving track is a track of real-time moving position points of the file pocket, and the second moving track is a track formed by position points corresponding to the response array, where the first moving track and the second moving track can determine whether the examination paper is qualified or not.
In this embodiment, the first movement trace 1:01-02 (t 11) -03-04 (t 33) -05, second movement trace 2:021 (t 1, t2 time) -041 (t 3, t4 time), at which time it is determined that the first movement trajectory overlaps the second movement trajectory.
If the second movement trace 2:021 (t 0, t2 time) -041 (t 3, t4 time), at this time, there is a trajectory overlapping with the first movement trajectory but there is a conflict in time, so that the examination paper is effectively judged to be acceptable.
In this embodiment, the trace auxiliary video is used to effectively determine whether the conflict existing in time can be eliminated, if so, it is determined that the test paper is qualified, and if not, it is determined that the test paper is unqualified.
The beneficial effects of the technical scheme are as follows: through obtaining first moving track, second moving track and supplementary auxiliary video, whether examination paper is qualified or not can be effectively confirmed, and follow-up intelligent examination paper reading is guaranteed.
The invention provides a method for intelligent examination paper and examination paper tracking based on 5G, which judges whether the corresponding examination paper is qualified or not according to a first moving track, a second moving track and a track auxiliary video, and comprises the following steps:
locking an entering position point and an exiting position point in the first moving track, and determining a tracking pair;
acquiring a moving pair in a second moving track, carrying out position matching management on the moving pair and a tracking pair, and determining whether the moving pair and the tracking pair are completely matched;
if the test paper is completely matched, judging that the corresponding test paper is qualified in tracking;
if the matching is incomplete, acquiring a mismatching management combination, wherein the mismatching management combination comprises a mismatching tracking pair and a mismatching moving pair;
analyzing a first location range of the unmatched tracking pair and a second location range of the unmatched moving pair;
extracting a first auxiliary track related to a first position range and a second auxiliary track related to a second position range according to the track auxiliary video;
And determining the association relation between the first auxiliary track and the second auxiliary track, and judging that the corresponding examination paper is unqualified when the association relation is irrelevant.
In this embodiment, the tracking pair refers to a pair of an entrance examination room and an exit examination room, and an entrance examination room and an exit examination room, that is, the entrance and the exit of the same room can form the tracking pair.
In this embodiment, the mobile pair is determined by a response array at the same location.
In this embodiment, the tracking pair: enter examination room 01, leave examination room 02, move pair: enter examination room 01 and leave examination room 02, and at this time, the matching is completed.
Tracking pairs: enter examination room 01, leave examination room 02, move pair: entering examination room 01, leaving examination room, and at this time, the tracking pair and the moving pair are unmatched management combinations.
In this embodiment, the first position range of the unmatched tracking pair is a range of examination location positions 01 to 02 and positions 02 to 01, and the second position range of the unmatched moving pair is a range of examination location positions 01 to 021.
In this embodiment, there will be a track intersection point 01 between the first track range and the second track range, and normally there will be two track intersection points 01, where there is only one.
In this embodiment, the first auxiliary track refers to a person-to-file pocket moving track, and the second auxiliary track refers to a person-to-test paper moving track.
In this embodiment, the first auxiliary track is a track obtained based on image capturing, the first auxiliary track includes two track intersections, the second auxiliary track is a track obtained by image capturing, and the second auxiliary track includes one track intersection; at this time, the association relationship between the first auxiliary track and the second auxiliary track can be obtained, and the association relationship is considered as irrelevant, and the examination paper is considered as unqualified, and may be caused by paper omission.
The beneficial effects of the technical scheme are as follows: the auxiliary track is determined by acquiring the unmatched pairs in the moving track, whether the test paper is qualified or not is determined by judging the correlation of the auxiliary track, the reliability of test paper tracking is effectively ensured, and a reliability basis is provided for follow-up examination paper reading.
The invention provides a method based on 5G intelligent examination paper and examination paper tracking, which is used for intelligently examining the target examination paper to obtain the fidelity examination result and the examination result to be confirmed of the same target examination paper, and comprises the following steps:
scanning reply contents corresponding to each test paper question in the target test paper, confirming the definition of each reply character in the reply contents, and constructing a definition list based on the reply contents;
Locking the position of the low-level definition of the definition list according to the definition classification, and obtaining non-definition distribution of the corresponding reply content according to the locking position;
