CN110083637B - Bridge disease rating data-oriented denoising method - Google Patents
Bridge disease rating data-oriented denoising method Download PDFInfo
- Publication number
- CN110083637B CN110083637B CN201910327313.0A CN201910327313A CN110083637B CN 110083637 B CN110083637 B CN 110083637B CN 201910327313 A CN201910327313 A CN 201910327313A CN 110083637 B CN110083637 B CN 110083637B
- Authority
- CN
- China
- Prior art keywords
- samples
- data set
- sample
- conflict
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention relates to a bridge disease rating data oriented denoising method, which is characterized by comprising the following steps: firstly, deleting the characteristics of the original data set, wherein the characteristic value sequence relation is difficult to distinguish; secondly, respectively comparing every two samples with different labels in the data set to obtain a conflict pair set consisting of samples with label conflicts; then, sequencing the samples from high to low according to the times of the samples appearing in the conflict pair set, sequentially calculating the outline coefficients of the samples t% before ranking, deleting the samples with the outline coefficients smaller than epsilon from the conflict pair set and the data set, sequencing the samples again according to the times of the samples appearing in the conflict pairs, calculating the outline coefficients and deleting the samples until the outline coefficients of the samples t% before ranking are not smaller than epsilon; and finally, obtaining a data set with noise removed, wherein the noise removal method can effectively improve the accuracy of the classification of the bridge disease data.
Description
Technical Field
The invention belongs to the technical field of data processing, and aims to design a noise removal method for rating data of bridge diseases.
Background
The improvement is opened, and the highway bridge of China is in the period of great development of the coming major construction. At present, the total number of highway bridges in China is close to 80 thousands, and the number and scale of the bridges are at the top of the world. However, the number of bridges in service, which are stepped into the maintenance period in China, is increasing. According to incomplete statistics, about 40 percent of bridges in service in China are in service for more than 20 years, the technology level of three or four types of bridge with diseases is as high as 30 percent, even more than 10 ten thousand bridges are dangerous bridges, and potential safety hazards are not ignored. The bridge disease condition rating is the basis for road and bridge management and maintenance. The traditional manual rating method is time-consuming and labor-consuming, is not high in accuracy, and urgently needs to automatically rate the in-service bridge diseases by using a machine learning technology. The existing bridge data set often contains a large amount of label noise data, and in order to effectively improve the prediction performance of a machine learning method for bridge disease rating, the label noise data in the original data set needs to be filtered. Currently, there are two main types of label noise filtering methods: (1) Directly adapting the classification algorithm to reduce the influence of label noise on algorithm performance; (2) And classifying and voting the data by adopting a pre-classifier, and then filtering out part of suspected noise data. However, the above two methods are not ideal in filtering noise data of the bridge disease label.
In view of the above, it is desirable to design a new tag noise filtering method to solve the above problems.
Disclosure of Invention
In view of the above, the invention discloses a bridge disease rating data denoising method, which effectively improves the prediction performance of a bridge disease rating method based on machine learning. Firstly, deleting characteristics of which the sequence relation of characteristic values is difficult to distinguish in an original data set to obtain a new data set, wherein the characteristic values of all the characteristics in the data set have the sequence relation, and the original data set comprises basic information of all bridges, bridge defect information of all kinds and corresponding bridge defect grade labels; secondly, comparing all samples in the data set pairwise, and forming a conflict pair by the two samples which conflict with each other, wherein all conflict pairs in the data set form a conflict pair set; thirdly, counting the occurrence times of the samples in the conflict pair set, and sequencing the occurrence times of the samples; fourthly, removing samples with a certain proportion from high to low in sequence according to the outline coefficient and the sample frequency of the samples to obtain a new data set; and finally, training the original data set and the new data set respectively by using a stacking method to obtain two models, evaluating and verifying bridge disease grade prediction performances of the two models to verify the effectiveness of the denoising method, and obtaining a relatively clean data set if the effectiveness is confirmed.
