CN112700425B - Determination method for X-ray image quality of power equipment - Google Patents
Determination method for X-ray image quality of power equipment Download PDFInfo
- Publication number
- CN112700425B CN112700425B CN202110017751.4A CN202110017751A CN112700425B CN 112700425 B CN112700425 B CN 112700425B CN 202110017751 A CN202110017751 A CN 202110017751A CN 112700425 B CN112700425 B CN 112700425B
- Authority
- CN
- China
- Prior art keywords
- ray
- image
- score
- network
- quality
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 39
- 238000001514 detection method Methods 0.000 claims abstract description 38
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000011176 pooling Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 7
- 230000008485 antagonism Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 4
- 238000012549 training Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000013441 quality evaluation Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 208000003464 asthenopia Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a judging method for the quality of X-ray images of power equipment, which is characterized in that through collecting pictures shot by X-rays to be scored, scoring is carried out on each picture shot by the X-rays by combining international standards on the basis of generating an countermeasure network and a target detection network, so as to obtain MOS values; preprocessing a picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network taking VGGNET as a framework, and calculating the quality fraction of the picture shot by the X-ray to be scored through the convolutional neural network; and calculating the final score of the picture shot by the X-ray through the MOS value and the mass score. The method provided by the invention utilizes the generated countermeasure network, the target detection network and the convolutional neural network taking VGGNET as a framework to judge the quality of different X-ray images, can effectively solve the problems existing in the existing X-ray image quality judging method, and improves the diagnosis efficiency and reliability of the detection images.
Description
Technical Field
The invention relates to the field of power equipment, in particular to a method for judging the quality of X-ray images of power equipment.
Background
The quality of the image obtained by the X-ray digital imaging is directly related to automatic detection and intelligent interpretation of the subsequent defects, whether the internal structure of the power equipment can be accurately and intuitively displayed is related to whether a detection person can accurately judge a detection result, the detection person is seriously dependent on the technical level of the detection person, the currently developed X-ray detection robot cannot autonomously judge whether the quality of a shot image meets the diagnosis requirement or not, the intelligent degree is seriously influenced, and the current X-ray image quality judging technology has the following defects: 1. the business is many, and the expert lacks: because of uneven shooting technology of each detector, the traditional X-ray image quality judging technology needs a large number of professional technicians to judge the image quality, has more services and low efficiency, and is also very short of the professional technicians. 2. The intelligent level is low, and the cost of labor is high: the remote control of more than 200 meters is not realized by the X-ray intelligent detection robot developed at present, and the shooting system does not have a shooting quality recognition function yet, so that the image quality needs to be judged manually, the time and the labor are wasted, and the problem is encountered or the assistance of a professional technician at a far end needs to be requested. 3. The current X-ray image quality judging method is time-consuming and labor-consuming, errors are easy to occur when the service volume is large, meanwhile, the problem of visual fatigue can occur when the image quality is judged manually, and the judging result is influenced. 4. At present, the identification technology of X-ray detection images in the medical industry is fast in development, but human body structures and tissue components are relatively fixed, the power equipment structure and materials are complex, the detection in a laboratory cannot be realized, on-site detection is needed, influence factors of the quality of the X-ray detection images are more, and great difficulty is brought to artificial intelligent identification of the image quality.
Therefore, how to design a method for determining the quality of an X-ray image of an electrical device with low cost, high efficiency and high reliability is a problem to be solved.
Disclosure of Invention
The invention provides a judging method for the X-ray image quality of power equipment, which aims to solve the problems of high cost, low efficiency and low reliability of the existing judging method.
The invention provides a judging method for the X-ray image quality of power equipment, which comprises the following steps:
S1: acquiring pictures shot by X-rays to be scored, scoring each picture shot by the X-rays based on a generated countermeasure network and a target detection network by combining with international standards, and obtaining an MOS value;
S2: preprocessing a picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network taking VGGNET as a framework, and calculating by the convolutional neural network taking VGGNET as the framework to obtain the mass fraction of the picture shot by the X-ray to be scored;
s3: and calculating the final score of the picture shot by the X-ray through the MOS value and the mass score, wherein the final score of the picture shot by the X-ray is used for judging the quality of the X-ray image.