when the unclear distribution is of a focusing unclear type, acquiring a residual range outside a focusing clear range, and carrying out clear focusing division on the residual range according to a set focusing condition;
splicing clear focusing division results corresponding to the same reply content to obtain content to be reviewed;
when the unclear distribution is of a non-focusing unclear type, controlling idle equipment to move to the maximum connection area according to the maximum connection area of the unclear distribution when the occupied area of the unclear distribution is smaller than a preset area, and scanning to obtain a first scanning result of the maximum connection area;
when the first scanning result still contains the unclear character, adjusting the scanning resolution of the idle equipment, scanning the largest connecting area again, and splicing the scanning result corresponding to the focusing clear range and the fuzzy result of the rest of the non-connecting area to obtain the content to be reviewed;
when clear characters do not exist in the first scanning result, splicing the first scanning result, the scanning result corresponding to the focusing clear range and the fuzzy result of the rest non-connection area to obtain contents to be reviewed;
When the occupation area of the unclear distribution is larger than or equal to a preset area, scanning resolution of the scanning equipment in a working state is adjusted to be high, and the reply content is scanned again to obtain the content to be reviewed;
extracting key items from the contents to be reviewed according to the mapping relation between the same test paper questions and standard answers, and obtaining corresponding fidelity review results when the definition of each extraction result meets the preset definition;
otherwise, when the definition of the extraction result of at least one key item does not meet the preset definition, obtaining the to-be-confirmed review result.
In this embodiment, the test paper frame is based on the distribution position of the test paper and the question including the objective question and the subjective question, and the question attribution type can be classified into an objective question or a subjective question, and the storage file of the objective question and the storage file of the main question are different.
In this embodiment, the test paper has question contents answered by the examinee, so that the definition of each answer character can be effectively determined by scanning the answer contents, and further a definition list can be obtained.
In this embodiment, explicit characters that are intuitive and clear are considered to be clear and the like.
In this embodiment, the sharpness list is obtained on the basis of the character position of each character in the reply content, and each character position is in one-to-one correspondence with the position in the sharpness list.
In this embodiment, different characters are different in definition, and therefore, a position with a low definition level is locked to obtain a non-clear distribution.
In this embodiment, the focusing condition is set to ensure that the non-clear range is clearly displayed during the scanning process, if the scanning result is clearly displayed only in the range corresponding to the touch point manually touched on the screen as the center, and the rest is not clearly displayed, then the remaining range outside the clear range needs to be clearly focused, for example, the regions outside the clear range a are B and C, where B and C can be regarded as the remaining ranges, and the focusing conditions are set for the regions B and C, respectively, so that the regions B and C are clearly displayed.
In this embodiment, the unfocused and unclear type means that the area reached by a certain position point is not clearly displayed, but is unclear, where the occupied area is determined according to the area of the area where the occupied position is located, and since there may be a continuous unclear area or a single unclear area in the unclear area, the maximum connection area is obtained to scan again, and normally, since the final result is affected if the examinee's handwriting is blurred, but if the feature is unclear, the score is reduced due to the evaluation error, and thus, the accuracy of the evaluation is ensured as much as possible.
In this embodiment, the preset area is preset, for example, 1 square centimeter, and at this time, the existing device (other than the device in the working state) is controlled to scan the non-clear portion, so as to obtain a first scan result.
In this embodiment, the scanning resolution is adjustable, normally in a state of normal use resolution.
In this embodiment, the high adjustment resolution is to rescan the reply content to obtain the content to be reviewed, and the content to be reviewed is the reply content which is finally based on artificial intelligence technology or manually needs to be reviewed.
In this embodiment, the preset definition is preset, that is, the fidelity review result can intuitively determine whether the extraction result is correct or not, and the review error due to blurring is not required.
In this embodiment, the mapping relationship is obtained based on the examination paper questions of the examination paper frame of the target examination paper, the attribution type of the questions and the answer analysis keys matched with the questions, that is, the questions 1 correspond to the items 1 and 2, the item 1 contains the key points 11, 12 and 13, and the item 2 contains the key points 21, 22 and 23.
In this embodiment, since the reply content corresponding to each question should be completely analyzed, when there is an entry that does not satisfy the preset definition, it is the result of the review to be confirmed.