The technical scheme of the invention is realized in the form that: a noise removal method for rating data of bridge diseases comprises the steps of firstly obtaining a conflict pair set through pairwise comparison of samples, then conducting noise data elimination according to the number of times of the samples appearing in the conflict pair set and combining with a contour coefficient of the samples to obtain a filtered data set, then training models on an original data set and a filtered new data set respectively by using the same machine learning method, and finally comparing the prediction performances of the two models, wherein the specific steps are as follows:
s1, preprocessing data in an original data set to obtain a data set W 1 Through the pair W 1 Removing the characteristics without the complete sequence relation to obtain a new data set W 2 ；
S2, based on the data set W 2 According to the characteristic a i Characteristic value a of i，j Comparing samples with different labels pairwise to construct conflict pairs c i ；
S3, based on conflict pair c i Construct conflicts set C = { C 1 ，c 2 ，...，c N N is the total number of conflict pairs contained in conflict set C;
s4, counting samples S in the conflict set C k Frequency of occurrence f k Obtaining dictionary D = { s = { [ s ] k ：f k }。
S5, sequencing the samples in the dictionary D from high to low according to frequency;
s6, aiming at the samples t% before the sorting, in the data set W 2 In which contour coefficients s (k) are calculated, samples s for which s (k) is smaller than epsilon are deleted k Obtaining a new filtered data set W 3 At the same time, the set C of the deletion conflict pairs contains the suspected noise sample s k The conflict pair of (1);
s7, repeating S4, S5 and S6 until no samples with S (i) smaller than epsilon exist in the step S62;
s8, using the same machine learning method and based on the data set W 1 And W 3 Respectively train out model M 1 And M 3 Evaluating and comparing model M 3 The predicted performance of (2).
Further, step S1 includes:
s11, based on the data set W 1 Complementing by using the value of the most similar sample by using a hot card filling methodThe measurement method of the most similar sample without characteristic value is
Wherein a is i，j For a feature value of a jth feature of an ith sample in a data set>
Is a missing characteristic value;
s12, deleting useless features which do not affect the label value;
s13, deleting the data set W 1 The medium characteristic value has no characteristic of the complete sequence relation to obtain a data set W 2 。
Further, step S2 includes:
s21, data set W 2 The feature set of (A) = { a = 1 ，a 2 ，...，a Ni Is the data set W 2 The total number of features of (a);
s22, data set characteristic a i Set of characteristic values of
Wherein N is a Is a data set W 2 Is the total number of samples of, and is also characteristic a i The total number of characteristic values of;
s23, firstly judging the labels of the two samples, if the labels are the same, skipping to compare the two samples, and if the labels are different, comparing the feature values of the two samples under all the characteristics in a one-to-one correspondence manner, wherein the calculation formula is as follows:
s24, selecting a first sample, sequentially comparing all the following samples with the first sample according to the mode of the step S23, constructing conflict pairs, sequentially proceeding until the last sample is iterated, then selecting a second sample, sequentially comparing all the following samples with the first sample according to the mode of the step S23, constructing conflict pairs, and sequentially proceeding until the last sample is iterated; similarly, the iteration is stopped until the selected penultimate sample is compared.
Further, step S3 includes:
s31, all conflict pairs constructed in the step S23, construct one conflict set C = { C = 1 ，c 2 ，...，c N N is the total number of conflict pairs of the conflict set C.
Further, step S4 includes:
s41, counting samples S in the conflict pair left element k Number of occurrences f lk ；
S42, counting samples S in the right element of the conflict pair k Number of occurrences f rk ；
S43, calculating the total frequency f k ＝f lk +f rk ；
S44, sampling S k And frequency f of its occurrence k A one-to-one mapping relation between the two is used for constructing a dictionary D = { s = {(s) } k ：f k }，k＝1，2，...，Na。
Further, step S5 includes:
s51, the samples in the dictionary D are processed according to the frequency f k Ordering from high to low.
Further, step S6 includes:
s61, according to the formula
Calculating a contour coefficient, wherein>
Is a sample s k Degree of intra-cluster dissimilarity of (a) i，k Is a sample s k The eigenvalues and rk of the ith characteristic are samples s k The label of (1); b (k) = min { b (k) 1 ，b(k) 2 ，...b(k) n Is the sample s k The degree of dissimilarity between clusters of,
is a sample s k Dissimilarity to the nth cluster (i.e., the class labeled rn);
s62, if the sample S k If the contour coefficient s (k) < epsilon, the sample s is recorded k Is counted and considered as a suspected noise sample in the data set W 2 In which the sample s is deleted k To obtain a new data set W 3 。
S63, deleting the suspected noise sample S containing the step S62 in the conflict pair set C k The pair of conflicts of (c).
Further, step S7 includes:
s71, repeating S4, S5, and S6 until there are no samples where S (i) is smaller than ∈ (in this patent, ∈ is set to 0) in step S62.
Further, step S8 includes:
s81, respectively combining the data sets W 1 And a new data set W 3 Dividing the test sample into three parts according to the same proportion, namely a training set, a verification set and a test set;
s82, respectively applying a stacking method to W 1 And W 3 Train out model M 1 And M 3 ；
S83, comparing and evaluating model M 3 The predicted performance of (2).
After the method strategy is adopted, the method has the advantages that:
(1) The invention designs a completely different noise elimination algorithm aiming at the label noise data appearing in the bridge disease data set, finds the difference of the probability of different samples as the noise data according to the sample conflict times by utilizing the label conflict between the samples, increases the accuracy of label noise filtration and improves the data quality of the data set.
(2) Compared with the traditional method for filtering the label noise by using a classification algorithm, the method has the advantages that the prediction performance of the machine learning model is finally trained by means of the specificity of the internal structure of the data set.
Drawings
The various aspects of the present invention will become more apparent to the reader upon reading the detailed description of the invention in reference to the accompanying drawings, which are some examples of the invention and from which others may be derived by those of ordinary skill in the art without inventive faculty.
Fig. 1 is a schematic flow diagram of an embodiment of the noise removal method for rating data of bridge defects according to the present invention.
Fig. 2 is a detailed flowchart of step S2 in the noise removing method for the rating data of bridge defects according to the embodiment of the present invention.
Fig. 3 is a schematic flowchart of step S4 in the embodiment of the noise removing method for bridge defect rating data according to the present invention.
Fig. 4 is a detailed flowchart of step S6 in the embodiment of the noise removing method for the bridge fault rating data according to the present invention.
Fig. 5 is a detailed flowchart of step S8 in the noise removing method for bridge defect rating data according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
S1, carrying out data preprocessing on data in an original data set to obtain a data set W 1 To W 1 Removing the characteristics of medium or non-full order relation to obtain a data set W 2 。
S11, using a hot card filling method, complementing missing characteristic values by using the values of the most similar samples, wherein the measurement method of the most similar samples is
Wherein ai, j is the characteristic value of the jth characteristic of the ith sample in the data set, which is greater than or equal to>
Is a missing feature value;
s12, deleting useless features which do not affect the label value;
s13, deleting the data set W 1 The medium characteristic value has no characteristic of the complete sequence relation to obtain a data set W 2 。
S2, according to the data set W 2 Based on the feature a i Characteristic value a of i，j Comparing the samples of different labels pairwise to construct a conflict pair c i 。
S21, data set W 2 The feature set of (A) = { a = 1 ，a 2 ，...，a Ni Is the data set W 2 The total number of features of (a);
s22, data set characteristics a i Is D = { a = i，1 ，a i，2 ，...，a i，Na Na is the dataset W 2 Is the total number of samples of, and is also characteristic a i The total number of characteristic values of;
s23, firstly judging the labels of the two samples, if the labels are the same, skipping to compare the two samples, and if the labels are different, comparing the feature values of the two samples under all the characteristics in a one-to-one correspondence manner, wherein the calculation formula is as follows:
S24, selecting a first sample, sequentially comparing all the following samples with the first sample according to the mode of the step S23, constructing a conflict pair, sequentially proceeding until the last sample is iterated, then selecting a second sample, sequentially comparing all the following samples with the first sample according to the mode of the step S23, constructing a conflict pair, and sequentially proceeding until the last sample is iterated; similarly, the iteration is stopped until the selected penultimate sample is compared.
S3, according to the conflict pair c i Construction conflict set C = { C = 1 ，c 2 ，...，c N Where N is the total number of conflict pairs for conflict set C.
S31, combining all conflict pairs constructed in the step S23 into a conflict set C = { C = { (C) } 1 ，c 2 ，...，c N Where N is the total number of conflict pairs for conflict set C.
S4, counting samples S in the conflict set C k Frequency of occurrence f k Obtaining a dictionary D = { s = { s } k ：f k }。
S41, counting samples S in the conflict pair left element k Number of occurrences f lk ；
S42, counting samples S in the right element of the conflict pair k Number of occurrences f rk ；
S43, calculating frequency f k ＝f lk +f rk ；
S44, sampling S k And frequency f of its occurrence k A one-to-one mapping relation between the two is used for constructing a dictionary D = { s = {(s) } k ：f k }，k＝1，2，...，Na。
And S5, sequencing the samples in the dictionary D from high to low according to the frequency.
S51, the samples in the dictionary D are processed according to the frequency f k Ordering from high to low.
S6, aiming at the samples t% at the front of the sorting in the data set W 2 The contour coefficients s (k) are calculated, s (samples s where k is smaller than epsilon) are deleted k To obtain a new filtered data set W 3 And simultaneously deleting suspected noise samples s contained in the conflict pair set C k The conflict pair of (3).
S61 according to the formula
Calculating a contour coefficient, wherein>
Is a sample s k Within clusters of (a) dissimilarity i，k Is a sample s k The eigenvalue and rk of the ith characteristic are samples s k The label of (1); b (k) = min { b (k) 1 ，b(k) 2 ，...b(k) n Is the sample s k The degree of dissimilarity between clusters of,
is a sample s k Dissimilarity to the nth cluster (i.e., the class labeled rn);
s62, if the sample S k If the contour coefficient s (k) < epsilon, the sample s is recorded k Is counted and considered as a suspected noise sample in the data set W 2 In which the sample s is deleted k To obtain a new data set W 3 ；
S63, deleting the suspected noise sample S containing the step S62 in the conflict pair set C k The pair of conflicts of (c).
And S7, repeating S4, S5 and S6 until no samples with S (i) smaller than epsilon exist in the step S62.
S71, repeating S4, S5, and S6 until there are no samples where S (i) is smaller than ∈ (in this patent, ∈ is set to 0) in step S62.
S8, based on the data set W 1 And W 3 Respectively training a model M by using a stacking method 1 And M 3 Evaluating and comparing model M 3 The predicted performance of (2).
S81, respectively combining the data sets W 1 And a new data set W 3 Dividing the test sample into three parts according to the same proportion, and respectively using the three parts as a training set, a verification set and a test set;
s82, respectively based on W by applying the same machine learning algorithm 1 And W 3 Training out model M from data set 1 And M 3 ；
S83, evaluating and comparing the model M 3 The predicted performance of (2).
Hereinbefore, the embodiments of the present invention are described with reference to the drawings. However, those skilled in the art will appreciate that various modifications and substitutions can be made to the specific embodiments of the present invention without departing from the spirit and scope of the invention. Such modifications and substitutions are intended to be within the scope of the present invention as defined by the appended claims.
Claims (8)
1. A noise removal method for bridge disease rating data includes the steps of firstly, comparing sample data pairwise to obtain a conflict pair set, then, conducting noise data elimination according to the number of times of appearance of a sample in the conflict pair set and combining a contour coefficient of the sample to obtain a filtered data set, then, training a model on an original data set and the filtered new data set respectively by using a stacking method, and finally evaluating and comparing prediction performances of the two models to verify the effectiveness of the noise removal method, wherein if the effectiveness is confirmed, a clean data set is obtained, and the specific steps are as follows:
s1, preprocessing data in an original data set to obtain a data set W 1 To W 1 Removing the characteristics of medium and non-full order relation to obtain a data set W 2 The original data set comprises basic information of each bridge, bridge defect information of each type and corresponding bridge defect grade labels;
s2, according to the data set W 2 Based on the feature a i Characteristic value a of i,j Comparing the samples with different labels pairwise to construct a conflict pair c i ；
S3, according to the conflict pair c i Construction conflict set C = { C = 1 ,c 2 ,…,c N N is the total number of conflict pairs of the conflict set C;
s4, counting samples S in the conflict set C k Frequency of occurrence f k Obtaining dictionary D = { s = { [ s ] k ∶f k }；
S5, sequencing the samples in the dictionary D from high to low according to frequency;
s6, sorting the samples with the first t% in the data set W 2 Calculating the contour coefficients s (k), deleting s: (k) Samples s smaller than ε k To obtain a new filtered data set W 3 And simultaneously deleting suspected noise samples s contained in the conflict pair set C k Wherein t is a threshold value for reducing the number of samples to be calculated;
s7, repeating S4, S5 and S6 until no sample with S (i) smaller than epsilon exists in the step S6, wherein the value of epsilon is 0;
s8, in the data set W 1 And W 3 Respectively training out the model M by using the same machine learning algorithm 1 And M 3 Evaluating and verifying the bridge disease grade prediction performance of the two models, and comparing the evaluation model M 3 The predicted performance of (2).
2. The bridge disease data-oriented denoising method according to claim 1, wherein the step S1 specifically comprises:
s11, based on the data set W 1 Using a hot card filling method, complementing the missing characteristic value by using the value of the most similar sample, wherein the measurement method of the most similar sample is
Wherein a is i,j For a feature value of a jth feature of an ith sample in a data set>
For missing feature values, na is the data set W 2 Total number of samples of i 0 Numbering the most similar samples;
s12, deleting useless features which do not affect the label value;
s13, deleting the data set W 1 The medium characteristic value has no characteristic of the complete sequence relation to obtain a data set W 2 。
3. The bridge disease data-oriented denoising method according to claim 1, wherein the step S2 specifically comprises:
s21, data set W 2 The feature set of (A) = { a = 1 ,a 2 ,…,a Ni Ni is the data set W 2 The total number of features of (a);
s22, data set characteristics a i Is D = { a = i,1 ,a i,2 ,…,a i,Na Is the data set W 2 Is the total number of samples of, and is also characteristic a i The total number of characteristic values of;
s23, firstly judging the labels of the two samples, if the labels are the same, skipping to compare the two samples, and if the labels are different, comparing the feature values of the two samples under all the characteristics in a one-to-one correspondence manner, wherein the calculation formula is as follows:
S24, selecting a first sample, sequentially comparing all the following samples with the first sample according to the mode of the step S23, constructing a conflict pair, sequentially proceeding until the last sample is iterated, then selecting a second sample, sequentially comparing all the following samples with the first sample according to the mode of the step S23, constructing a conflict pair, and sequentially proceeding until the last sample is iterated; similarly, the iteration is stopped until the selected penultimate sample is compared.
4. The bridge disease rating data-oriented denoising method according to claim 1, wherein the step S3 specifically comprises:
s31, all conflict pairs constructed in the step S23, construct one conflict set C = { C = 1 ,c 2 ,…,c N N is the total number of conflict pairs of the conflict set C.
5. The bridge disease data-oriented denoising method according to claim 1, wherein the step S4 specifically comprises:
s41, counting samples S in the conflict pair left element k Number of occurrences f lk ；
S42, counting samples S in the right element of the conflict pair k Number of occurrences f rk ；
S43, calculating the total frequency f k ＝f lk +f rk ；
S44, sampling S k And frequency f of its occurrence k A one-to-one mapping relation between the two is used for constructing a dictionary D = { s = {(s) } k :f k }，k＝1,2,…,Na。
6. The bridge disease data-oriented denoising method according to claim 1, wherein the step S5 specifically comprises:
s51, the samples in the dictionary D are processed according to the frequency f k Sorted from high to low.
7. The bridge disease data-oriented denoising method according to claim 1, wherein the step S6 specifically comprises:
s61, according to the formula
Calculating a contour factor, wherein>
Is a sample s k Degree of intra-cluster dissimilarity of (a) i,k As a sample s k The eigenvalue and rk of the ith characteristic are samples s k The label of (1); b (k) = min { b (k) 1 ,b(k) 2 ,…b(k) n Is the sample s k Is not similar between clusters, is greater than>
Is a sample s k Dissimilarity to the nth cluster;
s62, if the sample S k S (k) of the contour coefficient<E, recording the sampleThis s k Is counted and considered as a suspected noise sample in the data set W 2 In which the sample s is deleted k To obtain a new data set W 3 ；
S63, deleting the suspected noise sample S containing the step S62 in the conflict pair set C k The conflict pair of (3).
8. The bridge disease classification data-oriented denoising method according to claim 1, wherein the step S8 specifically comprises:
s81, respectively combining the data sets W 1 And a new data set W 3 Dividing the test sample into three parts according to the same proportion, namely a training set, a verification set and a test set;
s82, respectively applying the same machine learning algorithm based on W 1 And W 3 Training out model M from data set 1 And M 3 ；
S83, evaluating and comparing the model M 3 The predicted performance of (2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910327313.0A CN110083637B (en) | 2019-04-23 | 2019-04-23 | Bridge disease rating data-oriented denoising method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910327313.0A CN110083637B (en) | 2019-04-23 | 2019-04-23 | Bridge disease rating data-oriented denoising method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110083637A CN110083637A (en) | 2019-08-02 |
| CN110083637B true CN110083637B (en) | 2023-04-18 |
Family
ID=67416114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910327313.0A Active CN110083637B (en) | 2019-04-23 | 2019-04-23 | Bridge disease rating data-oriented denoising method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110083637B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111352966A (en) * | 2020-02-24 | 2020-06-30 | 交通运输部水运科学研究所 | Data tag calibration method in autonomous navigation |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5341142A (en) * | 1987-07-24 | 1994-08-23 | Northrop Grumman Corporation | Target acquisition and tracking system |
| CN201023495Y (en) * | 2007-04-29 | 2008-02-20 | 余亚莉 | Vehicle mounted railway traffic passenger transport capacity and operation safety intelligence monitoring and prewarning system |
| CN101639934A (en) * | 2009-09-04 | 2010-02-03 | 西安电子科技大学 | SAR image denoising method based on contour wave domain block hidden Markov model |
| CN102340884A (en) * | 2010-07-26 | 2012-02-01 | 中兴通讯股份有限公司 | Base station and method for preventing random access conflicts |
| CN102509083A (en) * | 2011-11-19 | 2012-06-20 | 广州大学 | Detection method for body conflict event |
| CN103150409A (en) * | 2013-04-08 | 2013-06-12 | 深圳市宜搜科技发展有限公司 | Method and system for recommending user search word |
| CN104077749A (en) * | 2014-06-17 | 2014-10-01 | 长江大学 | Seismic data denoising method based on contourlet transformation |
| CN104966090A (en) * | 2015-07-21 | 2015-10-07 | 公安部第三研究所 | Visual word generation and evaluation system and method for realizing image comprehension |
| CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
| CN108073940A (en) * | 2016-11-18 | 2018-05-25 | 北京航空航天大学 | A kind of method of 3D object instance object detections in unstructured moving grids |
| CN108549904A (en) * | 2018-03-28 | 2018-09-18 | 西安理工大学 | Difference secret protection K-means clustering methods based on silhouette coefficient |
| CN109389143A (en) * | 2018-06-19 | 2019-02-26 | 北京九章云极科技有限公司 | A kind of Data Analysis Services system and method for automatic modeling |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7508990B2 (en) * | 2004-07-30 | 2009-03-24 | Euclid Discoveries, Llc | Apparatus and method for processing video data |
| US10147180B2 (en) * | 2013-12-19 | 2018-12-04 | Axon Dx, Llc | Cell detection, capture and isolation methods and apparatus |
-
2019
- 2019-04-23 CN CN201910327313.0A patent/CN110083637B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5341142A (en) * | 1987-07-24 | 1994-08-23 | Northrop Grumman Corporation | Target acquisition and tracking system |