Optionally, the collecting the pictures of the X-ray to be scored, based on generating the countermeasure network and the target detection network, scoring each of the pictures of the X-ray based on the international standard, and obtaining the MOS value includes the following steps:
s11: collecting pictures shot by X-rays to be scored, and generating X-ray images with different distortion degrees through an antagonism network;
s12: the X-ray images with different distortion degrees and the original data form a data set;
S13: detecting the double-wire position of the X-ray image in the data set through a target detection network to obtain an image with a detection tag;
s14: and scoring the image with the detection label by combining with the international standard to obtain the MOS value.
Optionally, the preprocessing the image of the X-ray to be scored, inputting the image of the X-ray to a convolutional neural network with VGGNET as a skeleton, and calculating the mass fraction of the image of the X-ray to be scored by the convolutional neural network with VGGNET as the skeleton includes the following steps:
s21: preprocessing a picture shot by the X-ray to be scored, and intercepting an image block in the image;
S22: inputting the image block into a convolutional neural network taking VGGNET as a framework, and carrying out feature extraction on the image block by using the convolutional neural network taking VGGNET as the framework to obtain a feature vector;
s23: calculating weights and scores of the image blocks based on the feature vectors;
S24: and calculating the quality score of the image shot by the X-ray to be scored through weighting pooling based on the weight and the score of the image block.
Optionally, after the weighting and scoring based on the image blocks are pooled by weighting, calculating correction network parameters before calculating the quality score of the image shot by the X-ray to be scored, and feeding the correction network parameters back to the convolutional neural network taking VGGNET as a framework.
Optionally, the generating an countermeasure network includes a generator network and a discriminator network.
Alternatively, the MOS value is calculated by:
the average score of a single image is calculated as follows:
Where M i represents the average score of the ith image, S k,i represents the score of the kth tester for the ith image;
Judging whether the average score of a single image is in a 95% confidence interval, if not, the data is abnormal, deleting the data, otherwise, reserving the data;
and after removing the abnormal data, carrying out iterative computation again until the obtained data are within a 95% confidence interval, and taking a data mean value within the 95% confidence interval as an MOS value.
Optionally, the final score of the X-ray taken picture is an average of the MOS value and the mass score.
The invention provides a judging method for the X-ray image quality of power equipment, which comprises the following steps: acquiring pictures shot by X-rays to be scored, scoring each picture shot by the X-rays based on a generated countermeasure network and a target detection network by combining with international standards, and obtaining an MOS value; preprocessing a picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network taking VGGNET as a framework, and calculating by the convolutional neural network taking VGGNET as the framework to obtain the mass fraction of the picture shot by the X-ray to be scored; and calculating the final score of the picture shot by the X-ray through the MOS value and the mass score, wherein the final score of the picture shot by the X-ray is used for judging the quality of the X-ray image. The method provided by the invention utilizes the generated countermeasure network, the target detection network and the convolutional neural network taking VGGNET as a framework to judge the quality of different X-ray images, can effectively solve the problems existing in the existing X-ray image quality judging method, and improves the diagnosis efficiency and reliability of the detection images.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a flow chart of a method for determining the quality of an X-ray image of an electrical device according to the present invention;
FIG. 2 is a flowchart of S1 of a method for determining the quality of an X-ray image of an electrical device according to the present invention;
FIG. 3 is a flowchart of S2 of a method for determining the quality of an X-ray image of an electrical device according to the present invention;
FIG. 4 is a partial flow chart of a method for determining the quality of an X-ray image of an electrical device according to the present invention;
FIG. 5 is a flow chart of a method for determining X-ray image quality of an electrical device for generating an countermeasure network according to the present invention;
Fig. 6 is a schematic diagram of a convolutional neural network used for a method for determining the quality of an X-ray image of an electrical device according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made more fully hereinafter with reference to the accompanying drawings, in which it is shown, however, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways other than those described again, and persons skilled in the art will readily appreciate that the present invention is not limited to the specific embodiments disclosed below.