The beneficial effects of the technical scheme are as follows: the definition list is obtained by confirming the definition of the characters, and the contents to be reviewed can be effectively obtained by judging the types of the non-clear distribution and splicing the division results, wherein the reply contents are further determined by adjusting the resolution and setting the focusing conditions, and a reliability foundation is provided for the follow-up review.
The invention provides a method for tracking based on 5G intelligent examination papers and examination papers, which analyzes reliable reply content and unreliable reply content in each examination result to be confirmed, and comprises the following steps:
splitting according to the definition of the key item in the to-be-confirmed review result to obtain a reliable reply item and an unreliable reply item;
establishing a standard answer step list according to the mapping relation between the same test paper questions and standard answers, and carrying out first labeling on the standard answer step list based on the reliable answer items and second labeling on the standard answer step list based on the unreliable answer items;
and determining reliable reply content and unreliable reply content according to the first labeling result and the second labeling result.
In this embodiment, key entries 1, 2 and 3 exist in the to-be-confirmed review result, where after resolution of the key entries 1, 2 and 3, reliable reply entries 1 and 2 and unreliable reply entry 3 are obtained.
In the embodiment, a test paper frame of a target test paper is obtained to obtain test paper questions, standard answers matched with the test paper questions and answer steps of the standard answers are obtained from a preset database, and then a step list of obtaining the standard answers is established;
there are positions 1, 2, 3, 4 in the standard answer step list, with the first labeled position 1, 3, and 4 and the second labeled position 2.
In this embodiment, the reliable reply content is clear content, the unreliable reply content is unclear content, and is to obtain a corresponding position in the standard-answer-based step list.
The beneficial effects of the technical scheme are as follows: marking is carried out through the mapping relation, the determination of the positions of the reliable reply content and the unreliable reply content is preliminarily realized, and a foundation is provided for attribution of the follow-up unreliable reply content.
The invention provides a method for tracking based on 5G intelligent examination papers and examination papers, which determines reliable reply contents and unreliable reply contents according to a first labeling result and a second labeling result, and comprises the following steps:
Determining a single item to be analyzed and a continuous item to be analyzed based on the first labeling result and the second labeling result;
determining a left standard answer definition of a left adjacent item of the single item to be analyzed and a right standard answer definition of a right adjacent item of the single item to be analyzed;
calculating the marked adjacent correlation value of the single item to be analyzed, and carrying out first attribution on the single item to be analyzed;
,
wherein ,Z(gi-1 , g i+1 ) Representing a left answer definition g corresponding to a single item to be analyzed i-1 And right answer definition g i+1 A consistency function with the definition of the left standard answer and the definition of the right standard answer; g1 represents the step tightness coefficients corresponding to a single item to be analyzed and the left adjacent item;an entry weight representing a left adjacent entry; />An entry weight representing a right adjacent entry; g2 represents the step tightness coefficients corresponding to a single item to be analyzed and the right adjacent item;
if the value of the consistent function is 0, the first attribution of the corresponding single item to be analyzed is unreliable reply content;
if the value of the consistent function is 1, the first attribution corresponding to the single item to be analyzed is reliable reply content;
if the value of the coincidence function is 0.5, when When the calculation result of the corresponding single item to be analyzed is smaller than the preset result and the item weight of the single item to be analyzed is smaller than the preset weight, judging that the first attribution of the corresponding single item to be analyzed is reliable reply content, otherwise, judging that the first attribution of the corresponding single item to be analyzed is unreliable reply content;
performing second attribution on the continuous items to be analyzed;
when the reply results of the leftmost adjacent item and the rightmost adjacent item of the continuous items to be analyzed are correct, attributing each item in the continuous items to be analyzed as reliable reply content;
when the reply result corresponding to the leftmost adjacent item is correct and the reply result corresponding to the rightmost adjacent item is wrong, screening the item number according to the item number in the continuous items to be analyzed and the total weight of the continuous items, belonging to reliable reply content, and belonging to unreliable reply content of the rest items in the continuous items to be analyzed;
,
wherein ,qmin Representing the minimum entry weight corresponding to the continuous entry to be analyzed; sum (q) representsThe accumulated sum of all the item weights in the continuous items to be analyzed; []Representing a rounding symbol; n1 represents the number of entries in the continuous entries to be analyzed;
Determining reliable reply content and unreliable reply content according to the first attribution result and the second attribution result;
the first labeling results are all reliable reply contents, and the second labeling results are all unreliable reply contents or part of unreliable reply contents and part of reliable reply contents or all reliable reply contents.
In this embodiment, there are positions 1, 2, 3, and 4 in the standard answer step list, where the first label is positions 1, 3, and 4, and the second label is position 2, and at this time, the entry corresponding to position 2 is a single entry to be analyzed.
