CN111178199B - Interactive question-answering method and terminal based on real-time image recognition - Google Patents
Interactive question-answering method and terminal based on real-time image recognition Download PDFInfo
- Publication number
- CN111178199B CN111178199B CN201911321703.3A CN201911321703A CN111178199B CN 111178199 B CN111178199 B CN 111178199B CN 201911321703 A CN201911321703 A CN 201911321703A CN 111178199 B CN111178199 B CN 111178199B
- Authority
- CN
- China
- Prior art keywords
- answer
- questions
- question
- image
- result
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/418—Document matching, e.g. of document images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The invention discloses an interactive question-answering method and terminal based on real-time image recognition, which are used for creating test questions, wherein the test questions comprise questions and corresponding answer ranges which can be input; question according to the test questions; the answers including the images, which are made according to the questions and the corresponding answer ranges capable of being input, are obtained in real time, the content of the images are synchronously identified, the answers are evaluated according to the identification results, simple and accurate interactive questions and answers are achieved, the interference of irrelevant result items on the judgment results is avoided by limiting the answer ranges capable of being input by the test questions, the accuracy of the interactive questions and answers is improved, the image answers made according to the test questions are obtained and identified in real time by utilizing a real-time image identification technology, the problem that a user can finish the questions and answers under the condition that the terminal cannot be operated is effectively solved, the interactive questions and answers are simpler in process, and user experience is improved.
Description
Technical Field
The invention relates to the field of image recognition, in particular to an interactive question-answering method and terminal based on real-time image recognition.
Background
The conventional interactive question-answering terminal based on image recognition generally needs to input contents such as characters and voices to the terminal through an input device, in the question-answering process, the terminal recognizes a target picture by means of an image recognition technology, captures images by means of photographing, and exposes the limitation of use for users who cannot touch the device or cannot accurately express the contents by the characters.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the interactive question-answering method and the terminal based on the real-time image recognition can simply and accurately realize interactive question-answering.
In order to solve the technical problems, the invention adopts the following technical scheme:
an interactive question-answering method based on real-time image recognition comprises the following steps:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
s2, asking questions according to the test questions;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results.
In order to solve the technical problems, the invention adopts another technical scheme that:
an interactive question-answering terminal based on real-time image recognition, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
s2, asking questions according to the test questions;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results.
The invention has the beneficial effects that: creating a test question and a corresponding inputtable answer range thereof, asking the test question, acquiring answers comprising images, which are made according to the question and the corresponding inputtable answer range thereof, synchronously identifying the content of the images, and evaluating the answers according to an identification result, thereby realizing simple and accurate interactive question and answer. By limiting the answer range which can be input by the test questions, the identification of irrelevant answer items can be effectively avoided, the accuracy of answer content judgment of respondents in the question-answering process is improved, the overall effect of the questions-answering is effectively ensured, the image answers made according to the test questions are obtained and identified in real time by utilizing a real-time image identification technology, the image identification results are taken as question-answering results, the problem that respondents can finish the questions-answering under the condition that the respondents cannot operate a terminal is effectively solved, the interactive question-answering process is simpler, and the user experience is improved.
Drawings
FIG. 1 is a flow chart of steps of an interactive question-answering method based on real-time image recognition according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an interactive question-answering terminal based on real-time image recognition according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating an application of an interactive question-answering method based on real-time image recognition in interactive question-answering of children according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of an answer range selection process capable of being input by an interactive question-answering method based on real-time image recognition according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a test question creating process of an interactive question-answering method based on real-time image recognition according to an embodiment of the present invention;
FIG. 6 is a flow chart of a method for creating test questions based on an interactive question-answering method for real-time image recognition according to an embodiment of the present invention;
FIG. 7 is a block diagram of a process for generating an identification result set of an interactive question-answering method based on real-time image identification according to an embodiment of the present invention;
FIG. 8 is a block diagram showing an application relationship between an application example of an interactive question-answering method based on real-time image recognition and a model after training in accordance with an embodiment of the present invention;
FIG. 9 is a flowchart of a detection recognition result set of an interactive question-answering method based on real-time image recognition according to an embodiment of the present invention;
description of the reference numerals:
1. an interactive question-answering terminal based on real-time image recognition; 2. a memory; 3. a processor.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
Referring to fig. 1, an interactive question-answering method based on real-time image recognition, the method includes the steps of:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
s2, asking questions according to the test questions;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results.