Referring to fig. 1, the present invention provides a method for determining the quality of an X-ray image of an electrical device, comprising the steps of:
S1: acquiring pictures shot by X-rays to be scored, scoring each picture shot by the X-rays based on a generated countermeasure network and a target detection network by combining with international standards, and obtaining an MOS value;
S2: preprocessing a picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network taking VGGNET as a framework, and calculating by the convolutional neural network taking VGGNET as the framework to obtain the mass fraction of the picture shot by the X-ray to be scored;
s3: and calculating the final score of the picture shot by the X-ray through the MOS value and the mass score, wherein the final score of the picture shot by the X-ray is used for judging the quality of the X-ray image.
Referring to fig. 2, optionally, the collecting the pictures of the X-ray shots to be scored, based on generating an countermeasure network and a target detection network, scoring each of the pictures of the X-ray shots in combination with international standards, and obtaining the MOS value includes the following steps:
s11: collecting pictures shot by X-rays to be scored, and generating X-ray images with different distortion degrees through an antagonism network;
s12: the X-ray images with different distortion degrees and the original data form a data set;
S13: detecting the double-wire position of the X-ray image in the data set through a target detection network to obtain an image with a detection tag;
s14: and scoring the image with the detection label by combining with the international standard to obtain the MOS value.
Referring to fig. 3, optionally, preprocessing the image of the X-ray to be scored, inputting the image of the X-ray to a convolutional neural network with VGGNET as a skeleton, and calculating the mass fraction of the image of the X-ray to be scored by using the convolutional neural network with VGGNET as the skeleton includes the following steps:
s21: preprocessing a picture shot by the X-ray to be scored, and intercepting an image block in the image;
S22: inputting the image block into a convolutional neural network taking VGGNET as a framework, and carrying out feature extraction on the image block by using the convolutional neural network taking VGGNET as the framework to obtain a feature vector;
s23: calculating weights and scores of the image blocks based on the feature vectors;
S24: and calculating the quality score of the image shot by the X-ray to be scored through weighting pooling based on the weight and the score of the image block.
Optionally, after the weighting and scoring based on the image blocks are pooled by weighting, calculating correction network parameters before calculating the quality score of the image shot by the X-ray to be scored, and feeding the correction network parameters back to the convolutional neural network.
Optionally, the generating an countermeasure network includes a generator network and a discriminator network.
Alternatively, the MOS value is calculated by:
the average score of a single image is calculated as follows:
Where M i represents the average score of the ith image, S k,i represents the score of the kth tester for the ith image;
Judging whether the average score of a single image is in a 95% confidence interval, if not, the data is abnormal, deleting the data, otherwise, reserving the data;
and after removing the abnormal data, carrying out iterative computation again until the obtained data are within a 95% confidence interval, and taking a data mean value within the 95% confidence interval as an MOS value.
Optionally, the final score of the X-ray taken picture is an average of the MOS value and the mass score.
Convolutional neural networks are part of the field of artificial intelligence research and have wide application in many fields, and can automatically learn features from data and generalize results to unknown data.
Referring to fig. 4, a specific embodiment is as follows:
A plurality of X-ray shooting images are collected, X-ray images with different distortion degrees are generated by utilizing a generating countermeasure network, the generating countermeasure network and the original data form a data set together, and the generating countermeasure network consists of two parts, namely a generator and a discriminator. In the training process, the two are required to be well matched. The generation of the countermeasure network GAN has a strong capability in picture generation, and includes two parts, namely a generator network and a discriminator network. The generator inputs random noise for learning real distribution, the discriminator inputs the generated sample and the real sample, and judges the true or false of the generated image, the whole generated countermeasure network learns the real sample distribution in continuous countermeasure training, and the true data is generated.
After the data is generated, the positions of the double wires are detected through the target detection network, and an image with the detection tag is generated. By generating the countermeasure network, over 2000X-ray images of different quality are finally generated, and the X-ray images and the original data together form a required data set. And then preprocessing the data of the data set, and manually scoring the shot image according to expert experience (manual operation) to obtain MOS. The calculation of MOS is specifically divided into two steps, firstly, the average division of a single image is calculated, and the formula is as follows:
Where M i represents the average score of the ith image and S k,i represents the score of the kth tester for the ith image.
Some outliers are deleted successively. Firstly, calculating a confidence interval of M i at 95%, regarding data outside the confidence interval as abnormal data, removing the abnormal data, and performing iterative calculation again until the obtained data are all within the confidence interval. And finally taking the average value of the data as a final result. These results are ground truth as image quality.
In the model training stage, 32X32 image blocks are densely intercepted in the image, and then a convolutional neural network with VGGNET as a skeleton network is adopted to extract depth features of the image blocks to obtain depth representation vectors of the image blocks. Once the depth representation vector of the image block is obtained, a fusion sub-network is constructed, and based on the extracted features, on the one hand, the weights of the image block are learned, and on the other hand, the quality evaluation score of the image block is estimated. Finally, in order to obtain the quality evaluation scores of the whole image, a global pooling layer is added, and the quality evaluation scores of all the image blocks are subjected to weighted pooling operation to obtain a final image quality evaluation score. And adopting the average absolute error as a loss function optimization criterion, and feeding back and correcting network parameters. Finally, a spearman rank correlation coefficient (SROCC) and a Pearson Linear Correlation Coefficient (PLCC) are selected as measurement standard measurement model training results. The whole learning process is a supervised regression learning process.
The loss function is used for measuring the degree of difference between the image quality score predicted by the convolutional neural network and ground truth, if the difference is large, the back propagation is carried out to correct the parameters of the convolutional neural network, the loss function L is gradually reduced along with the training, and the prediction score is more and more accurate. The loss function here directs the correction of network parameters. Let q c represent the prediction score, q t represent ground truth, the loss function expression is:
L＝|qc-qt|
to increase the nonlinearity of the algorithm, an activation function is added to the convolutional neural network, which should have the following properties: non-linearities; continuous and microminiaturizable; the range is preferably unsaturated; monotonicity; approximately linear at the origin. The activating function selects ReLU, and the specific expression is:
f(x)＝max(0,x)
and x represents the output of the convolution layer, the sparsity of the convolution neural network can be well increased by the activation function, and the greater the sparsity is, the more representative the extracted features are, and the better the generalization capability is.
In general, distortion in an area that attracts the attention of a viewer is more serious than interference in other areas. This results in the idea of combining the visual saliency model with the IQM by weighting the local quality score y i of region i with the corresponding image block weight w i to the overall image quality Q. Therefore, in the convolutional neural network training process, on one hand, the weight of the image block is obtained through training, and on the other hand, the score of the image block is obtained through training, and the final evaluation score of the final whole image is calculated as follows:
Wherein w i represents the weight of the image block i, y i represents the score of the image block i, the final score of the whole image is obtained by solving the weighted average, and the quality of the image is judged by the score. The area that attracts the viewer's attention is weighted relatively more heavily, whereas the area is weighted relatively less heavily.
The invention designs a generating countermeasure network, and a flow chart refers to fig. 5, and X-ray image diagrams with different qualities are obtained through repeated countermeasure training of a generator and a discriminator.
The convolutional neural network designed by the invention, a network diagram referring to fig. 6, comprises 10 convolutional layers, 5 pooling layers and two fully connected layers. A pooling operation is performed for extracting the main features every two convolutions. The image features are converted into tensors by multiple convolutions. And finally, outputting the weights and scores of the image blocks through the two full-connection layers.
The invention expands the data volume by generating the countermeasure network, inputs the processed data into the neural network to extract the characteristics, and obtains the final weighted score through the operations such as characteristic fusion, pooling and the like.
The invention judges the image quality according to the score output by the neural network, the score is high, the image quality is good, the image quality is relatively poor if the score is low, and the score interval set by the invention is [1,10]. The image score predicted by the network is high and low, so that a plurality of images with poor quality can be effectively screened out, and images with good quality can be left.
The invention utilizes the convolutional neural network to automatically judge the quality of the X-ray detection image, and automatically evaluates the shooting quality of the image according to the predictive score of the convolutional neural network. The algorithm can be suitable for various environments and various different types of photographed images, can keep good stability for images which are not seen by a training set, and gives a predictive score which accords with a human visual system.
The advantages created by the invention have the following points:
1. The method has high efficiency, effectively solves the problems of more time and labor consumption of the service, reduces the cost of manual identification, and can effectively utilize various data acquired before to carry out iterative optimization of the algorithm model.
2. The method has high reliability, effectively solves the problem that the accuracy is reduced due to visual fatigue during manual judgment, and the judgment result is seriously dependent on the technical level of personnel, and is suitable for various image types and various environments.
3. The method adopts the generation of the expanded data volume of the countermeasure network, improves the generalization capability of the network, and simultaneously solves the problems of insufficient data volume and insufficient training samples under specific conditions.
4. The definition of the X-ray image of the equipment with complex structure and material can be measured under different thicknesses.
The invention provides a judging method for the X-ray image quality of power equipment, which comprises the following steps: acquiring pictures shot by X-rays to be scored, scoring each picture shot by the X-rays based on a generated countermeasure network and a target detection network by combining with international standards, and obtaining an MOS value; preprocessing the picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network, and calculating the quality fraction of the picture shot by the X-ray to be scored through the convolutional neural network; and calculating the final score of the picture shot by the X-ray through the MOS value and the mass score, wherein the final score of the picture shot by the X-ray is used for judging the quality of the X-ray image. The method provided by the invention utilizes the generation countermeasure network, the target detection network and the convolution neural network to judge the quality of different X-ray images, can effectively solve the problems existing in the existing X-ray image quality judging method, and improves the diagnosis efficiency and reliability of the detection images.
The foregoing is merely exemplary of the invention and it will be appreciated by those skilled in the art that variations and modifications may be made without departing from the principles of the invention, and it is intended that the invention also be limited to the specific embodiments shown.
Claims (5)
1. A method for determining the quality of an X-ray image of an electrical device, the method comprising the steps of:
S1: acquiring pictures shot by X-rays to be scored, scoring each picture shot by the X-rays based on a generated countermeasure network and a target detection network by combining with international standards, and obtaining an MOS value;
The method for acquiring the pictures shot by the X-rays to be scored, based on the generation of an countermeasure network and a target detection network, and combining with international standards, scoring each picture shot by the X-rays to obtain an MOS value comprises the following steps:
s11: collecting pictures shot by X-rays to be scored, and generating X-ray images with different distortion degrees through an antagonism network;
s12: the X-ray images with different distortion degrees and the original data form a data set;
S13: detecting the double-wire position of the X-ray image in the data set through a target detection network to obtain an image with a detection tag;
s14: scoring the image with the detection tag by combining with international standards to obtain an MOS value;
S2: preprocessing a picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network taking VGGNET as a framework, and calculating by the convolutional neural network taking VGGNET as the framework to obtain the mass fraction of the picture shot by the X-ray to be scored;
the preprocessing of the picture shot by the X-ray to be scored, inputting the picture shot by the X-ray into a convolutional neural network taking VGGNET as a framework, and calculating the mass fraction of the picture shot by the X-ray to be scored by the convolutional neural network taking VGGNET as the framework, wherein the mass fraction comprises the following steps:
s21: preprocessing a picture shot by the X-ray to be scored, and intercepting an image block in the image;
S22: inputting the image block into a convolutional neural network taking VGGNET as a framework, and carrying out feature extraction on the image block by using the convolutional neural network taking VGGNET as the framework to obtain a feature vector;
s23: calculating weights and scores of the image blocks based on the feature vectors;
S24: based on the weight and the score of the image block, calculating to obtain the quality score of the image shot by the X-ray to be scored through weighting pooling;
s3: and calculating the final score of the picture shot by the X-ray through the MOS value and the mass score, wherein the final score of the picture shot by the X-ray is used for judging the quality of the X-ray image.
2. The method for determining the quality of an X-ray image of an electrical device according to claim 1, wherein after the weighting and scoring based on the image blocks, calculating the quality score of the image of the X-ray to be scored before calculating the quality score further comprises calculating a correction network parameter, and feeding the correction network parameter back to the VGGNET convolutional neural network.
3. The method for determining the quality of an X-ray image of a power device of claim 1, wherein the generating an countermeasure network includes a generator network and a discriminator network.
4. The method for determining the quality of an X-ray image of an electrical device according to claim 1, wherein the MOS value is calculated by:
the average score of a single image is calculated as follows:
Where M i represents the average score of the ith image, S k,i represents the score of the kth tester for the ith image;
Judging whether the average score of a single image is in a 95% confidence interval, if not, the data is abnormal, deleting the data, otherwise, reserving the data;
and after removing the abnormal data, carrying out iterative computation again until the obtained data are within a 95% confidence interval, and taking a data mean value within the 95% confidence interval as an MOS value.
5. The method for determining the quality of an X-ray image of a power device according to claim 1, wherein the final score of the X-ray taken picture is an average of a MOS value and a mass score.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110017751.4A CN112700425B (en) | 2021-01-07 | 2021-01-07 | Determination method for X-ray image quality of power equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110017751.4A CN112700425B (en) | 2021-01-07 | 2021-01-07 | Determination method for X-ray image quality of power equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112700425A CN112700425A (en) | 2021-04-23 |
| CN112700425B true CN112700425B (en) | 2024-04-26 |
Family
ID=75514991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110017751.4A Active CN112700425B (en) | 2021-01-07 | 2021-01-07 | Determination method for X-ray image quality of power equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112700425B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107743225A (en) * | 2017-10-16 | 2018-02-27 | 杭州电子科技大学 | It is a kind of that the method for carrying out non-reference picture prediction of quality is characterized using multilayer depth |
| CN108648188A (en) * | 2018-05-15 | 2018-10-12 | 南京邮电大学 | A kind of non-reference picture quality appraisement method based on generation confrontation network |
| CN109215028A (en) * | 2018-11-06 | 2019-01-15 | 福州大学 | A kind of multiple-objection optimization image quality measure method based on convolutional neural networks |
| WO2019047949A1 (en) * | 2017-09-08 | 2019-03-14 | 众安信息技术服务有限公司 | Image quality evaluation method and image quality evaluation system |
| CN109559310A (en) * | 2018-11-30 | 2019-04-02 | 国网山东省电力公司电力科学研究院 | Power transmission and transformation inspection image quality evaluating method and system based on conspicuousness detection |
| CN109754391A (en) * | 2018-12-18 | 2019-05-14 | 北京爱奇艺科技有限公司 | A kind of image quality evaluating method, device and electronic equipment |
| CN110570353A (en) * | 2019-08-27 | 2019-12-13 | 天津大学 | Dense connection generation countermeasure network single image super-resolution reconstruction method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9741107B2 (en) * | 2015-06-05 | 2017-08-22 | Sony Corporation | Full reference image quality assessment based on convolutional neural network |
-
2021
- 2021-01-07 CN CN202110017751.4A patent/CN112700425B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019047949A1 (en) * | 2017-09-08 | 2019-03-14 | 众安信息技术服务有限公司 | Image quality evaluation method and image quality evaluation system |
| CN107743225A (en) * | 2017-10-16 | 2018-02-27 | 杭州电子科技大学 | It is a kind of that the method for carrying out non-reference picture prediction of quality is characterized using multilayer depth |
| CN108648188A (en) * | 2018-05-15 | 2018-10-12 | 南京邮电大学 | A kind of non-reference picture quality appraisement method based on generation confrontation network |
| CN109215028A (en) * | 2018-11-06 | 2019-01-15 | 福州大学 | A kind of multiple-objection optimization image quality measure method based on convolutional neural networks |
| CN109559310A (en) * | 2018-11-30 | 2019-04-02 | 国网山东省电力公司电力科学研究院 | Power transmission and transformation inspection image quality evaluating method and system based on conspicuousness detection |
| CN109754391A (en) * | 2018-12-18 | 2019-05-14 | 北京爱奇艺科技有限公司 | A kind of image quality evaluating method, device and electronic equipment |
| CN110570353A (en) * | 2019-08-27 | 2019-12-13 | 天津大学 | Dense connection generation countermeasure network single image super-resolution reconstruction method |
Non-Patent Citations (1)
| Title |
|---|
| 生成式对抗网络的图像超分辨率重建;王志强等;西安工业大学学报;第40卷(第1期);第102-108页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112700425A (en) | 2021-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921051B (en) | Pedestrian attribute identification network and technology based on cyclic neural network attention model | |
| CN111814741B (en) | Method for detecting embryo-sheltered pronucleus and blastomere based on attention mechanism | |
| CN110598736A (en) | Power equipment infrared image fault positioning, identifying and predicting method | |
| CN110378232B (en) | Improved test room examinee position rapid detection method of SSD dual-network | |
| CN114548278A (en) | In-service tunnel lining structure defect identification method and system based on deep learning | |
| CN115013298B (en) | Real-time performance online monitoring system and monitoring method of sewage pump | |
| CN111141653B (en) | Tunnel leakage rate prediction method based on neural network | |
| CN111161224A (en) | Casting internal defect grading evaluation system and method based on deep learning | |
| CN114359695A (en) | Insulator breakage identification method based on uncertainty estimation | |
| CN114154530A (en) | Training method and device for atrial fibrillation detection model of electrocardio timing signals | |
| CN112700425B (en) | Determination method for X-ray image quality of power equipment | |
| CN115830302B (en) | Multi-scale feature extraction fusion power distribution network equipment positioning identification method | |
| CN116662800A (en) | Rolling bearing fault diagnosis method based on self-adaptive attention mechanism | |
| CN110738208A (en) | efficient scale-normalized target detection training method | |
| CN112861670B (en) | Transmission line hardware detection method and system | |
| CN112488125B (en) | Reconstruction method and system based on high-speed visual diagnosis and BP neural network | |
| CN112508946B (en) | Cable tunnel anomaly detection method based on antagonistic neural network | |
| CN107341798A (en) | High Resolution SAR image change detection method based on global local SPP Net | |
| CN110223295B (en) | Significance prediction method and device based on deep neural network color perception | |
| CN117079416B (en) | Multi-person 5D radar falling detection method and system based on artificial intelligence algorithm | |
| CN117491357B (en) | Quality monitoring method and system for paint | |
| CN116719002B (en) | Quantitative precipitation estimation method, quantitative precipitation estimation device, electronic equipment and computer storage medium | |
| CN116148069B (en) | Bearing tensile strength detection device and detection method | |
| CN112396648B (en) | Target identification method and system capable of positioning mass center of target object | |
| CN114418941B (en) | Defect diagnosis method and system based on detection data of power inspection equipment |
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 |