If the items 1, 2, 3, 4 and 5 exist, wherein the items 1 and 5 are clear items, the items 2, 3 and 4 are non-clear items, and the items to be analyzed continuously are the items 2, 3 and 4.
In this embodiment, the left adjacent entry of the single entry to be analyzed is entry 1, and the right adjacent entry of the single entry to be analyzed is entry 3, where the left standard answer definition and the right standard answer definition are both obtained from the standard answer step list.
In this embodiment, the item weights are the importance degrees of the corresponding steps set according to the importance degrees of the key points in the reply contents corresponding to the questions, and the more important the corresponding steps are, the greater the matched item weights are.
In this embodiment, the degree of tightness between the different steps is different during the mathematical examination, for example, step 1 and step 2 are for calculating the value 1, and step 3 is for calculating the value 2, and the tightness between step 1 and step 2 is greater than the tightness between step 2 and step 3.
The beneficial effects of the technical scheme are as follows: the final attribution content of the corresponding items to be analyzed is determined through different analysis modes by determining the single items to be analyzed and the continuous items to be analyzed, so that the visible reply content and the unreliable reply content are obtained, wherein the attribution is realized through classifying and discussing the calculation of the marked adjacent correlation value of the single items to be analyzed, and the attribution is carried out through judging whether reply results of the rightmost adjacent items and the leftmost adjacent items are correct or not, so that convenience is provided for follow-up examination.
The invention provides a method for tracking based on 5G intelligent examination papers and examination papers, which obtains clear examination results according to analysis indexes of unreliable reply contents, and comprises the following steps:
extracting clear content and non-clear content in the unreliable reply content;
and according to the first review of the clear content and the second review of the clear content and the non-clear content, a clear review result is obtained.
Preferably, according to all the fidelity review results and the definitely review results of the same target test paper, obtaining the final review result and outputting and displaying the final review result, including:
according to the scoring rule frame of the same target test paper, obtaining a first scoring value of each fidelity scoring result and a second scoring value of each clear scoring result;
and accumulating all the first evaluation values and the second evaluation values to obtain a final evaluation result and outputting and displaying the final evaluation result.
In this embodiment, the unreliable reply content includes various key points in the corresponding item, so that the definition and the non-definition of the key points are determined to obtain the review result, and the review result is obtained by reviewing the reply content with reference to the standard answer.
In this embodiment, the scoring rules framework refers to scoring rules for different topics in the test paper.
In the embodiment, the fidelity review value can effectively determine the score value, and the score of the clear review result is strived for the examinee as much as possible in the review process, so that the accuracy of the score of the examination is ensured.
The beneficial effects of the technical scheme are as follows: and the final evaluation result is obtained by extracting clear content and non-clear content and acquiring values according to the scoring rule frame, so that the final evaluation score is conveniently and intuitively known, and the evaluation efficiency is ensured.
The invention provides a system based on 5G intelligent paper reading and paper tracking, as shown in figure 2, comprising:
the dynamic tracking module is used for dynamically tracking the test paper of the target test paper;
the system comprises a review result acquisition module, a test paper identification module and a test paper identification module, wherein the review result acquisition module is used for transmitting a target test paper to an online review platform constructed by a 5G network and combining with an artificial intelligence technology to intelligently review the target test paper to acquire a fidelity review result and a to-be-confirmed review result of the same target test paper, each test question in each target test paper corresponds to a question file, and the question file comprises test questions and reply contents;
the analysis module is used for analyzing the reliable reply content and the unreliable reply content in each review result to be confirmed and acquiring clear review results according to the analysis indexes of the unreliable reply content;
and the output display module is used for obtaining the final evaluation result and outputting and displaying the final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper.
The present invention provides an electronic device including: a processor and a memory for storing one or more program instructions; the processor is configured to execute one or more program instructions to perform any of the methods recited in any of the claims.
The beneficial effects of the technical scheme are as follows: by intelligently evaluating the examination paper and distinguishing reliable and unreliable reply contents from the evaluation result, the intelligent examination paper is realized, the examination paper evaluation efficiency is improved, and the reliability of the intelligent evaluation result and the accuracy of the intelligent examination paper are further ensured.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims (9)

