CN110428368A - A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing - Google Patents

A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing Download PDF

Info

Publication number
CN110428368A
CN110428368A CN201910700408.2A CN201910700408A CN110428368A CN 110428368 A CN110428368 A CN 110428368A CN 201910700408 A CN201910700408 A CN 201910700408A CN 110428368 A CN110428368 A CN 110428368A
Authority
CN
China
Prior art keywords
evaluation
image
target
target image
original image
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.)
Pending
Application number
CN201910700408.2A
Other languages
Chinese (zh)
Inventor
解晓丽
张明慧
那瀚文
陈金
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.)
Beijing Kingsoft Cloud Network Technology Co Ltd
Beijing Kingsoft Cloud Technology Co Ltd
Original Assignee
Beijing Kingsoft Cloud Network Technology Co Ltd
Beijing Kingsoft Cloud 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 Beijing Kingsoft Cloud Network Technology Co Ltd, Beijing Kingsoft Cloud Technology Co Ltd filed Critical Beijing Kingsoft Cloud Network Technology Co Ltd
Priority to CN201910700408.2A priority Critical patent/CN110428368A/en
Publication of CN110428368A publication Critical patent/CN110428368A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00Geometric image transformations in the plane of the image
    • G06T3/40Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4053Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

The embodiment of the present application provides a kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing, method includes: to obtain original image and target image to be processed, evaluation request is sent to pre-set user terminal, the address of the evaluation page is carried in evaluation request, original image and target image are shown in the evaluation page, so that the user for logging in pre-set user terminal evaluates according to the content that the evaluation page is shown, it obtains what pre-set user terminal returned, indicates the evaluation result of the treatment effect of target image Processing Algorithm.Based on above-mentioned processing, evaluation efficiency can be improved, and improve the reliability of evaluation result.