From the above description, the beneficial effects of the invention are as follows: creating a test question and a corresponding inputtable answer range thereof, asking the test question, acquiring answers comprising images, which are made according to the question and the corresponding inputtable answer range thereof, synchronously identifying the content of the images, and evaluating the answers according to an identification result, thereby realizing simple and accurate interactive question and answer. By limiting the answer range which can be input by the test questions, the identification of irrelevant answer items can be effectively avoided, the accuracy of answer content judgment of respondents in the question-answering process is improved, the overall effect of the questions-answering is effectively ensured, the image answers made according to the test questions are obtained and identified in real time by utilizing a real-time image identification technology, the image identification results are taken as question-answering results, the problem that respondents can finish the questions-answering under the condition that the respondents cannot operate a terminal is effectively solved, the interactive question-answering process is simpler, and the user experience is improved.
Further, the step S1 includes:
receiving an answer range which can be input;
receiving a preset number of questions and answer items for each question, wherein the answer items are determined from the input answer range;
and creating test questions according to the received questions and corresponding answer items thereof and storing the test questions.
The answer range which can be input by the test questions is limited, and the answer item aiming at each question is determined in the answer range which can be input, so that the identification of irrelevant answer items can be effectively avoided, the accuracy of answer content judgment of respondents in the question-answering process is improved, and the overall effect of interactive questions-answering is effectively ensured.
Further, the identifying the content of the image in step S3, and evaluating the answer according to the identification result includes:
analyzing the identified image content to obtain an identification result set, judging whether an effective result item exists in the identification result set through preset detection logic, if so, taking the effective result item as an answer to the question, judging whether the answer is correct according to the answer corresponding to the question, and if not, not processing until the question is overtime.
According to the description, all possible recognition results are listed through the recognition result set, the problem that deviation between the result item and an actual image exists due to the fact that a single result item is recognized only according to image content is avoided, the content of the recognition result set is judged through preset detection logic, irrelevant interference items can be effectively eliminated, effective result items are screened out, the effective result items are used as answers of the questioning at the moment, accuracy of image recognition results is improved, when the result detected by the recognition result set is no effective result item, processing is not performed, and the fact that a terminal gives an incorrect judgment result due to the fact that the irrelevant result item is received is avoided is facilitated.
Further, in the step S3, the parsing the identified image content to obtain an identification result set includes:
and matching the identified image content with a preset model obtained by training a plurality of pictures by a training script, and storing the matched result item and the corresponding matching degree thereof into an identification result set.
According to the description, the recognized image content is matched with a large number of preset models, recognition results of all possible target images are analyzed, smooth question-answer real-time monitoring is achieved, errors in recognition of the image content are avoided, the matched result items and the corresponding matching degrees are stored in a recognition result set, the association between the matched result items and the images can be intuitively displayed according to the matching degrees, and the accuracy of image recognition is improved.
Further, in the step S3, the determining, by a preset detection logic, whether a valid result item exists in the recognition result set includes:
s31, judging whether undetected result items exist in the identification result set, if so, acquiring the result item with highest matching degree, executing the step S32, and if not, judging that no effective result item exists in the identification result set;
s32, judging whether the matching degree of the result item is greater than the lowest matching degree, if so, executing the step S33, and if not, judging that no effective result item exists in the identification result set;
s33, judging whether the result item is in the answer range which can be input, if so, determining that the result item is a valid result item, if not, marking the current result item as a detected result item, and returning to the step S31.
According to the description, the result items in the recognition result set are detected according to the sequence from high to low of the matching degree, the higher the matching degree is, the closer the result items are to the target image, meanwhile, the lowest matching degree is set, the result items lower than the lowest matching degree are excluded, the detection reliability of the recognition result set is guaranteed, the range of the inputtable answers of the questions and answers is given, the interference of irrelevant result items can be effectively avoided by combining the matching degree of the result items, the result items meeting the conditions are used as content judgment, and the accuracy of the interactive questions and answers is improved.
Referring to fig. 2, an interactive question-answering terminal based on real-time image recognition includes a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the following steps when executing the computer program:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
s2, asking questions according to the test questions;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results.
From the above description, the beneficial effects of the invention are as follows: creating a test question and a corresponding inputtable answer range thereof, asking the test question, acquiring answers comprising images, which are made according to the question and the corresponding inputtable answer range thereof, synchronously identifying the content of the images, and evaluating the answers according to an identification result, thereby realizing simple and accurate interactive question and answer. By limiting the answer range which can be input by the test questions, the identification of irrelevant answer items can be effectively avoided, the accuracy of answer content judgment of respondents in the question-answering process is improved, the overall effect of the questions-answering is effectively ensured, the image answers made according to the test questions are obtained and identified in real time by utilizing a real-time image identification technology, the image identification results are taken as question-answering results, the problem that respondents can finish the questions-answering under the condition that the respondents cannot operate a terminal is effectively solved, the interactive question-answering process is simpler, and the user experience is improved.