1. A method based on 5G intelligent paper reading and paper tracking is characterized by comprising the following steps:
step 1: dynamically tracking the target test paper;
step 2: if the target test paper is qualified, transmitting the target test paper to an online review platform built by a 5G network and combining with an artificial intelligence technology, and intelligently reviewing the target test paper to obtain a fidelity review result and a to-be-confirmed review result of the same target test paper, wherein each examination question in each target test paper corresponds to a question file, and the question file comprises examination questions and reply contents;
Step 3: analyzing reliable reply content and unreliable reply content in each to-be-confirmed review result, and acquiring clear review results according to analysis indexes of the unreliable reply content;
step 4: obtaining a final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper, and outputting and displaying the final evaluation result;
the method for intelligently evaluating the target test paper to obtain the fidelity evaluating result and the evaluating result to be confirmed of the same target test paper comprises the following steps:
scanning reply contents corresponding to each test paper question in the target test paper, confirming the definition of each reply character in the reply contents, and constructing a definition list based on the reply contents;
locking the position of the low-level definition of the definition list according to the definition classification, and obtaining non-definition distribution of the corresponding reply content according to the locking position;
when the unclear distribution is of a focusing unclear type, acquiring a residual range outside a focusing clear range, and carrying out clear focusing division on the residual range according to a set focusing condition;
splicing clear focusing division results corresponding to the same reply content to obtain content to be reviewed;
When the unclear distribution is of a non-focusing unclear type, controlling idle equipment to move to the maximum connection area according to the maximum connection area of the unclear distribution when the occupied area of the unclear distribution is smaller than a preset area, and scanning to obtain a first scanning result of the maximum connection area;
when the first scanning result still contains the unclear character, adjusting the scanning resolution of the idle equipment, scanning the largest connecting area again, and splicing the scanning result corresponding to the focusing clear range and the fuzzy result of the rest of the non-connecting area to obtain the content to be reviewed;
when clear characters do not exist in the first scanning result, splicing the first scanning result, the scanning result corresponding to the focusing clear range and the fuzzy result of the rest non-connection area to obtain contents to be reviewed;
when the occupation area of the unclear distribution is larger than or equal to a preset area, scanning resolution of the scanning equipment in a working state is adjusted to be high, and the reply content is scanned again to obtain the content to be reviewed;
extracting key items from the contents to be reviewed according to the mapping relation between the same test paper questions and standard answers, and obtaining corresponding fidelity review results when the definition of each extraction result meets the preset definition;
Otherwise, when the definition of the extraction result of at least one key item does not meet the preset definition, obtaining the to-be-confirmed review result.
2. The method for 5G intelligent paper marking and paper tracking according to claim 1, wherein dynamically tracking the target paper comprises:
acquiring the current position of each test paper file pocket in real time based on a positioner arranged on each test paper file pocket, and constructing a first moving track based on the test paper file pocket;
acquiring a unique identification code of each examination paper, and setting a response array for each examination paper, wherein the response array comprises: decoding information leaving different positions and code scanning information entering different positions;
determining a second moving track of each examination paper according to the response array;
simultaneously, capturing a first personnel video for driving the corresponding examination paper to leave the corresponding position and a second personnel video for driving the examination paper to enter the corresponding position, and obtaining a track auxiliary video;
and judging whether the corresponding examination paper is qualified or not according to the first moving track, the second moving track and the track auxiliary video.
3. The method for tracking test paper based on 5G intelligent paper according to claim 2, wherein determining whether the corresponding test paper is qualified according to the first moving track, the second moving track and the track auxiliary video comprises:
Locking an entering position point and an exiting position point in the first moving track, and determining a tracking pair;