Description

A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing
Technical field
This application involves technical field of image processing, more particularly to a kind of algorithm evaluation method, device, electronic equipment and Readable storage medium storing program for executing.
Background technique
Image Super-resolution processing, refers to and is handled according to image of the Image Super-resolution to low resolution, obtained The image of high-resolution.The image of processing can be picture, alternatively, being also possible to the video comprising multiple images frame.In addition, The treatment effect of Image Super-resolution can also be evaluated according to the picture quality of treated image.
In a kind of mode, original image to be processed is uploaded to server by terminal by evaluation personnel, and is manually booted The corresponding service of target image Processing Algorithm to be evaluated, in the image processing algorithm that server is selected according to evaluation personnel to original Beginning image procossing is completed, and after generating treated image (being properly termed as target image), evaluation personnel can will be under target image It is loaded onto terminal, then, evaluation personnel can observe the picture quality for comparing original image and target image, obtain for target figure As the evaluation result of Processing Algorithm.
As it can be seen that in the related technology, evaluation personnel it is cumbersome, evaluation efficiency is lower, and is only observed by the evaluation personnel Compare original image and target image, the reliability of obtained evaluation result is lower.
Summary of the invention
The embodiment of the present application is designed to provide a kind of algorithm evaluation method, device, electronic equipment and readable storage medium Matter can be improved evaluation efficiency, and improve the reliability of evaluation result.Specific technical solution is as follows:
In a first aspect, in order to achieve the above object, the embodiment of the present application discloses a kind of algorithm evaluation method, the method Include:
Obtain original image and target image to be processed, wherein the target image is to handle to calculate according to target image The image that method obtains after handling the original image;
Evaluation request is sent to pre-set user terminal, the address of the evaluation page, institute's commentary are carried in the evaluation request The original image and the target image are shown in the valence page, so as to log in the user of the pre-set user terminal according to institute The content that the commentary valence page is shown is evaluated;
It obtains what the pre-set user terminal returned, indicates the evaluation knot of the treatment effect of the target image Processing Algorithm Fruit.
It is optionally, described to obtain original image and target image to be processed, comprising:
Receive evaluation instruction, wherein the algorithm mark of the target image Processing Algorithm is carried in the evaluation instruction;
Obtain original image to be processed, and according to the target image Processing Algorithm to the original image at Reason, obtains the target image.
Optionally, described to obtain original image to be processed, including following one:
The evaluation instruction in carry image identification in the case where, obtain be locally stored with the evaluation instruct in The corresponding image of the image identification of carrying, as the original image;
In the case where carrying image in evaluation instruction, the image carried in the evaluation instruction is obtained, as The original image.
Optionally, it is sent described to pre-set user terminal before evaluating request, the method also includes:
Receive evaluation instruction, wherein the index mark of evaluation index is also carried in the evaluation instruction;
Show that the index identifies corresponding objective appraisal index in the evaluation page, so as to log in the default use The user of family terminal comments the picture quality of the original image and the target image according to the objective appraisal index Valence.
Optionally, the target indicator includes at least one of: picture noise, image blocking artifact, image ringing effect, Edge definition and image detail richness.
Optionally, the evaluation result includes that the original image and the target image obtain on the target indicator Fractional value;
It obtains what the pre-set user terminal returned described, indicates the treatment effect of the target image Processing Algorithm After evaluation result, the method also includes:
The average value for calculating the fractional value that the original image obtains on the target indicator, as the original image Evaluation score;
The average value for calculating the fractional value that the target image obtains on the target indicator, as the target image Evaluation score;
According to the evaluation score of the evaluation score of the original image and the target image, determine at the target image The treatment effect grade of adjustment method, the treatment effect grade indicate what the target image Processing Algorithm handled image Effect.
Optionally, after the treatment effect grade of the determination target image Processing Algorithm, the method is also wrapped It includes:
Generate evaluation result report, wherein include in the appraisal report original image, the target image, Algorithm mark, the index of the target indicator of the target image Processing Algorithm identify, the original image is in the target The fractional value and target image processing that fractional value, the target image obtained in index obtains on the target indicator The treatment effect grade of algorithm.
Optionally, the original image is picture or video image.
Second aspect, in order to achieve the above object, the embodiment of the present application also disclose a kind of algorithm evaluating apparatus, the dress It sets and includes:
First obtains module, for obtaining original image and target image to be processed, wherein the target image is root The image obtained after handling according to target image Processing Algorithm the original image;
Sending module carries the evaluation page in the evaluation request for sending evaluation request to pre-set user terminal Address, the original image and the target image are shown in the evaluation page, so that it is whole to log in the pre-set user The user at end evaluates according to the content that the evaluation page is shown;
Second obtains module, for obtaining the pre-set user terminal return, indicates the target image Processing Algorithm Treatment effect evaluation result.
Optionally, described first module is obtained, is specifically used for receiving evaluation instruction, wherein carried in the evaluation instruction There is the algorithm of the target image Processing Algorithm to identify;
Obtain original image to be processed, and according to the target image Processing Algorithm to the original image at Reason, obtains the target image.
Optionally, described first module is obtained, is specifically used for executing following one:
The evaluation instruction in carry image identification in the case where, obtain be locally stored with the evaluation instruct in The corresponding image of the image identification of carrying, as the original image;
In the case where carrying image in evaluation instruction, the image carried in the evaluation instruction is obtained, as The original image.
Optionally, described device further include:
Display module, for receiving evaluation instruction, wherein the index mark of evaluation index is also carried in the evaluation instruction Know;
Show that the index identifies corresponding objective appraisal index in the evaluation page, so as to log in the default use The user of family terminal comments the picture quality of the original image and the target image according to the objective appraisal index Valence.
Optionally, the target indicator includes at least one of: picture noise, image blocking artifact, image ringing effect, Edge definition and image detail richness.
Optionally, the evaluation result includes that the original image and the target image obtain on the target indicator Fractional value;
Described device further include:
Processing module is made for calculating the average value for the fractional value that the original image obtains on the target indicator For the evaluation score of the original image;
The average value for calculating the fractional value that the target image obtains on the target indicator, as the target image Evaluation score;
According to the evaluation score of the evaluation score of the original image and the target image, determine at the target image The treatment effect grade of adjustment method, at the treatment effect grade can indicate the target image Processing Algorithm to image The effect of reason.
Optionally, the processing module is also used to generate evaluation result report, wherein include in the appraisal report The original image, the target image, the target image Processing Algorithm algorithm mark, the target indicator index mark Know, fractional value, the target image that the original image obtains on the target indicator obtain on the target indicator Fractional value and the target image Processing Algorithm treatment effect grade.
Optionally, the original image is picture or video image.
The third aspect, in order to achieve the above object, the embodiment of the present application also disclose a kind of electronic equipment, and the electronics is set Standby includes memory and processor;
The memory, for storing computer program;