Further, the step S1 includes:
receiving an answer range which can be input;
receiving a preset number of questions and answer items for each question, wherein the answer items are determined from the input answer range;
and creating test questions according to the received questions and corresponding answer items thereof and storing the test questions.
The answer range which can be input by the test questions is limited, and the answer item aiming at each question is determined in the answer range which can be input, so that the identification of irrelevant answer items can be effectively avoided, the accuracy of answer content judgment of respondents in the question-answering process is improved, and the overall effect of interactive questions-answering is effectively ensured.
Further, the identifying the content of the image in step S3, and evaluating the answer according to the identification result includes:
analyzing the identified image content to obtain an identification result set, judging whether an effective result item exists in the identification result set through preset detection logic, if so, taking the effective result item as an answer to the question, judging whether the answer is correct according to the answer corresponding to the question, and if not, not processing until the question is overtime.
According to the description, all possible recognition results are listed through the recognition result set, the problem that deviation between the result item and an actual image exists due to the fact that a single result item is recognized only according to image content is avoided, the content of the recognition result set is judged through preset detection logic, irrelevant interference items can be effectively eliminated, effective result items are screened out, the effective result items are used as answers of the questioning at the moment, accuracy of image recognition results is improved, when the result detected by the recognition result set is no effective result item, processing is not performed, and the fact that a terminal gives an incorrect judgment result due to the fact that the irrelevant result item is received is avoided is facilitated.
Further, in the step S3, the parsing the identified image content to obtain an identification result set includes:
and matching the identified image content with a preset model obtained by training a plurality of pictures by a training script, and storing the matched result item and the corresponding matching degree thereof into an identification result set.
According to the description, the recognized image content is matched with a large number of preset models, recognition results of all possible target images are analyzed, smooth question-answer real-time monitoring is achieved, errors in recognition of the image content are avoided, the matched result items and the corresponding matching degrees are stored in a recognition result set, the association between the matched result items and the images can be intuitively displayed according to the matching degrees, and the accuracy of image recognition is improved.
Further, in the step S3, the determining, by a preset detection logic, whether a valid result item exists in the recognition result set includes:
s31, judging whether undetected result items exist in the identification result set, if so, acquiring the result item with highest matching degree, executing the step S32, and if not, judging that no effective result item exists in the identification result set;
s32, judging whether the matching degree of the result item is greater than the lowest matching degree, if so, executing the step S33, and if not, judging that no effective result item exists in the identification result set;
s33, judging whether the result item is in the answer range which can be input, if so, determining that the result item is a valid result item, if not, marking the current result item as a detected result item, and returning to the step S31.
According to the description, the result items in the recognition result set are detected according to the sequence from high to low of the matching degree, the higher the matching degree is, the closer the result items are to the target image, meanwhile, the lowest matching degree is set, the result items lower than the lowest matching degree are excluded, the detection reliability of the recognition result set is guaranteed, the range of the inputtable answers of the questions and answers is given, the interference of irrelevant result items can be effectively avoided by combining the matching degree of the result items, the result items meeting the conditions are used as content judgment, and the accuracy of the interactive questions and answers is improved.
Example 1
Referring to fig. 1, an interactive question-answering method based on real-time image recognition specifically includes the steps of:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
specifically, the step S1 includes:
receiving an answer range which can be input;
receiving a preset number of questions and answer items for each question, wherein the answer items are determined from the input answer range;
and creating test questions according to the received questions and corresponding answer items thereof and storing the test questions.
Specifically, the interactive question-answering method based on real-time image recognition can be applied to a mobile terminal, a specific application scenario is shown in an application schematic diagram of the interactive question-answering method based on real-time image recognition in fig. 3 in an interactive question-answering of children, in the question-answering terminal, parents check answer items corresponding to all preset test questions in the test at the time as an inputtable answer range, for example, the inputtable answer range of the test is set to be banana, corn, bottle, apple or panda, as shown in an inputtable answer range selection process schematic diagram in fig. 4, a real object, model or card corresponding to the inputtable answer range is prepared, parents create test questions, for example, which is corn? Recording the corresponding voice of the question or editing the text of the question, picking up the answer corn corresponding to the test question in the input answer range, and storing the test question, wherein the process can refer to a schematic diagram of the test question creation process in FIG. 5, and FIG. 6 is a flow chart for creating the test question;
s2, asking questions according to the test questions;
specifically, playing pre-recorded problem voice or directly displaying the problem content on the terminal in the form of characters;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results.