acquiring a moving pair in a second moving track, carrying out position matching management on the moving pair and a tracking pair, and determining whether the moving pair and the tracking pair are completely matched;
if the test paper is completely matched, judging that the corresponding test paper is qualified in tracking;
if the matching is incomplete, acquiring a mismatching management combination, wherein the mismatching management combination comprises a mismatching tracking pair and a mismatching moving pair;
analyzing a first location range of the unmatched tracking pair and a second location range of the unmatched moving pair;
extracting a first auxiliary track related to a first position range and a second auxiliary track related to a second position range according to the track auxiliary video;
and determining the association relation between the first auxiliary track and the second auxiliary track, and judging that the corresponding examination paper is unqualified when the association relation is irrelevant.
4. The method for 5G intelligent paper and paper tracking according to claim 1, wherein analyzing the reliable reply content and the unreliable reply content in each to-be-confirmed review result comprises:
splitting according to the definition of the key item in the to-be-confirmed review result to obtain a reliable reply item and an unreliable reply item;
Establishing a standard answer step list according to the mapping relation between the same test paper questions and standard answers, and carrying out first labeling on the standard answer step list based on the reliable answer items and second labeling on the standard answer step list based on the unreliable answer items;
and determining reliable reply content and unreliable reply content according to the first labeling result and the second labeling result.
5. The method for 5G intelligent paper and paper tracking according to claim 4, wherein determining reliable reply content and unreliable reply content according to the first labeling result and the second labeling result comprises:
determining a single item to be analyzed and a continuous item to be analyzed based on the first labeling result and the second labeling result;
determining a left standard answer definition of a left adjacent item of the single item to be analyzed and a right standard answer definition of a right adjacent item of the single item to be analyzed;
calculating the marked adjacent correlation value of the single item to be analyzed, and carrying out first attribution on the single item to be analyzed;
;
wherein ,representing the left answer definition +. >And right answer definitionA consistency function with the definition of the left standard answer and the definition of the right standard answer; />Representing the step tightness coefficients corresponding to a single item to be analyzed and a left adjacent item; />An entry weight representing a left adjacent entry; />An entry weight representing a right adjacent entry; />Representing the step tightness coefficients corresponding to a single item to be analyzed and a right adjacent item;
if the value of the consistent function is 0, the first attribution of the corresponding first item to be analyzed is unreliable reply content;
if the consistent function takes a value of 1, the first attribution corresponding to the first item to be analyzed is reliable reply content;
if the value of the coincidence function is 0.5, whenWhen the calculation result of the first item to be analyzed is smaller than the preset result and the item weight of the first item to be analyzed is smaller than the preset weight, judging that the first attribution of the corresponding first item to be analyzed is reliable reply content, otherwise, judging that the first attribution of the corresponding first item to be analyzed is unreliable reply content;
performing second attribution on the continuous items to be analyzed;
when the reply results of the leftmost adjacent item and the rightmost adjacent item of the continuous items to be analyzed are correct, attributing each item in the continuous items to be analyzed as reliable reply content;
When the reply result corresponding to the leftmost adjacent item is correct and the reply result corresponding to the rightmost adjacent item is wrong, screening the item number according to the item number in the continuous items to be analyzed and the total weight of the continuous items, belonging to reliable reply content, and belonging to unreliable reply content of the rest items in the continuous items to be analyzed;
；
wherein ,representing the minimum entry weight corresponding to the continuous entry to be analyzed; />Representing the accumulated sum of all the item weights in the continuous items to be analyzed; />Representing a rounding symbol; />Representing the number of entries in the continuous entries to be analyzed;
determining reliable reply content and unreliable reply content according to the first attribution result and the second attribution result;
the first labeling results are all reliable reply contents, and the second labeling results are all unreliable reply contents or part of unreliable reply contents and part of reliable reply contents or all reliable reply contents.
6. The method for 5G intelligent paper and paper tracking according to claim 1, wherein obtaining the clear review result according to the analysis index of the unreliable reply content comprises:
Extracting clear content and non-clear content in the unreliable reply content;
and according to the first review of the clear content and the second review of the clear content and the non-clear content, a clear review result is obtained.
7. The method for 5G intelligent paper and paper tracking according to claim 1, wherein obtaining the final review result and outputting and displaying the final review result according to all the fidelity review results and the explicit review results of the same target paper comprises:
according to the scoring rule frame of the same target test paper, obtaining a first scoring value of each fidelity scoring result and a second scoring value of each clear scoring result;
and accumulating all the first evaluation values and the second evaluation values to obtain a final evaluation result and outputting and displaying the final evaluation result.
8. A system based on 5G intelligent paper reading and paper tracking is characterized by comprising:
the dynamic tracking module is used for dynamically tracking the test paper of the target test paper;
the system comprises a review result acquisition module, a test paper identification module and a test paper identification module, wherein the review result acquisition module is used for transmitting a target test paper to an online review platform constructed by a 5G network and combining with an artificial intelligence technology to intelligently review the target test paper to acquire a fidelity review result and a to-be-confirmed review result of the same target test paper, each test question in each target test paper corresponds to a question file, and the question file comprises test questions and reply contents;
The analysis module is used for analyzing the reliable reply content and the unreliable reply content in each review result to be confirmed and acquiring clear review results according to the analysis indexes of the unreliable reply content;
the output display module is used for obtaining a final evaluation result and outputting and displaying the final evaluation result according to all the fidelity evaluation results and the clear evaluation results of the same target test paper;
the review result acquisition module is used for:
scanning reply contents corresponding to each test paper question in the target test paper, confirming the definition of each reply character in the reply contents, and constructing a definition list based on the reply contents;
locking the position of the low-level definition of the definition list according to the definition classification, and obtaining non-definition distribution of the corresponding reply content according to the locking position;
when the unclear distribution is of a focusing unclear type, acquiring a residual range outside a focusing clear range, and carrying out clear focusing division on the residual range according to a set focusing condition;
splicing clear focusing division results corresponding to the same reply content to obtain content to be reviewed;
When the unclear distribution is of a non-focusing unclear type, controlling idle equipment to move to the maximum connection area according to the maximum connection area of the unclear distribution when the occupied area of the unclear distribution is smaller than a preset area, and scanning to obtain a first scanning result of the maximum connection area;
when the first scanning result still contains the unclear character, adjusting the scanning resolution of the idle equipment, scanning the largest connecting area again, and splicing the scanning result corresponding to the focusing clear range and the fuzzy result of the rest of the non-connecting area to obtain the content to be reviewed;
when clear characters do not exist in the first scanning result, splicing the first scanning result, the scanning result corresponding to the focusing clear range and the fuzzy result of the rest non-connection area to obtain contents to be reviewed;
when the occupation area of the unclear distribution is larger than or equal to a preset area, scanning resolution of the scanning equipment in a working state is adjusted to be high, and the reply content is scanned again to obtain the content to be reviewed;
extracting key items from the contents to be reviewed according to the mapping relation between the same test paper questions and standard answers, and obtaining corresponding fidelity review results when the definition of each extraction result meets the preset definition;
Otherwise, when the definition of the extraction result of at least one key item does not meet the preset definition, obtaining the to-be-confirmed review result.
9. An electronic device, comprising: a processor and a memory for storing one or more program instructions; the processor being configured to execute one or more program instructions for performing the method of any of claims 1-7.
CN202310603251.8A 2023-05-26 2023-05-26 Method, system and equipment based on 5G intelligent paper reading and paper tracking Active CN116308012B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310603251.8A CN116308012B (en) 2023-05-26 2023-05-26 Method, system and equipment based on 5G intelligent paper reading and paper tracking