The processor when for executing the program stored on memory, realizes the calculation as described in above-mentioned first aspect Method evaluation method step.
Fourth aspect, in order to achieve the above object, the embodiment of the present application also disclose a kind of computer readable storage medium, It is stored with computer program in the computer readable storage medium, when the computer program is executed by processor, realizes such as Algorithm evaluation method step described in above-mentioned first aspect.
5th aspect, in order to achieve the above object, the embodiment of the present application also discloses a kind of computer journey comprising instruction Sequence product, when run on a computer, so that computer executes algorithm evaluation method step described in above-mentioned first aspect.
The embodiment of the present application provides a kind of algorithm evaluation method, available original image to be processed and target figure Picture, wherein target image is the image obtained after being handled according to target image Processing Algorithm original image, to default use Family terminal send evaluation request, evaluation request in carry evaluation the page address, evaluate the page in show original image and Target image obtains default use so that the user for logging in pre-set user terminal evaluates according to the content that the evaluation page is shown What family terminal returned, indicate the evaluation result of the treatment effect of target image Processing Algorithm.
Based on the present processes, evaluation personnel need to only determine original image and target image Processing Algorithm to be evaluated, Electronic equipment can send evaluation request to pre-set user terminal, and obtain evaluation result, eliminate evaluation personnel from server Target image, and the step of comparing original image and target image are downloaded, can be improved evaluation efficiency, in addition, relative to existing The evaluation result that original image and target image obtain only is compared by evaluation personnel observation in technology, the embodiment of the present application can obtain The evaluation result for logging in multiple users of pre-set user terminal is taken, reduces the personal subjectivity of evaluation result to a certain extent, Improve the reliability of evaluation result.
Certainly, implement the application any product or method it is not absolutely required to and meanwhile reach all the above excellent Point.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow chart of algorithm evaluation method provided by the embodiments of the present application;
Fig. 2 is the schematic diagram that a kind of evaluation task provided by the embodiments of the present application generates interface;
Fig. 3 is a kind of schematic diagram of evaluation result report provided by the embodiments of the present application;
Fig. 4 is a kind of exemplary flow chart of algorithm evaluation method provided by the embodiments of the present application;
Fig. 5 is a kind of structure chart of algorithm evaluating apparatus provided by the embodiments of the present application;
Fig. 6 is the structure chart of a kind of electronic equipment provided by the embodiments of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall in the protection scope of this application.
In the related technology, evaluation personnel is cumbersome, and evaluation efficiency is lower, and only more former by evaluation personnel observation The reliability of beginning image and target image, obtained evaluation result is lower.
To solve the above-mentioned problems, the embodiment of the present application provides a kind of algorithm evaluation method, can be applied to electronics and sets Standby, which can be terminal, or server.Electronic equipment locally can store multiple images processing and calculate Method, the multiple images Processing Algorithm of storage may include image processing algorithm to be evaluated.
Electronic equipment available original image and target image to be processed, wherein target image is according to target figure The image obtained after handling as Processing Algorithm original image, and evaluation request is sent to pre-set user terminal, evaluation is asked The address of the evaluation page is carried in asking, evaluates in the page and shows original image and target image, correspondingly, logging in default use The user of family terminal, the content that can be shown according to the evaluation page is evaluated, and is returned to by user terminal to electronic equipment Return the evaluation result for indicating the treatment effect of target image Processing Algorithm.
Based on the present processes, evaluation personnel need to only determine original image and target image Processing Algorithm to be evaluated, Electronic equipment can obtain evaluation result from pre-set user terminal, eliminate evaluation personnel from server and download target image, and The step of comparing original image and target image, can be improved evaluation efficiency, in addition, only by appraiser in compared with the existing technology The evaluation result that original image and target image obtain is compared in member's observation, and the embodiment of the present application, which can obtain, logs in pre-set user end The evaluation result of multiple users at end reduces evaluation result individual subjectivity to a certain extent, improves the reliable of evaluation result Property.
Referring to Fig. 1, Fig. 1 is a kind of flow chart of algorithm evaluation method provided by the embodiments of the present application, and this method can wrap Include following steps:
S101: original image and target image to be processed are obtained.
Wherein, target image is the image obtained after being handled according to target image Processing Algorithm original image.Mesh Logo image Processing Algorithm is current image processing algorithm to be evaluated, and target image Processing Algorithm can be one, or It is multiple.
In a kind of implementation, electronic equipment can directly acquire the original image and target image of evaluation personnel input. Original image can be picture, alternatively, original image may be video image.
Target image Processing Algorithm can be the image processing algorithm that electronic equipment is locally stored, that is, evaluation personnel can be with The selection target image processing algorithm from the image processing algorithm that electronic equipment is locally stored, alternatively, evaluation personnel can also be to The target image Processing Algorithm of electronic equipment input needs assessment.
Optionally, S101 may comprise steps of: receive evaluation instruction;Obtain original image to be processed, and according to Target image Processing Algorithm handles original image, obtains target image.
Wherein, the algorithm mark of target image Processing Algorithm is carried in evaluation instruction.
Electronic equipment locally can store multiple images Processing Algorithm, evaluation personnel can select according to demand currently to The image processing algorithm (i.e. target image Processing Algorithm) of evaluation.
Evaluation personnel can be carried target image processing to electronic equipment input and be calculated by the input part of electronic equipment The evaluation instruction of the algorithm mark of method, correspondingly, electronic equipment can receive evaluation instruction, extraction obtains algorithm mark, into And electronic equipment can be identified according to the algorithm, determine the target image Processing Algorithm of evaluation personnel needs assessment, it in turn, can To handle original image, target image is obtained.
S102: sending evaluation request to pre-set user terminal, so as to log in the user of pre-set user terminal according to evaluation page The content that face is shown is evaluated.
Wherein, the address that the evaluation page is carried in evaluation request, evaluates in the page and shows original image and target figure Picture, pre-set user terminal can be multiple user terminals.
After electronic equipment obtains target image, electronic equipment, which then can be generated, shows original image and target image The page is evaluated, then, electronic equipment can send the evaluation request for carrying the address of the evaluation page to pre-set user terminal.
Correspondingly, pre-set user terminal can be commented according to the address, access after pre-set user terminal receives the address The valence page will evaluate the content shown in the page and be presented to user.User is browsed in the evaluation page by pre-set user terminal It, can be by pre-set user terminal to the picture quality of original image and target image after the original image and target image of display It is compared.For example, user the picture quality to original image and target image can give a mark respectively, fractional value is higher can be with table Show that picture quality is higher.
In a kind of mode, electronic equipment can send the address of the evaluation page to pre-set user terminal, and be by mail Each user terminal distribution has the temporary account of default effective time.Correspondingly, pre-set user terminal then can be according to distribution Holding account, the access evaluation page, in turn, user then can according to the content that shows in the evaluation page, to original image and The picture quality of target image is compared.
S103: obtaining what pre-set user terminal returned, indicates the evaluation result of the treatment effect of target image Processing Algorithm.
It is compared, is determined respective by picture quality of the pre-set user terminal to original image and target image in user Picture quality after, electronic equipment then it is available indicate target image Processing Algorithm treatment effect evaluation result.
As it can be seen that being based on the present processes, evaluation personnel is eliminated from server and downloads target image, and compares original graph The step of picture and target image, can be improved evaluation efficiency, in addition, only being compared by evaluation personnel observation in compared with the existing technology The evaluation result that original image and target image obtain, the embodiment of the present application can obtain the multiple use for logging in pre-set user terminal The evaluation result at family reduces the personal subjectivity of evaluation result to a certain extent, improves the reliability of evaluation result.
Optionally, the index mark of evaluation index can also be carried in evaluation instruction, evaluation index compares for determining The index of the picture quality of image.
In a kind of mode, evaluation index can simply be divided into " picture quality is good ", " picture quality is bad ".Correspondingly, For original image and treated target image, user can visually more respective picture quality, and then it is more former The picture quality of beginning image and target image, obtain " original image quality is relatively good " or " objective image quality is relatively good " or " original image and objective image quality are similar " can determine the treatment effect of target image Processing Algorithm in turn.
Alternatively, evaluation index also may include other indexs such as picture noise, image detail richness.For example, user's root According to this index of picture noise, give a mark to original image and target image, and/or, user according to image detail richness this Index gives a mark to original image and target image.The score value of image is higher, can indicate that the picture quality of image is better.Into And the treatment effect of target image Processing Algorithm can be determined according to the score value of each evaluation index.
In a kind of implementation, evaluation personnel can be logged in and be matched in advance in electronic equipment by way of account and password The algorithm evaluation platform set evaluates task by the algorithm evaluation platform generating algorithm.
Referring to fig. 2, Fig. 2 is the schematic diagram that a kind of evaluation task provided by the embodiments of the present application generates interface.
Evaluation personnel can choose creation algorithm evaluation task, correspondingly, the algorithm is commented after logging in algorithm evaluation platform Valence platform can then show that evaluation task shown in Fig. 2 generates interface.
Evaluation personnel can appoint for indicating that this algorithm is evaluated in input in the text input box of " evaluation and test task names " The title of business, for example, evaluation personnel this need to evaluate image processing algorithm 1, then evaluation personnel can be in the text " for the evaluation task of image processing algorithm 1 " is inputted in input frame.
In Fig. 2, " picture " and " video " corresponding option can be set in the choice box after " evaluation and test type ", that is, such as Fruit evaluation personnel needs evaluate image processing algorithm according to picture, and evaluation personnel can select afterwards in " evaluation and test type " " picture " option;If evaluation personnel needs evaluate image processing algorithm according to video, evaluation personnel can be in " evaluation and test Type " selects " video " option afterwards.That is, original image can be picture, or video.
In Fig. 2, " evaluation and test dimension " is for indicating the evaluation method that evaluation personnel can choose, for example, evaluation personnel can be with It selects " tripartite division ", correspondingly, when the picture quality to original image and target image is compared, it is available to include " original image quality is relatively good ", " objective image quality is relatively good " or " original image and objective image quality are similar " three As a result.
For another example, be also provided with " the five-grade marking system " in " evaluation and test dimension ", " the five-grade marking system " may include " very poor ", " slightly poor ", " similar ", " slightly good " and " fine " five fractional levels.For example, this index of picture noise can be based on, determine respectively original Image and target image are belonging respectively to which above-mentioned fractional levels.It can also be based on this index of image detail richness, point Not Que Ding original image and target image which above-mentioned fractional levels be belonging respectively to.
In Fig. 2, way of contrast includes " original image VS result set " and " result set VS result set ".If evaluation personnel needs root According to original image and target image, the treatment effect of a certain image processing algorithm is evaluated, then evaluation personnel can choose " original image VS Result set ";If opposite treatment effect between evaluation personnel needs assessment multiple images Processing Algorithm, evaluation personnel can be with It selects " result set VS result set ".
In Fig. 2, evaluation personnel can select original image to be processed according to " original image " item.Evaluation personnel can basis The drop down button of choice box selects original image to be processed.Shown original image can be in the drop down button of choice box It is the image that algorithm evaluation platform is locally stored.
It is understood that if algorithm evaluation platform evaluation personnel is locally not present needed for original image, evaluate Personnel can click " upload " button on right side, and required original image is uploaded to algorithm evaluation platform.
In Fig. 2, evaluation personnel can select the corresponding algorithm model of image processing algorithm according to " result set " item.For example, Algorithm model may include that oversubscription repairs high version model, the oversubscription height of matching with version model, oversubscription reparation light weight version model, oversubscription light weight Version model, above-mentioned each algorithm model can be handled original image according to different super resolution algorithms.
After electronic equipment obtains original image, the target image that electronic equipment can be specified according to evaluation personnel, which is handled, to be calculated Method handles original image, the image (i.e. target image) that obtains that treated.
The mode that electronic equipment obtains original image can be diversified.
In a kind of mode, in the case where carrying image identification in evaluation instruction, referring to evaluation of being locally stored is obtained The corresponding image of the image identification carried in order, as original image.
For example, picture library has locally can be set in electronic equipment, pre-set image can store in picture library, when evaluation personnel needs When evaluating target image Processing Algorithm, evaluation personnel can select in the pre-set image that electronic equipment is locally stored Original image.That is, evaluation personnel can input the evaluation instruction for carrying image identification to electronic equipment.
Correspondingly, electronic equipment then can be according to the image identification carried in evaluation instruction, in the default figure being locally stored As in inquired, will pre-set image corresponding with image identification as original image.
In the method, Fig. 2 can be referred to, evaluation personnel can select electronic equipment to be locally stored according to " original image " item Image is as original image.
In addition, electronic equipment can also support user to upload original image, a kind of in order to meet the individual demand of user In mode, in the case where carrying image in evaluation instruction, the image carried in evaluation instruction is obtained, as original image.
In the method, can refer to Fig. 2, if electronic equipment evaluation personnel is locally not present needed for original image, Evaluation personnel can click " upload " button on right side, required original image is uploaded to electronic equipment, that is, evaluation personnel can To carry the evaluation instruction of original image to electronic equipment input.
Correspondingly, electronic equipment can extract the image carried in evaluation instruction, as original image.It is understood that It is that after the image carried in obtaining evaluation instruction, which can also be stored in local picture library by electronic equipment, with reality The update of existing picture library, so that evaluation personnel can directly select the image in picture library, as original image next time.
Optionally, if also carrying the index mark of evaluation index in evaluation instruction, this method is also before S102 It may comprise steps of: showing that index identifies corresponding objective appraisal index in the evaluation page, so as to log in pre-set user The user of terminal evaluates according to picture quality of the objective appraisal index to original image and target image.
If carrying index mark in evaluation instruction, it can be identified according to index if electronic equipment and determine that evaluation personnel refer to Fixed evaluation index (i.e. objective appraisal index), and the displaying target evaluation index in the evaluation page.
Correspondingly, user can be according to objective appraisal index when user is by the pre-set user terminal browsing evaluation page The picture quality of original image and target image is compared.