| CN201023495Y (en) * | 2007-04-29 | 2008-02-20 | 余亚莉 | Vehicle mounted railway traffic passenger transport capacity and operation safety intelligence monitoring and prewarning system |
| CN101639934A (en) * | 2009-09-04 | 2010-02-03 | 西安电子科技大学 | SAR image denoising method based on contour wave domain block hidden Markov model |
| CN102340884A (en) * | 2010-07-26 | 2012-02-01 | 中兴通讯股份有限公司 | Base station and method for preventing random access conflicts |
| CN102509083A (en) * | 2011-11-19 | 2012-06-20 | 广州大学 | Detection method for body conflict event |
| CN103150409A (en) * | 2013-04-08 | 2013-06-12 | 深圳市宜搜科技发展有限公司 | Method and system for recommending user search word |
| CN104077749A (en) * | 2014-06-17 | 2014-10-01 | 长江大学 | Seismic data denoising method based on contourlet transformation |
| CN104966090A (en) * | 2015-07-21 | 2015-10-07 | 公安部第三研究所 | Visual word generation and evaluation system and method for realizing image comprehension |
| CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
| CN108073940A (en) * | 2016-11-18 | 2018-05-25 | 北京航空航天大学 | A kind of method of 3D object instance object detections in unstructured moving grids |
| CN108549904A (en) * | 2018-03-28 | 2018-09-18 | 西安理工大学 | Difference secret protection K-means clustering methods based on silhouette coefficient |
| CN109389143A (en) * | 2018-06-19 | 2019-02-26 | 北京九章云极科技有限公司 | A kind of Data Analysis Services system and method for automatic modeling |
Non-Patent Citations (2)
| Title |
|---|
| "De-noising and event extraction for silicon pore sensors using matrix decomposition";P. Sattigeri 等;《 Sensor Signal Processing for Defence》;20130708;第1-4页 * |
| "我国网络舆情热点话题发现研究综述";游丹丹 等;《现代情报》;20170315;第165-171页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110083637A (en) | 2019-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255506B (en) | Internet financial user loan overdue prediction method based on big data | |
| WO2017143919A1 (en) | Method and apparatus for establishing data identification model | |
| CN110503245B (en) | Prediction method for large-area delay risk of airport flight | |
| CN108665093B (en) | Deep learning-based expressway traffic accident severity prediction method | |
| CN110263166A (en) | Public sentiment file classification method based on deep learning | |
| CN110120218A (en) | Expressway oversize vehicle recognition methods based on GMM-HMM | |
| CN112687349A (en) | Construction method of model for reducing octane number loss | |
| CN111079941B (en) | Credit information processing method, credit information processing system, terminal and storage medium | |
| CN111767398A (en) | Secondary equipment fault short text data classification method based on convolutional neural network | |
| CN111177010B (en) | Software defect severity identification method | |
| CN104850868A (en) | Customer segmentation method based on k-means and neural network cluster | |
| CN110826785A (en) | High-risk road section identification method based on k-medoids clustering and Poisson inverse Gaussian | |
| CN115795351A (en) | Elevator big data risk early warning method based on residual error network and 2D feature representation | |
| CN112800232A (en) | Big data based case automatic classification and optimization method and training set correction method | |
| CN113674862A (en) | Acute renal function injury onset prediction method based on machine learning | |
| CN110807601B (en) | Park road degradation analysis method based on tail cutting data | |
| CN110083637B (en) | Bridge disease rating data-oriented denoising method | |
| CN113516228A (en) | Network anomaly detection method based on deep neural network | |
| CN108090635B (en) | Road performance prediction method based on cluster classification | |
| CN110515836B (en) | Weighted naive Bayes method for software defect prediction | |
| WO2022188425A1 (en) | Deep learning fault diagnosis method integrating prior knowledge | |
| CN114926299A (en) | Prediction method for predicting vehicle accident risk based on big data analysis | |
| CN114139624A (en) | Method for mining time series data similarity information based on integrated model | |
| CN111625578B (en) | Feature extraction method suitable for time series data in cultural science and technology fusion field | |
| CN112884570A (en) | Method, device and equipment for determining model security |
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 |