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310603251.8A CN116308012B (en) 2023-05-26 2023-05-26 Method, system and equipment based on 5G intelligent paper reading and paper tracking

Publications (2)

Publication Number Publication Date
CN116308012A CN116308012A (en) 2023-06-23
CN116308012B true CN116308012B (en) 2023-08-04

Family

ID=86824346

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310603251.8A Active CN116308012B (en) 2023-05-26 2023-05-26 Method, system and equipment based on 5G intelligent paper reading and paper tracking

Country Status (1)

Country Link
CN (1) CN116308012B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101334814A (en) * 2008-04-28 2008-12-31 华北电力大学 Automatic scanning and reading system and reading method
CN204759496U (en) * 2015-07-15 2015-11-11 江苏省教育考试院 Storehouse management system is rolled up to intelligence
KR20160061753A (en) * 2014-11-24 2016-06-01 오종현 Examination paper recognition and scoring system
WO2021180442A1 (en) * 2020-03-07 2021-09-16 Immundiagnostik Ag Calibration of a digital camera for use as a scanner
CN113505786A (en) * 2021-06-24 2021-10-15 作业帮教育科技（北京）有限公司 Test question photographing and judging method and device and electronic equipment
CN114821596A (en) * 2022-05-17 2022-07-29 北京百度网讯科技有限公司 Text recognition method and device, electronic equipment and medium
CN114915778A (en) * 2022-04-07 2022-08-16 江西盛泰精密光学有限公司 Camera testing method and system for paper marking

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101334814A (en) * 2008-04-28 2008-12-31 华北电力大学 Automatic scanning and reading system and reading method
KR20160061753A (en) * 2014-11-24 2016-06-01 오종현 Examination paper recognition and scoring system
CN204759496U (en) * 2015-07-15 2015-11-11 江苏省教育考试院 Storehouse management system is rolled up to intelligence
WO2021180442A1 (en) * 2020-03-07 2021-09-16 Immundiagnostik Ag Calibration of a digital camera for use as a scanner
CN113505786A (en) * 2021-06-24 2021-10-15 作业帮教育科技（北京）有限公司 Test question photographing and judging method and device and electronic equipment
CN114915778A (en) * 2022-04-07 2022-08-16 江西盛泰精密光学有限公司 Camera testing method and system for paper marking
CN114821596A (en) * 2022-05-17 2022-07-29 北京百度网讯科技有限公司 Text recognition method and device, electronic equipment and medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
河北省全国计算机等级考试网上评卷的实践与思考;尹建;;河南科技(16);全文 *

Also Published As

Publication number Publication date
CN116308012A (en) 2023-06-23

Similar Documents

Publication Publication Date Title
CN111340457A (en) Building fire-fighting examination and acceptance platform system based on artificial intelligence and method thereof
CN113822847A (en) Image scoring method, device, equipment and storage medium based on artificial intelligence
CN114625901B (en) Multi-algorithm integration method and device
CN117726958B (en) Intelligent detection and hidden danger identification method for inspection image target of unmanned aerial vehicle of distribution line
CN113822907B (en) Image processing method and device
CN115035328A (en) Converter image increment automatic machine learning system and establishment training method thereof
CN110929032A (en) User demand processing system and method for software system
CN116308012B (en) Method, system and equipment based on 5G intelligent paper reading and paper tracking
CN115984158A (en) Defect analysis method and device, electronic equipment and computer readable storage medium
US20160019803A1 (en) System, method and computer-accessible medium for scalable testing and evaluation
CN116932523B (en) Platform for integrating and supervising third party environment detection mechanism
CN115147248B (en) Travel information consultation system and method based on big data
CN114742522B (en) Method, system, device and storage medium for automatically comparing survey design drawings
KR20220161601A (en) System for determining defect of image inspection target using deep learning model
CN114118908A (en) Electric power material acceptance judging system and method
KR102297047B1 (en) Method and system for diagnosing Workplace Learning Agility of Knowledge Workers
CN116468395B (en) Audit correction system
CN116596716B (en) Network learning management method, system, storage medium and intelligent terminal
CN115063753B (en) Safety belt wearing detection model training method and safety belt wearing detection method
CN115146243B (en) Online VR examination examinee information checking and managing system based on data analysis
CN115146964A (en) Tourism management system and method capable of realizing intelligent analysis
CN115330574A (en) Public service target crowd active discovery system of smart community
KR20220144131A (en) Method for servicing branding management based on branding solution kit
CN113128357A (en) Vehicle fault processing method and system
CN113822266A (en) Scanning identification method and device for automatic identification of question number

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant