CN108873401B - Liquid crystal display response time prediction method based on big data - Google Patents
Liquid crystal display response time prediction method based on big data Download PDFInfo
- Publication number
- CN108873401B CN108873401B CN201810650117.2A CN201810650117A CN108873401B CN 108873401 B CN108873401 B CN 108873401B CN 201810650117 A CN201810650117 A CN 201810650117A CN 108873401 B CN108873401 B CN 108873401B
- Authority
- CN
- China
- Prior art keywords
- data
- liquid crystal
- characteristic data
- crystal display
- column
- 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
- G02—OPTICS
- G02F—OPTICAL DEVICES OR ARRANGEMENTS FOR THE CONTROL OF LIGHT BY MODIFICATION OF THE OPTICAL PROPERTIES OF THE MEDIA OF THE ELEMENTS INVOLVED THEREIN; NON-LINEAR OPTICS; FREQUENCY-CHANGING OF LIGHT; OPTICAL LOGIC ELEMENTS; OPTICAL ANALOGUE/DIGITAL CONVERTERS
- G02F1/00—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics
- G02F1/01—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour
- G02F1/13—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour based on liquid crystals, e.g. single liquid crystal display cells
- G02F1/1306—Details
- G02F1/1309—Repairing; Testing
Landscapes
- Physics & Mathematics (AREA)
- Nonlinear Science (AREA)
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Liquid Crystal (AREA)
Abstract
The invention discloses a method for predicting response time of a liquid crystal display based on big data, which solves the technical problem of lower prediction precision in the prior art and comprises the following steps: obtaining a liquid crystal display production process data set I1(ii) a Judging the data set I1Whether there is missing feature data in (a); to I1Filling the missing characteristic data; extracting key characteristic data in the complete characteristic data set; constructing a liquid crystal display response time prediction model; and predicting the response time of the liquid crystal display. According to the invention, the characteristic data missing column is filled according to the parameter data of the equipment used in each process of the liquid crystal display in the production process and the discreteness of the corresponding characteristic data missing column, so that the interference of the data filled with the missing value on the prediction model is avoided, the prediction precision of the liquid crystal display prediction model is further improved, and the method can be used for predicting the response time of the liquid crystal display.
Description
Technical Field
The invention belongs to the technical field of engineering, relates to a method for predicting the response time of a liquid crystal display, in particular to a method for predicting the response time of the liquid crystal display based on big data, and can be used for predicting the response time of the liquid crystal display.
Background
The semiconductor industry is an industry with a high degree of industrialization and informatization. The high degree of industrialization and informatization creates the possibility for data analysis. However, the production process of the lcd is complicated, and tens of thousands of lcds are produced in a production line every day, which includes hundreds of processes, and each process may affect the quality of the lcd. Although the requirements for the performance of lcd devices in different fields are different, the technical challenges of lcd devices are wide viewing angle, high contrast, wide color gamut, low power consumption and fast response time, and the response time of lcd devices is the focus of attention.
The response time of the lcd is the response speed of each pixel point of the lcd to the input signal, i.e. the time required for the pixel point to change from dark to light or from light to dark. The time required for the pixel point to change from dark to light is called uplink time, and the time required for the pixel point to change from light to dark is called downlink time. When the response time of the liquid crystal display is long, an object moving in the screen leaves "a smear of an image". As represented in a computer, sometimes a mouse is used and the position of the cursor is not visible.
The liquid crystal display response time prediction is a technology which is mainly used for analyzing data in the production process of the liquid crystal display, establishing a model by utilizing various evaluation algorithms and predicting the liquid crystal display response time. The response time of the liquid crystal display is correctly predicted, the predicted response time of the liquid crystal display can be obtained, and the working procedures corresponding to the key characteristic data can be monitored in a key mode according to the key characteristic data which have large influence on the response time of the liquid crystal display, so that the product quality of the liquid crystal display is improved.
The main idea of predicting the response time of the liquid crystal display at present is to change the composition materials of the liquid crystal display and predict the response time of the liquid crystal display according to the physical characteristics of the liquid crystal by acquiring the data generated in the detection link in the production process of the liquid crystal display. For example, 2011 in the eastern south university report, volume 41, discloses a wavelet transform based LCD response time estimation method, which constructs an LCD luminance response under a flickering background light by actually measuring a liquid crystal response between five-level gray scales under continuous background light and a background light response of the flickering background light, and predicts the liquid crystal display response time by reducing a liquid crystal response signal by utilizing a wavelet transform filtering method according to data generated in a liquid crystal display detection link through wavelet decomposition, threshold processing and inverse transformation reconstruction. The method reduces the error between the predicted value and the true value of the response time of the liquid crystal display, but has the defect that when missing data is filled, only brightness data generated in the detection link of the liquid crystal display is utilized, and the characteristic data of the liquid crystal display in the production process is not fully utilized, so that the error between the predicted value and the true value of the response time of the liquid crystal display is larger.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a method for predicting the response time of a liquid crystal display based on big data, which is used for solving the technical problem of low prediction precision in the prior art.
In order to achieve the purpose, the technical scheme adopted by the invention comprises the following steps:
(1) obtaining a liquid crystal display production process data set I1：
Removing the duplicate of the collected characteristic data of the liquid crystal display in the production process and the parameter data of the equipment used in each procedure to obtain a liquid crystal display production process data set I1；
(2) Judging the data set I1Whether there is missing characteristic data in (1):
statistics I1The number n of medium characteristic data and the number m of non-empty characteristic data, if n is not equal to m, I1The missing characteristic data exists in the data;
(3) to I1Filling in the missing characteristic data:
(3a) setting I1The discrete threshold of the middle feature data is β;
(3b) according to I1Calculating the length of a column with missing data in the characteristic data, counting the type of the characteristic data in the missing data column, and calculating the discreteness coefficient of the missing data column by the length of the missing data column and the type of the characteristic data in the column;
(3c) when the dispersion coefficient of the characteristic data missing column is larger than β, the mode of the characteristic data missing column data is used for I1Filling the missing characteristic data, and when the discreteness coefficient of the missing characteristic data column is less than or equal to β, using the mean value of the missing characteristic data column data to pair I1Filling in the missing characteristic dataFilling to obtain a complete characteristic data set;
(4) extracting key characteristic data in the complete characteristic data set:
(4a) normalizing the complete characteristic data set to obtain a normalized characteristic data set I2；
(4b) To I2Is subjected to one-way analysis of variance:
calculation of I2Will be greater than a preset threshold value FaThe feature data corresponding to the statistical test quantity are merged to obtain a feature data set I3；
(4c) To I3Carrying out distance correlation analysis on the two characteristic data:
calculation of I3The distance correlation coefficient between every two middle feature data is combined with the feature data corresponding to the distance correlation coefficient smaller than a preset threshold value a to obtain a key feature data set I4；
(5) Constructing a liquid crystal display response time prediction model:
(5a) key feature data set I4Taking 70% of the data as training set data train and the rest as test set data test;
(5b) training the training set data train by adopting an XGboost algorithm to obtain a liquid crystal display response time prediction model;
(6) predicting the response time of the liquid crystal display:
and inputting the test set data test into a liquid crystal display response time prediction model to obtain the response time of the liquid crystal display.
Compared with the prior art, the invention has the following advantages:
the method is based on the collected characteristic data and the parameter data of the equipment used in each process, fills the characteristic data missing columns according to the parameter data of the equipment used in each process of the liquid crystal display in the production process and the discreteness of the corresponding characteristic data missing columns, and constructs the liquid crystal display prediction model by using the obtained complete characteristic data set.
Drawings
FIG. 1 is a flow chart of an implementation of the present invention.
Detailed description of the invention
The invention is described in further detail below with reference to the figures and specific examples.
Referring to fig. 1, a method for predicting response time of a liquid crystal display based on big data includes the steps of:
step 1) obtaining a data set I of a liquid crystal display production process1：
Removing the duplicate of the collected characteristic data of the liquid crystal display in the production process and the parameter data of the equipment used in each procedure to obtain a liquid crystal display production process data set I1；
The production process of the liquid crystal display comprises a plurality of process links, each process link comprises a plurality of working procedures, each working procedure can influence the response time of the liquid crystal display, the characteristic data of the liquid crystal display and the parameter data of the equipment used in each working procedure in the production process are collected, all data possibly comprising the influence on the response time of the liquid crystal display can be ensured to be collected, and a better data base is provided for filling of subsequent missing data. The method avoids that the response time of the liquid crystal display is predicted by constructing a prediction model only by using data generated in a single process step, and the accuracy of the prediction model is influenced due to the loss of data which are generated in other processes and influence on the response time of the liquid crystal display.
The characteristic data of the liquid crystal display in the production process comprises the display size, the liquid crystal viscosity coefficient, the liquid crystal dielectric coefficient, the voltage, the current, the liquid crystal inclination angle, the metal film thickness, the coating film thickness, the box making data, the substrate distortion value, the liquid crystal dropping amount, the liquid crystal box thickness and the etching inclination angle.
The parameter data of the equipment used in each process comprises the voltage, the power and the consumed time of the equipment used in each process.
The production process of the liquid crystal display comprises hundreds of processes, wherein the production is carried out by taking the processes as units, and characteristic data and equipment parameter data used in each process in the production process of the liquid crystal display are collected by an equipment sensor and a manual recording mode.
When the characteristic data of the liquid crystal display and the parameter data of the equipment used in each process are collected in the production process, the characteristic data of the liquid crystal display and the parameter data of the equipment are collected after the current process is finished, but the characteristic data and the equipment parameter data which are not changed in the current process are collected again, so that repeated characteristic data can be generated in the collected characteristic data of the liquid crystal display in the production process. If the collected characteristic data of the liquid crystal display in the production process is directly utilized, the time required for predicting the response time of the liquid crystal display is increased.
In the invention, the collected characteristic data of the liquid crystal display in the production process and the parameter data of the equipment used in each process are subjected to duplication elimination according to the mic function in the R language to obtain a liquid crystal display production process data set I1。
Step 2) judging the data set I1Whether there is missing characteristic data in (1):
statistics I1The number n of medium characteristic data and the number m of non-empty characteristic data, if n is not equal to m, I1The missing characteristic data exists in the data;
when collecting characteristic data of a liquid crystal display and parameter data of equipment used in each process in the production process, missing data may be generated due to omission in data recording due to uncertainty of a manual recording mode, so that a data set I is required to be subjected to1And judging whether the missing characteristic data exists.
Step 3) to I1Filling in the missing characteristic data:
step 3a) setting I1The discrete threshold of the middle feature data is β;
step 3b) according to I1Calculating the length of a column with missing data in the characteristic data, counting the type of the characteristic data in the missing data column, and calculating the discreteness coefficient of the missing data column by the length of the missing data column and the type of the characteristic data in the column;
step 3c) when the discreteness coefficient of the characteristic data missing column is larger than β, the mode of the characteristic data missing column data is used for pairing I1Filling the missing characteristic data, and when the discreteness coefficient of the missing characteristic data column is less than or equal to β, using the mean value of the missing characteristic data column data to pair I1Filling the missing characteristic data to obtain a complete characteristic data set;
step 4), extracting key characteristic data in the complete characteristic data set:
step 4a) normalizing the complete characteristic data set to obtain a normalized characteristic data set I2；
And normalizing the complete characteristic data set to eliminate dimension influence caused by data unit inconsistency among different characteristic data in the complete characteristic data set, so that all the characteristic data are in the same reference system.
The invention adopts min-max standardization, also called dispersion standardization, which is to perform linear change on original data to map the result between [0-1], and the calculation formula is as follows:
x*＝(x-xmin)/(xmax-xmin)
wherein xmaxIs the maximum value of the sample, xminIs the minimum value of the sample, x is the original data, x*Is the data after normalization.
Step 4b) for I2Is subjected to one-way analysis of variance:
calculation of I2Will be greater than a preset threshold value FaThe feature data corresponding to the statistical test quantity are merged to obtain a feature data set I3；
One-way anova: and analyzing the test data, and checking whether the plurality of normal overall mean values with equal variances are equal or not so as to judge whether the influence of each factor on a single test index is obvious or not.
In the invention, I is paired according to aov function in R language2Performing single-factor variance analysis on each feature data to obtain a feature data set I3。
Step 4c) for I3Carrying out distance correlation analysis on the two characteristic data:
calculation of I3The distance correlation coefficient between every two middle feature data is combined with the feature data corresponding to the distance correlation coefficient smaller than a preset threshold value a to obtain a key feature data set I4；
Distance correlation analysis: a statistical analysis method for studying the correlation between two or more equally positioned random variables.
Calculation of I3The distance correlation coefficient between every two middle feature data is realized by the following steps:
step 4c1) calculating I3Distance matrix a between feature data in middle X columnj,kAnd obtaining a distance matrix b from a distance matrix between the characteristic data in the Y columnj,kThe calculation formulas are respectively as follows:
aj,k＝||Xj-Xk||,j,k＝1,2,...,n
bj,k＝||Yj-Yk||,j,k＝1,2,...,n
wherein j represents I3J-th row of (a), k represents I3Line k of (1), Xj、XkRespectively representing the characteristic data of the j-th line in X and the characteristic data of the k-th line in X, Yj、YkRespectively representing the characteristic data of the jth line in Y and the characteristic data of the kth line in Y;
step 4c2) calculating I3Center distance matrix A of feature data in middle X columnj,kAnd a center distance matrix B of feature data in Y columnsj,kThe calculation formulas are respectively as follows:
wherein
Is aj,kThe average value of the j-th row in (c),
is aj,kThe average value of the k-th row in (c),
is a matrix aj,kThe overall average value of (a) of (b),
is bj,kThe average value of the j-th row in (c),
is bj,kThe average value of the k-th row in (c),
is a matrix bj,kThe overall mean value of;
step 4c3) calculating I3Covariance of distance between feature data in column X and feature data in column Y
The calculation formula is as follows:
step 4c4) calculating I3Distance variance of feature data in middle X column
Distance variance from feature data in Y column
The calculation formula is as follows:
step 4c5) calculating I3Distance correlation coefficient between feature data in X column and feature data in Y column
The calculation formula is as follows:
step 5) constructing a liquid crystal display response time prediction model:
step 5a) Key feature dataset I4Taking 70% of the data as training set data train and the rest as test set data test;
step 5b), training the training set data train by adopting an XGboost algorithm to obtain a liquid crystal display response time prediction model;
the method comprises the following implementation steps:
step 5b1) setting iteration parameters of the XGBoost algorithm;
step 5b2) leading the training set data train into an XGboost algorithm, realizing the fitting of the training set data train and obtaining a liquid crystal display response time prediction model;
the advantage of ensemble learning over individual models is that ensemble learning integrates many "individual models" in a suitable way, resulting in a final model performance that is better than the performance of the individual models. XGboost is one of boosting family algorithms in ensemble learning, is a weak classifier lifting algorithm with different training sets, and has the advantages of high regression fitting precision, high speed and the like.
Step 6) predicting the response time of the liquid crystal display:
and inputting the test set data test into a liquid crystal display response time prediction model to obtain the response time of the liquid crystal display.
The effects of the present invention can be further illustrated by the following practices:
1) the modeling practice is completed on an RStudio running platform under an Intel (R) core (TM) i5-4210U @2.10GHz and a Windows 7 (x 64) system;
2) modeling practice is carried out by using characteristic data of the liquid crystal display in the production process and parameter data of equipment used in each process, which are acquired in the production field of the liquid crystal display of a certain enterprise;
3) in the data preprocessing, a min-max standardization method is adopted to normalize the characteristic data of the liquid crystal display in the production process and the parameter data of the equipment used in each process to obtain a normalized data set, and missing characteristic data is filled according to the discreteness of missing columns of the characteristic data to obtain a complete characteristic data set;
4) performing one-factor analysis of variance, setting a threshold value of the statistical test quantity to be 0.05, and obtaining a data set formed by the characteristic data corresponding to the statistical test quantity larger than 0.05;
5) performing distance correlation analysis, setting a distance correlation coefficient threshold value to be 0.1, and obtaining a key feature data set formed by feature data corresponding to the distance correlation coefficient smaller than 0.1;
6) according to a training set divided by a key feature data set, constructing a liquid crystal display response time prediction model through an integrated learning-based XGboost regression model, and setting the iteration times n _ estimators as 10000;
7) and inputting the test set divided by the key characteristic data set into a liquid crystal display response time prediction model to obtain a predicted value of the liquid crystal display response time.
Claims (3)
1. A method for predicting the response time of a liquid crystal display based on big data is characterized by comprising the following steps:
(1) obtaining a liquid crystal display production process data set I1：
Removing the duplicate of the collected characteristic data of the liquid crystal display in the production process and the parameter data of the equipment used in each procedure to obtain a liquid crystal display production process data set I1；
(2) Judging the data set I1Whether there is missing characteristic data in (1):
statistics I1The number n of medium characteristic data and the number m of non-empty characteristic data, if n is not equal to m, I1The missing characteristic data exists in the data;
(3) to I1Filling in the missing characteristic data:
(3a) setting I1The discrete threshold of the middle feature data is β;
(3b) according to I1Calculating the length of a column with missing data in the characteristic data, counting the type of the characteristic data in the missing data column, and calculating the discreteness coefficient of the missing data column by the length of the missing data column and the type of the characteristic data in the column;
(3c) when the dispersion coefficient of the characteristic data missing column is larger than β, the mode of the characteristic data missing column data is used for I1Filling the missing characteristic data, and when the discreteness coefficient of the missing characteristic data column is less than or equal to β, using the mean value of the missing characteristic data column data to pair I1Filling the missing characteristic data to obtain a complete characteristic data set;
(4) extracting key characteristic data in the complete characteristic data set:
(4a) normalizing the complete characteristic data set to obtain a normalized characteristic data set I2；
(4b) To I2Is subjected to one-way analysis of variance:
calculation of I2Will be greater than a preset thresholdValue FaThe feature data corresponding to the statistical test quantity are merged to obtain a feature data set I3；
(4c) To I3Carrying out distance correlation analysis on the two characteristic data:
calculation of I3The distance correlation coefficient between every two middle feature data is combined with the feature data corresponding to the distance correlation coefficient smaller than a preset threshold value α to obtain a key feature data set I4；
(5) Constructing a liquid crystal display response time prediction model:
(5a) key feature data set I4Taking 70% of the data as training set data train and the rest as test set data test;
(5b) training the training set data train by adopting an XGboost algorithm to obtain a liquid crystal display response time prediction model;
(6) predicting the response time of the liquid crystal display:
and inputting the test set data test into a liquid crystal display response time prediction model to obtain the response time of the liquid crystal display.
2. The method according to claim 1, wherein the characteristic data of the lcd in the production process and the parameter data of the equipment used in each process in the step (1) comprise the display size, the liquid crystal viscosity coefficient, the liquid crystal dielectric coefficient, the voltage, the current, the liquid crystal tilt angle, the metal film thickness, the coating film thickness, the cell making data, the substrate distortion value, the liquid crystal dropping amount, the liquid crystal cell thickness and the etching tilt angle; the parameter data of the equipment includes voltage, power and consumed time of the equipment used in each process.
3. The big data based LCD response time prediction method of claim 1, wherein the calculating I in step (4c)3The distance correlation coefficient between every two middle feature data is realized by the following steps:
(4c1) calculation of I3Distance matrix a between feature data in middle X columnj，kAnd obtaining a distance matrix b from a distance matrix between the characteristic data in the Y columnj，kThe calculation formulas are respectively as follows:
aj，k＝||Xj-Xk||，j，k＝1，2，...，n
bj，k＝||Yj-Yk||，j，k＝1，2，...，n
wherein j represents I3J-th row of (a), k represents I3Line k of (1), Xj、XkRespectively representing the characteristic data of the j-th line in X and the characteristic data of the k-th line in X, Yj、YkRespectively representing the characteristic data of the jth line in Y and the characteristic data of the kth line in Y;
(4c2) calculation of I3Center distance matrix A of feature data in middle X columnj，kAnd a center distance matrix B of feature data in Y columnsj，kThe calculation formulas are respectively as follows:
wherein
Is aj，kThe average value of the j-th row in (c),
is aj，kThe average value of the k-th row in (c),
is a matrix aj，kThe overall average value of (a) of (b),
is bj，kThe average value of the j-th row in (c),
is bj，kThe average value of the k-th row in (c),
is a matrix bj，kThe overall mean value of;
(4c3) calculation of I3Covariance of distance between feature data in column X and feature data in column Y
The calculation formula is as follows:
(4c4) calculation of I3Distance variance of feature data in middle X column
Distance variance from feature data in Y column
The calculation formula is as follows:
(4c5) calculation of I3Distance correlation coefficient between feature data in X column and feature data in Y column
The calculation formula is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810650117.2A CN108873401B (en) | 2018-06-22 | 2018-06-22 | Liquid crystal display response time prediction method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810650117.2A CN108873401B (en) | 2018-06-22 | 2018-06-22 | Liquid crystal display response time prediction method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108873401A CN108873401A (en) | 2018-11-23 |
| CN108873401B true CN108873401B (en) | 2020-10-09 |
Family
ID=64340848
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810650117.2A Active CN108873401B (en) | 2018-06-22 | 2018-06-22 | Liquid crystal display response time prediction method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108873401B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768045A (en) * | 2020-07-03 | 2020-10-13 | 上海积成能源科技有限公司 | Method for supplementing resident electricity consumption missing data by applying multiple interpolation in resident electricity consumption management |
| CN117157576A (en) * | 2022-03-30 | 2023-12-01 | 京东方科技集团股份有限公司 | Model training method, performance prediction method, device, equipment and medium |
| CN116416884B (en) * | 2023-06-12 | 2023-08-18 | 深圳市彤兴电子有限公司 | Testing device and testing method for display module |
| CN116484269B (en) * | 2023-06-25 | 2023-09-01 | 深圳市彤兴电子有限公司 | Parameter processing method, device and equipment of display screen module and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102361014A (en) * | 2011-10-20 | 2012-02-22 | 上海大学 | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process |
| CN103236383A (en) * | 2013-05-04 | 2013-08-07 | 四川虹欧显示器件有限公司 | Method for dynamically determining and optimizing parameter importance in manufacturing process of plasma display screen |
| CN104239722A (en) * | 2014-09-18 | 2014-12-24 | 山东大学 | Forecasting method based on recognition of correlational relationship between factors |
| CN105354208A (en) * | 2015-09-21 | 2016-02-24 | 江苏讯狐信息科技有限公司 | Big data information mining method |
| CN105469108A (en) * | 2015-11-17 | 2016-04-06 | 深圳先进技术研究院 | Clustering method, clustering system, clustering result evaluation method and clustering result evaluation system based on biological data |
| CN107193876A (en) * | 2017-04-21 | 2017-09-22 | 美林数据技术股份有限公司 | A kind of missing data complementing method based on arest neighbors KNN algorithms |
| CN107909433A (en) * | 2017-11-14 | 2018-04-13 | 重庆邮电大学 | A kind of Method of Commodity Recommendation based on big data mobile e-business |
-
2018