Specifically, before the articles related to corn are lifted by children and put on a camera, an image capturing module captures images in real time, the content of the images is identified through an image identification module, and the answer is judged according to the identification result;
specifically, the identifying the content of the image in step S3, and evaluating the answer according to the identification result includes:
analyzing the identified image content to obtain an identification result set, judging whether an effective result item exists in the identification result set through preset detection logic, if so, taking the effective result item as an answer to the question, judging whether the answer is correct according to the answer corresponding to the question, and if not, not processing until the question is overtime.
Specifically, the image recognition module is utilized to analyze the acquired image content to obtain a recognition result set { [ corn, 98% ], [ banana, 43% ], [ lemon, 25% ] }, whether an effective result item exists in the recognition result set or not is judged through a detection logic preset in the input judgment module, the effective result item [ corn, 98% ] exists in the recognition result set after detection, the effective result item is used as an answer of the question, the answer is judged according to an answer item preset when the test question is created, and because the answer item preset by the test question is corn, the answer is judged to be correct at the time, the next question is entered, and if the corn is not placed in front of a camera, other recognition results are ignored until the question is overtime;
specifically, the parsing in step S3 to obtain the recognition result set according to the recognized image content includes:
and matching the identified image content with a preset model obtained by training a plurality of pictures by a training script, and storing the matched result item and the corresponding matching degree thereof into an identification result set.
Specifically, a machine learning framework is built, a label sample formed by a large number of pictures is trained through a preset training script based on the framework to obtain a model related to the pictures, image content obtained in real time is matched with the model through an image recognition module, a result item with matching degree is obtained according to the superposition condition of characteristic parameters of the image content and the model, the result item is stored in a recognition result set to obtain { [ corn, 98% ], [ banana, 43% ], [ lemon, 25% ] }, fig. 7 is a block diagram of a recognition result set generation process, and fig. 8 is an application relation block diagram of an application example of an interactive answer method based on real-time image recognition and a model after training is completed;
specifically, in step S3, the determining, by a preset detection logic, whether a valid result item exists in the recognition result set includes:
s31, judging whether undetected result items exist in the identification result set, if so, acquiring the result item with highest matching degree, executing the step S32, and if not, judging that no effective result item exists in the identification result set;
specifically, inputting a recognition result set { [ corn, 98% ], [ banana, 43% ], [ lemon, 25% ] }, judging whether undetected result items exist in the recognition result set, assuming that none of the result items in the recognition result set is detected, obtaining a result item [ corn, 98% ] with highest matching degree, and executing step S32;
s32, judging whether the matching degree of the result items is larger than the lowest matching degree, if so, executing the step S33, and if not, judging that no effective result items exist in the identification junction set;
specifically, assuming that the lowest matching degree is 80%,98% >80%, the result item [ corn, 98% ] satisfies the condition, step S33 is performed;
s33, judging whether the result item is in the answer range which can be input, if so, determining that the result item is a valid result item, if not, marking the current result item as a detected result item, and returning to the step S31.
Specifically, whether the result item [ corn, 98% ] is within a preset inputtable answer range is judged, and the result item [ corn, 98% ] is judged to be an effective result item because the preset inputtable answer range is banana, corn, bottle, apple and panda, and the result item [ corn, 98% ] is within the inputtable answer range, and fig. 9 is a flow chart of a detection and identification result set;
in another preferred embodiment, the statistical result is stored after the statistics are performed on the test questions, and as shown in fig. 3, learning content may be recommended for the stored statistical result, and a voice question may be performed according to the recommended learning content.
Example two
Referring to fig. 2, an interactive question-answering terminal 1 based on real-time image recognition, the terminal 1 includes a memory 2, a processor 3 and a computer program stored in the memory 2 and executable on the processor 3, wherein the processor 3 implements the steps in the first embodiment when executing the computer program.