Optionally, in order to realize the overall merit to image, objective appraisal index may include at least one of: image Noise, image blocking artifact, image ringing effect, edge definition and image detail richness.
In a kind of implementation, in the evaluation page, for original image and target image, electronic equipment can be shown respectively Show above-mentioned five evaluation indexes, for each evaluation index, electronic equipment shows five choice boxs in the evaluation page, respectively table Show " very poor ", " slightly poor ", " similar ", " slightly good " and " fine " five fractional levels, correspondingly, user can be according to visual impression By determining the corresponding fractional levels of five evaluation indexes of original image and target image.
Alternatively, for original image and target image, electronic equipment can show above-mentioned five respectively in the evaluation page Evaluation index, for each evaluation index, electronic equipment shows optional fractional value range in the evaluation page, and fractional value is higher Indicate that picture quality is higher, correspondingly, user can determine five evaluations of original image and target image according to visual experience The corresponding fractional value of index.
If the evaluation result that pre-set user terminal returns includes that original image and target image obtain on target indicator Fractional value, after S103, this method may also comprise the following steps::
Step 1 calculates the average value for the fractional value that original image obtains on target indicator, as commenting for original image Valence score calculates the average value for the fractional value that target image obtains on target indicator, the evaluation score as target image.
The case where in order to avoid maliciously scoring, after the evaluation result for getting user terminal return, electronic equipment can be with Fractional value in evaluation result is further processed.
Electronic equipment can calculate the average value for each fractional value that original image obtains on target indicator, as original graph The evaluation score of picture, and the average value for each fractional value that target image obtains on target indicator is calculated, as target image Evaluation score.
In addition, electronic equipment can also screen fractional value before the average value for calculating each fractional value, for example, Electronic equipment can delete excessively high and too low fractional value, and calculate the average value of remaining fractional value, as evaluation score.
It is understood that being commented if in evaluation result including the corresponding fractional value of multiple evaluation indexes for each Valence index, electronic equipment can calculate the average value of the corresponding each fractional value of the evaluation index, as image in the evaluation index Evaluation score, in turn, the evaluation score of the available each evaluation index of electronic equipment.
Alternatively, electronic equipment can also calculate the average value of the corresponding all fractional values of each evaluation index, as image Evaluation score.
Step 2 determines that target image processing is calculated according to the evaluation score of the evaluation score of original image and target image The treatment effect grade of method.
Wherein, treatment effect grade can indicate the effect that target image Processing Algorithm handles image.
In a kind of implementation, if the evaluation score of original image is less than the evaluation score of target image, electronic equipment The treatment effect grade that can determine target image Processing Algorithm is " outstanding ";If the evaluation score of original image is greater than target The evaluation score of image, electronic equipment can determine that the treatment effect grade of target image Processing Algorithm is " poor ".
In addition, being directed to original image and target image, the evaluation score of multiple evaluation indexes is calculated in electronic equipment, It is then directed to each evaluation index, if electronic equipment determines that original image is less than target image in the evaluation score of the evaluation index In the evaluation score of the evaluation index, then electronic equipment can determine the evaluation index as positive evaluation index, and in turn, electronics is set The total number (being properly termed as the first total number) of standby available positive evaluation index, and calculate the first total number and evaluation index Total number ratio (being properly termed as the first ratio).
When the first ratio is greater than the first default ratio, electronic equipment can determine the processing effect of target image Processing Algorithm Fruit grade is " outstanding ";When the first ratio is less than the second default ratio, electronic equipment can determine target image Processing Algorithm Treatment effect grade be " poor ".
In addition, electronic equipment can also show evaluation result to evaluation personnel, optionally, target image processing calculation is being determined After the treatment effect grade of method, this method may also comprise the following steps:: generating evaluation result report.
It wherein, include original image, target image, the algorithm mark of target image Processing Algorithm, mesh in appraisal report Fractional value, the target image that the index mark of mark index, original image obtain on target indicator obtain on target indicator The treatment effect grade of fractional value and target image Processing Algorithm.
Referring to Fig. 3, Fig. 3 is a kind of schematic diagram of evaluation result report provided by the embodiments of the present application,
In Fig. 3, evaluation and test type is picture, and evaluation and test dimension is tripartite division.Application model scene are as follows: oversubscription repairs height and matches version, Indicate that image processing algorithm to be evaluated is that oversubscription repairs height with the corresponding algorithm of version model.Participating in evaluation and test number is 1, indicates pre- If user terminal is 1.Evaluating and testing sample size is 3, and expression original image is 3 pictures.Task evaluates and tests the time: 2 divide 24 when 2 Second, it can indicate to generate evaluation task from evaluation personnel, report duration used to evaluation result is generated.
Two images shown in Fig. 3, left side are original image, right side is that treated target image, two images It can be what electronic equipment was screened according to evaluation score.Numerical value above original image and target image indicates respective and comments Valence score.
Referring to fig. 4, Fig. 4 is a kind of exemplary flow chart of algorithm evaluation method provided by the embodiments of the present application, and this method can With the following steps are included:
S401: evaluation instruction is received.
Wherein, evaluation instruction in carry target image Processing Algorithm algorithm mark, evaluation index index mark and Image identification.
S402: the image corresponding with the image identification carried in evaluation instruction being locally stored is obtained, as original image.
S403: target image Processing Algorithm corresponding with the algorithm mark carried in evaluation instruction is determined, and according to target Image processing algorithm handles original image, the target image that obtains that treated.
S404: evaluation request is sent to pre-set user terminal, so that the user for logging in pre-set user terminal comments according to target Valence index is compared the picture quality of original image and target image.
Wherein, the address that the evaluation page is carried in evaluation request, evaluates in the page and shows original image, target image Corresponding objective appraisal index is identified with index.
S405: obtaining what pre-set user terminal returned, indicates the evaluation result of the treatment effect of target image Processing Algorithm.
Wherein, evaluation result includes the fractional value that original image and target image obtain on target indicator.
S406: the average value for the fractional value that original image obtains on target indicator, the evaluation as original image are calculated Score, and the average value for the fractional value that target image obtains on target indicator is calculated, the evaluation score as target image.
S407: according to the evaluation score of the evaluation score of original image and target image, target image Processing Algorithm is determined Treatment effect grade.
Wherein, treatment effect grade indicates the effect that target image Processing Algorithm handles image.
S408: evaluation result report is generated.
It wherein, include original image, target image, the algorithm mark of target image Processing Algorithm, mesh in appraisal report Fractional value, the target image that the index mark of mark index, original image obtain on target indicator obtain on target indicator The treatment effect grade of fractional value and target image Processing Algorithm.
Corresponding with the embodiment of the method for Fig. 1, referring to Fig. 5, Fig. 5 is a kind of algorithm evaluation dress provided by the embodiments of the present application The structure chart set, the apparatus may include:
First obtains module 501, for obtaining original image and target image to be processed, wherein target image is root The image obtained after handling according to target image Processing Algorithm original image;
Sending module 502 carries the evaluation page for sending evaluation request to pre-set user terminal in evaluation request Address is evaluated in the page and shows original image and target image, so as to log in the user of pre-set user terminal according to evaluation page The content that face is shown is evaluated;