- 2018-06-22 CN CN201810650117.2A patent/CN108873401B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102361014A (en) * | 2011-10-20 | 2012-02-22 | 上海大学 | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process |
| CN103236383A (en) * | 2013-05-04 | 2013-08-07 | 四川虹欧显示器件有限公司 | Method for dynamically determining and optimizing parameter importance in manufacturing process of plasma display screen |
| CN104239722A (en) * | 2014-09-18 | 2014-12-24 | 山东大学 | Forecasting method based on recognition of correlational relationship between factors |
| CN105354208A (en) * | 2015-09-21 | 2016-02-24 | 江苏讯狐信息科技有限公司 | Big data information mining method |
| CN105469108A (en) * | 2015-11-17 | 2016-04-06 | 深圳先进技术研究院 | Clustering method, clustering system, clustering result evaluation method and clustering result evaluation system based on biological data |
| CN107193876A (en) * | 2017-04-21 | 2017-09-22 | 美林数据技术股份有限公司 | A kind of missing data complementing method based on arest neighbors KNN algorithms |
| CN107909433A (en) * | 2017-11-14 | 2018-04-13 | 重庆邮电大学 | A kind of Method of Commodity Recommendation based on big data mobile e-business |
Non-Patent Citations (1)
| Title |
|---|
| 基于小波变换的闪烁背光LCD响应时间估算;夏振平等;《东南大学学报》;20110131;第41卷(第1期);第37-40页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108873401A (en) | 2018-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108873401B (en) | Liquid crystal display response time prediction method based on big data | |
| CN109325550B (en) | No-reference image quality evaluation method based on image entropy | |
| CN107845087A (en) | The detection method and system of the uneven defect of liquid crystal panel lightness | |
| CN109598220B (en) | People counting method based on multi-input multi-scale convolution | |
| CN109087277B (en) | Method for measuring PM2.5 of fine air particles | |
| CN110751644B (en) | Road surface crack detection method | |
| CN109117774B (en) | Multi-view video anomaly detection method based on sparse coding | |
| CN110838090B (en) | Backlight diffusion method for image processing based on residual error network | |
| CN106651829B (en) | A kind of non-reference picture method for evaluating objective quality based on energy and texture analysis | |
| CN116340796B (en) | Time sequence data analysis method, device, equipment and storage medium | |
| CN114612714A (en) | Curriculum learning-based non-reference image quality evaluation method | |
| Wiedemann et al. | Disregarding the big picture: Towards local image quality assessment | |
| CN114863348A (en) | Video target segmentation method based on self-supervision | |
| Chi et al. | Blind tone mapped image quality assessment with image segmentation and visual perception | |
| CN114677377A (en) | Display screen defect detection method, training method, device, equipment and medium | |
| Yang et al. | Unsupervised blind image quality assessment via joint spatial and transform features | |
| CN110400307A (en) | A kind of screen picture method for evaluating quality based on area differentiation | |
| CN112861443B (en) | Advanced learning fault diagnosis method integrated with priori knowledge | |
| CN113313682A (en) | No-reference video quality evaluation method based on space-time multi-scale analysis | |
| CN113255789B (en) | Video quality evaluation method based on confrontation network and multi-tested electroencephalogram signals | |
| CN108776958A (en) | Mix the image quality evaluating method and device of degraded image | |
| CN109741313A (en) | The non-reference picture quality appraisement method of independent component analysis and convolutional neural networks | |
| Wu et al. | Effective quality metric for contrast-distorted images based on SVD | |
| CN111325720B (en) | Multi-visual feature integrated reference-free super-resolution image quality evaluation method | |
| CN113435464B (en) | Abnormal data detection method and device, electronic equipment and computer storage medium |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230724 Address after: East side of 1st floor, Building 6, Ruiyun, No. 99 Furong Middle Third Road, Xishan District, Wuxi City, Jiangsu Province, 214191 Patentee after: Wuxi Qigong Data Technology Co.,Ltd. Address before: 710071 Taibai South Road, Yanta District, Xi'an, Shaanxi Province, No. 2 Patentee before: XIDIAN University |