In summary, the interactive question-answering method and terminal based on real-time image recognition provided by the invention can effectively avoid wrong judging results given by the terminal due to the interference of irrelevant result items on judgment by defining the answer range which can be input by the test questions and determining answer items aiming at each question in the answer range which can be input, improve the accuracy of answer content judgment of respondents in the question-answering process, ensure the overall effect of questions and answers, question the test questions in a voice mode, and simultaneously obtain image contents in real time by means of an image recognition technology, take image recognition results as question-answering results, effectively solve the problem that a user can finish questions and answers under the condition that the user cannot touch the terminal, enable the interactive question-answering process to be simpler, improve user experience, match the recognized image contents with a preset large number of models, obtain result items with matching degree, improve the accuracy of image recognition, and screen the answer items with the matching degree and the lowest rejection degree according to the given answer range which can be input by carrying out logic detection on the recognition result sets, and the answer items with the relevance degree of the irrelevant result items can be screened as the answer items accurately.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (6)
1. An interactive question-answering method based on real-time image recognition is characterized by comprising the following steps:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
s2, asking questions according to the test questions;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results;
the step S3 of obtaining, in real time, an answer including an image according to the question and the corresponding answer range capable of being input, and synchronously identifying the content of the image includes:
before a user lifts an article and puts the article on a camera, an image capturing module captures an image in real time, and the content of the image is identified through an image identification module;
the step S3 of identifying the content of the image, and evaluating the answer according to the identification result includes:
analyzing the identified image content to obtain an identification result set, judging whether an effective result item exists in the identification result set through preset detection logic, if so, taking the effective result item as an answer to the question, judging whether the answer is correct according to the answer corresponding to the question, and if not, not processing until the question is overtime;
in the step S3, the determining, by a preset detection logic, whether a valid result item exists in the recognition result set includes:
s31, judging whether undetected result items exist in the identification result set, if so, acquiring the result item with highest matching degree, executing the step S32, and if not, judging that no effective result item exists in the identification result set;
s32, judging whether the matching degree of the result item is greater than the lowest matching degree, if so, executing the step S33, and if not, judging that no effective result item exists in the identification result set;
s33, judging whether the result item is in the answer range which can be input, if so, determining that the result item is a valid result item, if not, marking the current result item as a detected result item, and returning to the step S31.
2. The interactive question-answering method based on real-time image recognition according to claim 1, wherein the step S1 comprises:
receiving an answer range which can be input;
receiving a preset number of questions and answer items for each question, wherein the answer items are determined from the input answer range;
and creating test questions according to the received questions and corresponding answer items thereof and storing the test questions.
3. The interactive question-answering method according to claim 1, wherein the parsing the identified image content in step S3 to obtain an identification result set includes:
and matching the identified image content with a preset model obtained by training a plurality of pictures by a training script, and storing the matched result item and the corresponding matching degree thereof into an identification result set.
4. An interactive question-answering terminal based on real-time image recognition, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the following steps when executing the computer program:
s1, creating test questions, wherein the test questions comprise questions and corresponding answer ranges capable of being input;
s2, asking questions according to the test questions;
s3, acquiring answers comprising images, which are made according to the questions and the corresponding answer ranges capable of being input, in real time, synchronously identifying the content of the images, and evaluating the answers according to the identification results;
the step S3 of obtaining, in real time, an answer including an image according to the question and the corresponding answer range capable of being input, and synchronously identifying the content of the image includes:
before a user lifts an article and puts the article on a camera, an image capturing module captures an image in real time, and the content of the image is identified through an image identification module;
the step S3 of identifying the content of the image, and evaluating the answer according to the identification result includes:
analyzing the identified image content to obtain an identification result set, judging whether an effective result item exists in the identification result set through preset detection logic, if so, taking the effective result item as an answer to the question, judging whether the answer is correct according to the answer corresponding to the question, and if not, not processing until the question is overtime;
in the step S3, the determining, by a preset detection logic, whether a valid result item exists in the recognition result set includes:
s31, judging whether undetected result items exist in the identification result set, if so, acquiring the result item with highest matching degree, executing the step S32, and if not, judging that no effective result item exists in the identification result set;
s32, judging whether the matching degree of the result item is greater than the lowest matching degree, if so, executing the step S33, and if not, judging that no effective result item exists in the identification result set;
s33, judging whether the result item is in the answer range which can be input, if so, determining that the result item is a valid result item, if not, marking the current result item as a detected result item, and returning to the step S31.
5. An interactive question-answering terminal based on real-time image recognition according to claim 4, wherein step S1 comprises:
receiving an answer range which can be input;
receiving a preset number of questions and answer items for each question, wherein the answer items are determined from the input answer range;
and creating test questions according to the received questions and corresponding answer items thereof and storing the test questions.
6. An interactive question-answering terminal based on real-time image recognition according to claim 4, wherein the parsing in step S3 according to the recognized image content to obtain a recognition result set includes:
and matching the identified image content with a preset model obtained by training a plurality of pictures by a training script, and storing the matched result item and the corresponding matching degree thereof into an identification result set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911321703.3A CN111178199B (en) | 2019-12-20 | 2019-12-20 | Interactive question-answering method and terminal based on real-time image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911321703.3A CN111178199B (en) | 2019-12-20 | 2019-12-20 | Interactive question-answering method and terminal based on real-time image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111178199A CN111178199A (en) | 2020-05-19 |
| CN111178199B true CN111178199B (en) | 2023-04-28 |
Family
ID=70657609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911321703.3A Active CN111178199B (en) | 2019-12-20 | 2019-12-20 | Interactive question-answering method and terminal based on real-time image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111178199B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108040239A (en) * | 2017-12-06 | 2018-05-15 | 重庆欧派信息科技有限责任公司 | Knowledge training system and method based on image recognition |
| CN108536299A (en) * | 2018-04-12 | 2018-09-14 | 广州多益网络股份有限公司 | Human-computer interaction result generation method based on matrix and system |
| CN109448453A (en) * | 2018-10-23 | 2019-03-08 | 北京快乐认知科技有限公司 | Point based on image recognition tracer technique reads answering method and system |
| CN109670065A (en) * | 2018-09-25 | 2019-04-23 | 平安科技（深圳）有限公司 | Question and answer processing method, device, equipment and storage medium based on image recognition |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2554435B (en) * | 2016-09-27 | 2019-10-23 | Univ Leicester | Image processing |
-
2019
- 2019-12-20 CN CN201911321703.3A patent/CN111178199B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108040239A (en) * | 2017-12-06 | 2018-05-15 | 重庆欧派信息科技有限责任公司 | Knowledge training system and method based on image recognition |
| CN108536299A (en) * | 2018-04-12 | 2018-09-14 | 广州多益网络股份有限公司 | Human-computer interaction result generation method based on matrix and system |
| CN109670065A (en) * | 2018-09-25 | 2019-04-23 | 平安科技（深圳）有限公司 | Question and answer processing method, device, equipment and storage medium based on image recognition |
| CN109448453A (en) * | 2018-10-23 | 2019-03-08 | 北京快乐认知科技有限公司 | Point based on image recognition tracer technique reads answering method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111178199A (en) | 2020-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109271401B (en) | Topic searching and correcting method and device, electronic equipment and storage medium | |
| CN111144191B (en) | Font identification method, font identification device, electronic equipment and storage medium | |
| CN108304793B (en) | Online learning analysis system and method | |
| CN110427859A (en) | A kind of method for detecting human face, device, electronic equipment and storage medium | |
| CN109637286A (en) | A kind of Oral Training method and private tutor's equipment based on image recognition | |
| CN110222678B (en) | Topic analysis method, system, readable storage medium and electronic device | |
| CN111339801B (en) | Personnel attention detection method, device, equipment and system | |
| US20220207266A1 (en) | Methods, devices, electronic apparatuses and storage media of image processing | |
| CN111368808A (en) | Method, device and system for acquiring answer data and teaching equipment | |
| CN109446891A (en) | A kind of interactive learning methods based on image recognition, electronic equipment and storage medium | |
| CN111814587A (en) | Human behavior detection method, teacher behavior detection method, and related system and device | |
| CN108109453A (en) | Work correction device | |
| CN114708525A (en) | Deep learning-based student classroom behavior identification method and system | |
| CN111768170A (en) | Method and device for displaying operation correction result | |
| CN113705510A (en) | Target identification tracking method, device, equipment and storage medium | |
| CN110490173B (en) | Intelligent action scoring system based on 3D somatosensory model | |
| CN110248235B (en) | Software teaching method, device, terminal equipment and medium | |
| CN111814733A (en) | Concentration degree detection method and device based on head posture | |
| CN111353439A (en) | Method, device, system and equipment for analyzing teaching behaviors | |
| CN111046852A (en) | Personal learning path generation method, device and readable storage medium | |
| KR20120075587A (en) | Studying system and method using a recognition of handwriting mathematical expression | |
| Seneviratne et al. | Student and lecturer performance enhancement system using artificial intelligence | |
| CN111178199B (en) | Interactive question-answering method and terminal based on real-time image recognition | |
| CN111639630B (en) | Operation modifying method and device | |
| US5835631A (en) | Method and apparatus for evaluating an individual using character recognition processing of input handwritten characters |
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 |