Second obtains module 503, for obtaining the return of pre-set user terminal, indicates the processing of target image Processing Algorithm The evaluation result of effect.
Optionally, first module 501 is obtained, is specifically used for receiving evaluation instruction, wherein carry target in evaluation instruction The algorithm of image processing algorithm identifies;
Original image to be processed is obtained, and original image is handled according to target image Processing Algorithm, obtains mesh Logo image.
Optionally, first module 501 is obtained, is specifically used for executing following one:
In the case where carrying image identification in evaluation instruction, obtain be locally stored with the figure that carries in evaluation instruction As identifying corresponding image, as original image;
In the case where carrying image in evaluation instruction, the image carried in evaluation instruction is obtained, as original image.
Optionally, device further include:
Display module, for receiving evaluation instruction, wherein the index mark of evaluation index is also carried in evaluation instruction;
Show that index identifies corresponding objective appraisal index in the evaluation page, so as to log in the use of pre-set user terminal Family is evaluated according to picture quality of the objective appraisal index to original image and target image.
Optionally, target indicator includes at least one of: picture noise, image blocking artifact, image ringing effect, image Marginal definition and image detail richness.
Optionally, evaluation result includes the fractional value that original image and target image obtain on target indicator;
Device further include:
Processing module, for calculating the average value for the fractional value that original image obtains on target indicator, as original graph The evaluation score of picture;
Calculate the average value for the fractional value that target image obtains on target indicator, the evaluation score as target image;
According to the evaluation score of the evaluation score of original image and target image, the processing of target image Processing Algorithm is determined Gradation of effects, treatment effect grade can indicate the effect that target image Processing Algorithm handles image.
Optionally, processing module, be also used to generate evaluation result report, wherein include in appraisal report original image, Target image, the algorithm mark of target image Processing Algorithm, the index of target indicator mark, original image obtain on target indicator To fractional value, the fractional value that target image obtains on target indicator and target image Processing Algorithm treatment effect grade.
Optionally, original image is picture or video image.
The embodiment of the present application also provides a kind of electronic equipment, as shown in fig. 6, including memory 601 and processor 602;
Memory 601, for storing computer program;
Processor 602 when for executing the program stored on memory 601, realizes calculation provided by the embodiments of the present application Method evaluation method.
Specifically, above-mentioned algorithm evaluation method, comprising:
Obtain original image and target image to be processed, wherein target image is according to target image Processing Algorithm pair The image that original image obtains after being handled;
Evaluation request is sent to pre-set user terminal, the address of the evaluation page is carried in evaluation request, is evaluated in the page Original image and target image are shown, so that the user for logging in pre-set user terminal carries out according to the content that the evaluation page is shown Evaluation;
It obtains what pre-set user terminal returned, indicates the evaluation result of the treatment effect of target image Processing Algorithm.
It should be noted that other implementations of above-mentioned algorithm evaluation method are identical as preceding method embodiment part, Which is not described herein again.
Above-mentioned electronic equipment can have the communication interface realized and communicated between above-mentioned electronic equipment and other equipment.
Above-mentioned processor, communication interface, memory complete mutual communication by communication bus, referring herein to it is logical Letter bus can be Peripheral Component Interconnect standard (Peripheral Component Interconnect, abbreviation PCI) bus or Expanding the industrial standard structure (Extended Industry Standard Architecture, abbreviation EISA) bus etc..This is logical Letter bus can be divided into address bus, data/address bus, control bus etc..
Memory may include random access memory (Random Access Memory, abbreviation RAM), also may include Nonvolatile memory (Non-Volatile Memory, abbreviation NVM), for example, at least a magnetic disk storage.Optionally, it deposits Reservoir can also be that at least one is located remotely from the storage device of aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit, Abbreviation CPU), network processing unit (Network Processor, abbreviation NP) etc.;It can also be digital signal processor (Digital Signal Processing, abbreviation DSP), specific integrated circuit (Application Specific Integrated Circuit, abbreviation ASIC), field programmable gate array (Field-Programmable Gate Array, Abbreviation FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware components.
The embodiment of the present application also provides a kind of computer readable storage medium, stored in the computer readable storage medium There is instruction, when run on a computer, so that computer executes algorithm evaluation method provided by the embodiments of the present application.
Specifically, above-mentioned algorithm evaluation method, comprising:
Obtain original image and target image to be processed, wherein target image is according to target image Processing Algorithm pair The image that original image obtains after being handled;
Evaluation request is sent to pre-set user terminal, the address of the evaluation page is carried in evaluation request, is evaluated in the page Original image and target image are shown, so that the user for logging in pre-set user terminal carries out according to the content that the evaluation page is shown Evaluation;
It obtains what pre-set user terminal returned, indicates the evaluation result of the treatment effect of target image Processing Algorithm.
It should be noted that other implementations of above-mentioned algorithm evaluation method are identical as preceding method embodiment part, Which is not described herein again.
The embodiment of the present application also provides a kind of computer program products comprising instruction, when it runs on computers When, so that computer executes algorithm evaluation method provided by the embodiments of the present application.
Specifically, above-mentioned algorithm evaluation method, comprising:
Obtain original image and target image to be processed, wherein target image is according to target image Processing Algorithm pair The image that original image obtains after being handled;
Evaluation request is sent to pre-set user terminal, the address of the evaluation page is carried in evaluation request, is evaluated in the page Original image and target image are shown, so that the user for logging in pre-set user terminal carries out according to the content that the evaluation page is shown Evaluation;
It obtains what pre-set user terminal returned, indicates the evaluation result of the treatment effect of target image Processing Algorithm.
It should be noted that other implementations of above-mentioned algorithm evaluation method are identical as preceding method embodiment part, Which is not described herein again.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program Product includes one or more computer instructions.When loading on computers and executing the computer program instructions, all or It partly generates according to process or function described in the embodiment of the present application.The computer can be general purpose computer, dedicated meter Calculation machine, computer network or other programmable devices.The computer instruction can store in computer readable storage medium In, or from a computer readable storage medium to the transmission of another computer readable storage medium, for example, the computer Instruction can pass through wired (such as coaxial cable, optical fiber, number from a web-site, computer, server or data center User's line (DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or Data center is transmitted.The computer readable storage medium can be any usable medium that computer can access or It is comprising data storage devices such as one or more usable mediums integrated server, data centers.The usable medium can be with It is magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid State Disk (SSD)) etc..
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that There is also other identical elements in process, method, article or equipment including the element.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device, For electronic equipment, computer readable storage medium, computer program product embodiments, implement since it is substantially similar to method Example, so being described relatively simple, the relevent part can refer to the partial explaination of embodiments of method.
The foregoing is merely the preferred embodiments of the application, are not intended to limit the protection scope of the application.It is all Any modification, equivalent replacement, improvement and so within spirit herein and principle are all contained in the protection scope of the application It is interior.

Claims (18)

1. a kind of algorithm evaluation method, which is characterized in that the described method includes:
Obtain original image and target image to be processed, wherein the target image is according to target image Processing Algorithm pair The image that the original image obtains after being handled;
Evaluation request is sent to pre-set user terminal, the address of the evaluation page, the evaluation page are carried in the evaluation request The original image and the target image are shown in face, so as to log in the user of the pre-set user terminal according to institute's commentary The content that the valence page is shown is evaluated;
It obtains what the pre-set user terminal returned, indicates the evaluation result of the treatment effect of the target image Processing Algorithm.
2. the method according to claim 1, wherein described obtain original image and target image to be processed, Include:
Receive evaluation instruction, wherein the algorithm mark of the target image Processing Algorithm is carried in the evaluation instruction;
Original image to be processed is obtained, and the original image is handled according to the target image Processing Algorithm, is obtained To the target image.
3. according to the method described in claim 2, it is characterized in that, described obtain original image to be processed, including it is following it One:
The evaluation instruction in carry image identification in the case where, obtain be locally stored with the evaluation instruct in carry The corresponding image of image identification, as the original image;
In the case where carrying image in evaluation instruction, the image carried in the evaluation instruction is obtained, as described Original image.
4. the method according to claim 1, wherein sending evaluation request to pre-set user terminal described Before, the method also includes:
Receive evaluation instruction, wherein the index mark of evaluation index is also carried in the evaluation instruction;
Show that the index identifies corresponding objective appraisal index in the evaluation page, so that it is whole to log in the pre-set user The user at end evaluates the picture quality of the original image and the target image according to the objective appraisal index.
5. according to the method described in claim 4, it is characterized in that, the target indicator includes at least one of: image is made an uproar Sound, image blocking artifact, image ringing effect, edge definition and image detail richness.
6. according to the method described in claim 4, it is characterized in that, the evaluation result includes the original image and the mesh The fractional value that logo image obtains on the target indicator;
It obtains what the pre-set user terminal returned described, indicates the evaluation of the treatment effect of the target image Processing Algorithm As a result after, the method also includes:
The average value for calculating the fractional value that the original image obtains on the target indicator, as commenting for the original image Valence score;
The average value for calculating the fractional value that the target image obtains on the target indicator, as commenting for the target image Valence score;
According to the evaluation score of the evaluation score of the original image and the target image, determine that the target image processing is calculated The treatment effect grade of method, the treatment effect grade indicate the effect that the target image Processing Algorithm handles image Fruit.
7. according to the method described in claim 6, it is characterized in that, the determination target image Processing Algorithm processing After gradation of effects, the method also includes:
Generate evaluation result report, wherein include the original image in the appraisal report, the target image, described Algorithm mark, the index of the target indicator of target image Processing Algorithm identify, the original image is in the target indicator On the fractional value that is obtained on the target indicator of obtained fractional value, the target image and the target image Processing Algorithm Treatment effect grade.
8. method according to claim 1-7, which is characterized in that the original image is picture or video Image.
9. a kind of algorithm evaluating apparatus, which is characterized in that described device includes:
First obtains module, for obtaining original image and target image to be processed, wherein the target image is according to mesh The image that logo image Processing Algorithm obtains after handling the original image;
Sending module carries the ground of the evaluation page for sending evaluation request to pre-set user terminal in the evaluation request Location shows the original image and the target image in the evaluation page, so as to log in the pre-set user terminal User evaluates according to the content that the evaluation page is shown;
Second obtains module, for obtaining the pre-set user terminal return, indicates the place of the target image Processing Algorithm Manage the evaluation result of effect.
10. device according to claim 9, which is characterized in that described first obtains module, refers to specifically for receiving evaluation It enables, wherein the algorithm mark of the target image Processing Algorithm is carried in the evaluation instruction;
Original image to be processed is obtained, and the original image is handled according to the target image Processing Algorithm, is obtained To the target image.
11. device according to claim 10, which is characterized in that described first obtains module, is specifically used for executing following One of:
The evaluation instruction in carry image identification in the case where, obtain be locally stored with the evaluation instruct in carry The corresponding image of image identification, as the original image;
In the case where carrying image in evaluation instruction, the image carried in the evaluation instruction is obtained, as described Original image.
12. device according to claim 9, which is characterized in that described device further include:
Display module, for receiving evaluation instruction, wherein the index mark of evaluation index is also carried in the evaluation instruction;
Show that the index identifies corresponding objective appraisal index in the evaluation page, so that it is whole to log in the pre-set user The user at end evaluates the picture quality of the original image and the target image according to the objective appraisal index.
13. device according to claim 12, which is characterized in that the target indicator includes at least one of: image Noise, image blocking artifact, image ringing effect, edge definition and image detail richness.
14. device according to claim 12, which is characterized in that the evaluation result includes the original image and described The fractional value that target image obtains on the target indicator;
Described device further include:
Processing module, for calculating the average value for the fractional value that the original image obtains on the target indicator, as institute State the evaluation score of original image;
The average value for calculating the fractional value that the target image obtains on the target indicator, as commenting for the target image Valence score;
According to the evaluation score of the evaluation score of the original image and the target image, determine that the target image processing is calculated The treatment effect grade of method, the treatment effect grade can indicate what the target image Processing Algorithm handled image Effect.
15. device according to claim 14, which is characterized in that the processing module is also used to generate evaluation result report It accuses, wherein include the original image, the target image, the target image Processing Algorithm in the appraisal report The fractional value, described that algorithm mark, the index of target indicator mark, the original image obtain on the target indicator The treatment effect grade of fractional value and the target image Processing Algorithm that target image obtains on the target indicator.
16. according to the described in any item devices of claim 9-15, which is characterized in that the original image is picture or view Frequency image.
17. a kind of electronic equipment, which is characterized in that including memory and processor;
Memory, for storing computer program;
Processor when for executing the program stored on memory, realizes any method and step of claim 1-8.
18. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium Program realizes claim 1-8 any method and step when the computer program is executed by processor.
CN201910700408.2A 2019-07-31 2019-07-31 A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing Pending CN110428368A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910700408.2A CN110428368A (en) 2019-07-31 2019-07-31 A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910700408.2A CN110428368A (en) 2019-07-31 2019-07-31 A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing

Publications (1)

Publication Number Publication Date
CN110428368A true CN110428368A (en) 2019-11-08

Family

ID=68413260

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910700408.2A Pending CN110428368A (en) 2019-07-31 2019-07-31 A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing

Country Status (1)

Country Link
CN (1) CN110428368A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111223054A (en) * 2019-11-19 2020-06-02 深圳开立生物医疗科技股份有限公司 Ultrasonic image evaluation method and device
CN111709922A (en) * 2020-06-10 2020-09-25 北京百度网讯科技有限公司 Image quality comparison method, device, equipment and storage medium
CN112767318A (en) * 2020-12-31 2021-05-07 科大讯飞股份有限公司 Image processing effect evaluation method and device, storage medium and equipment
CN116701923A (en) * 2022-10-13 2023-09-05 荣耀终端有限公司 Operator performance evaluation method and device
CN118014392A (en) * 2024-02-01 2024-05-10 中国铁塔股份有限公司 Algorithm evaluation method, device, electronic equipment and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106296669A (en) * 2016-08-01 2017-01-04 微梦创科网络科技(中国)有限公司 A kind of image quality evaluating method and device
CN108074241A (en) * 2018-01-16 2018-05-25 深圳大学 Quality score method, apparatus, terminal and the storage medium of target image
WO2018214115A1 (en) * 2017-05-25 2018-11-29 华为技术有限公司 Face makeup evaluation method and device
CN109034242A (en) * 2018-07-26 2018-12-18 北京小米移动软件有限公司 Methods of marking, the apparatus and system of image processing algorithm
CN109522960A (en) * 2018-11-21 2019-03-26 泰康保险集团股份有限公司 Image evaluation method, device, electronic equipment and computer-readable medium
CN109525888A (en) * 2018-09-28 2019-03-26 Oppo广东移动通信有限公司 Image display method, device, electronic equipment and storage medium
CN109684581A (en) * 2018-09-07 2019-04-26 平安科技(深圳)有限公司 Evaluation method, device, equipment and storage medium based on user characteristic data
CN109885790A (en) * 2018-12-30 2019-06-14 贝壳技术有限公司 The method and apparatus for obtaining satisfaction evaluation data

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106296669A (en) * 2016-08-01 2017-01-04 微梦创科网络科技(中国)有限公司 A kind of image quality evaluating method and device
WO2018214115A1 (en) * 2017-05-25 2018-11-29 华为技术有限公司 Face makeup evaluation method and device
CN108074241A (en) * 2018-01-16 2018-05-25 深圳大学 Quality score method, apparatus, terminal and the storage medium of target image
CN109034242A (en) * 2018-07-26 2018-12-18 北京小米移动软件有限公司 Methods of marking, the apparatus and system of image processing algorithm
CN109684581A (en) * 2018-09-07 2019-04-26 平安科技(深圳)有限公司 Evaluation method, device, equipment and storage medium based on user characteristic data
CN109525888A (en) * 2018-09-28 2019-03-26 Oppo广东移动通信有限公司 Image display method, device, electronic equipment and storage medium
CN109522960A (en) * 2018-11-21 2019-03-26 泰康保险集团股份有限公司 Image evaluation method, device, electronic equipment and computer-readable medium
CN109885790A (en) * 2018-12-30 2019-06-14 贝壳技术有限公司 The method and apparatus for obtaining satisfaction evaluation data

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111223054A (en) * 2019-11-19 2020-06-02 深圳开立生物医疗科技股份有限公司 Ultrasonic image evaluation method and device
CN111223054B (en) * 2019-11-19 2024-03-15 深圳开立生物医疗科技股份有限公司 Ultrasonic image evaluation method and device
CN111709922A (en) * 2020-06-10 2020-09-25 北京百度网讯科技有限公司 Image quality comparison method, device, equipment and storage medium
CN111709922B (en) * 2020-06-10 2023-07-04 北京百度网讯科技有限公司 Image quality comparison method, device, equipment and storage medium
CN112767318A (en) * 2020-12-31 2021-05-07 科大讯飞股份有限公司 Image processing effect evaluation method and device, storage medium and equipment
CN112767318B (en) * 2020-12-31 2023-07-25 科大讯飞股份有限公司 Image processing effect evaluation method, device, storage medium and equipment
CN116701923A (en) * 2022-10-13 2023-09-05 荣耀终端有限公司 Operator performance evaluation method and device
CN116701923B (en) * 2022-10-13 2024-05-17 荣耀终端有限公司 Operator performance evaluation method and device
CN118014392A (en) * 2024-02-01 2024-05-10 中国铁塔股份有限公司 Algorithm evaluation method, device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN110428368A (en) A kind of algorithm evaluation method, device, electronic equipment and readable storage medium storing program for executing
US8856215B2 (en) Detecting visible display of content
CN105630977B (en) Application program recommended method, apparatus and system
CN106886491A (en) Webpage compatibility test method and device
CN108829597A (en) Software crowd surveys method and device, computer installation and readable storage medium storing program for executing
CN105912550A (en) Method and device for information recommendation of mobile terminal
CN110348519A (en) Financial product cheats recognition methods and the device of clique
US20150154564A1 (en) Weighted evaluation comparison
CN109670963A (en) Assets information screening technique, device, equipment and storage medium
CN108696399A (en) The test method and device of business service
US20150135019A1 (en) Automated testing of application programs from an application program ecosystem
CN108124271B (en) Network quality evaluation method and device based on user perception
CN108510204A (en) Methods of marking, device and electric terminal
CN109729433A (en) A kind of video playing appraisal procedure and device
CN104572851A (en) Method and device for acquiring recommend information
US20180048860A1 (en) Method and system for measuring quality of video call
CN110505490A (en) Multi-medium data recognition methods and system
CN114880310A (en) User behavior analysis method and device, computer equipment and storage medium
CN105794150B (en) Method for measuring end-to-end Internet application performance
CN110087103A (en) A kind of video recommendation system, method, apparatus and computer
CN108062401A (en) Using recommendation method, apparatus and storage medium
CN108074108A (en) A kind of display methods and its terminal of net recommendation
US20190289357A1 (en) Camera system
CN107067128A (en) Task quality detecting method, the method and device for determining dereferenced user
CN112633619A (en) Risk assessment method and device

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20191108

RJ01 Rejection of invention patent application after publication