CN115580721A - Quality detection method, device, equipment and medium for monitoring video of substation equipment - Google Patents
Quality detection method, device, equipment and medium for monitoring video of substation equipment Download PDFInfo
- Publication number
- CN115580721A CN115580721A CN202211461253.XA CN202211461253A CN115580721A CN 115580721 A CN115580721 A CN 115580721A CN 202211461253 A CN202211461253 A CN 202211461253A CN 115580721 A CN115580721 A CN 115580721A
- Authority
- CN
- China
- Prior art keywords
- picture
- detected
- quality detection
- video
- image 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.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/004—Diagnosis, testing or measuring for television systems or their details for digital television systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- 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
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The application relates to the technical field of video detection, and provides a quality detection method and device for a monitoring video of substation equipment, computer equipment, a storage medium and a computer program product. According to the method and the device, the accuracy and the efficiency of determining the video quality detection result of the video can be improved. The method comprises the steps of obtaining a video to be detected, inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected, inputting the feature vector into a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected, and determining the video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
Description
Technical Field
The application relates to the technical field of video detection, in particular to a quality detection method and device for a monitoring video of substation equipment, computer equipment, a storage medium and a computer program product.
Background
With the improvement of the automation and intelligence level of power grid dispatching, the video monitoring technology is widely applied to a transformer substation system. In the process of processing video information, situations such as distortion and the like occur inevitably, which cause damage to image quality, such as black screen, snowflake, stripe and the like, and the video shot under severe weather conditions such as fog, haze and rain also has the problem of image quality reduction, such as shading, dim color, blur and the like. Therefore, how to perform video quality detection on a video becomes an important research direction for determining a video quality detection result of the video.
The conventional technology generally determines the video quality detection result of the video by manually detecting the definition of the video, but the method mainly depends on manual detection, so that the efficiency of determining the video quality detection result of the video is low.
Disclosure of Invention
In view of the above, it is necessary to provide a quality detection method and apparatus for substation equipment monitoring video, a computer device, a computer readable storage medium, and a computer program product.
In a first aspect, the application provides a quality detection method for a substation equipment monitoring video. The method comprises the following steps:
acquiring a video to be detected; the video to be detected comprises a picture to be detected;
inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected;
inputting the characteristic vector into a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected;
and determining a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
In one embodiment, before the image to be detected is input to the feature vector calculation model to obtain the feature vector of the image to be detected, the method further includes:
performing resolution reduction processing on the picture to be detected to obtain a low-resolution picture of the picture to be detected;
inputting the picture to be detected into the feature vector calculation model to obtain the feature vector of the picture to be detected, and the method comprises the following steps:
inputting the picture to be detected into a feature vector calculation model to obtain a first feature vector of the picture to be detected;
inputting the low-resolution picture into the feature vector calculation model to obtain a second feature vector of the low-resolution picture;
and determining the characteristic vector of the picture to be detected according to the first characteristic vector and the second characteristic vector.
In one embodiment, inputting the feature vector into a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected, includes:
inputting the characteristic vectors into a pre-trained image quality detection model, and obtaining an image quality detection sub-result of each picture to be detected through the pre-trained image quality detection model;
and determining the image quality detection result of the picture to be detected according to the image quality detection sub-result of each picture to be detected.
In one embodiment, before acquiring the video to be detected, the method further includes:
acquiring an original video; the original video comprises an original picture;
inputting an original video into a pre-trained quality abnormity identification model, deleting a quality abnormity picture in an original picture through the pre-trained quality abnormity identification model, and obtaining the original video with the quality abnormity picture deleted;
and taking the original video with the quality abnormal picture deleted as the video to be detected.
In one embodiment, the pre-trained quality anomaly recognition model is trained by:
acquiring a normal-quality sample picture and an abnormal-quality sample picture;
and training the quality abnormity identification model to be trained by utilizing the sample quality normal picture and the sample quality abnormity picture to obtain a pre-trained quality abnormity identification model.
In one embodiment, the pre-trained image quality detection model is trained by:
acquiring a sample characteristic vector and a real image quality detection result of the sample characteristic vector;
and training the image quality detection model to be trained by using the sample characteristic vector and the real image quality detection result to obtain a pre-trained image quality detection model.
In one embodiment, the method further comprises:
selecting a picture to be detected, of which the image quality detection result meets a preset image quality detection condition, from the pictures to be detected;
and taking the picture to be detected with the image quality detection result meeting the preset image quality detection condition as a preview picture of the video to be detected.
In a second aspect, the application further provides a quality detection device for the substation equipment monitoring video. The device comprises:
the video acquisition module is used for acquiring a video to be detected; the video to be detected comprises a picture to be detected;
the picture input module is used for inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected;
the vector input module is used for inputting the characteristic vector to a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected;
and the result determining module is used for determining the video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:
acquiring a video to be detected; the video to be detected comprises a picture to be detected; inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected; inputting the characteristic vector into a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected; and determining a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring a video to be detected; the video to be detected comprises a picture to be detected; inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected; inputting the characteristic vector into a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected; and determining a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:
acquiring a video to be detected; the video to be detected comprises a picture to be detected; inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected; inputting the characteristic vector into a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected; and determining a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
According to the quality detection method and device for the substation equipment monitoring video, the computer equipment, the storage medium and the computer program product, the video to be detected is obtained, the picture to be detected is contained in the video to be detected, the picture to be detected is input into the feature vector calculation model, the feature vector of the picture to be detected is obtained, the feature vector is input into the pre-trained image quality detection model, the image quality detection result of the picture to be detected is obtained, and the video quality detection result of the video to be detected is determined according to the image quality detection result of the picture to be detected. According to the scheme, the video is obtained, the characteristic vector of the picture in the video is calculated, so that the judgment basis of the picture quality is quickly obtained, the image quality detection result of the picture is determined according to the characteristic vector, so that the quality detection result of each picture in the video is accurately obtained, the video quality detection result of the video is automatically determined by combining the quality detection result of each picture in the video, and therefore the accuracy and the efficiency of determining the video quality detection result of the video are improved.
Drawings
Fig. 1 is a schematic flow chart of a quality detection method of a substation equipment monitoring video in one embodiment;
fig. 2 is a schematic flow chart of a quality detection method of a substation equipment monitoring video in another embodiment;
fig. 3 is a block diagram of a quality detection apparatus for a substation equipment monitoring video according to an embodiment;
FIG. 4 is a diagram of the internal structure of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application 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 present application and are not intended to limit the present application.
In an embodiment, as shown in fig. 1, a quality detection method for a substation equipment monitoring video is provided, and this embodiment is exemplified by applying the method to a terminal, and includes the following steps:
and S101, acquiring a video to be detected.
In this step, the video to be detected comprises a picture to be detected; the video can be a substation equipment monitoring video; the picture to be detected may be a video frame of the video to be detected.
Specifically, the terminal selects the video to be detected from the substation equipment monitoring video.
And S102, inputting the picture to be detected into the feature vector calculation model to obtain the feature vector of the picture to be detected.
In this step, the feature vector calculation model may be a model for calculating a feature vector of the to-be-detected picture, for example, a model for calculating an Image feature vector based on a BRISQUE algorithm (a blank referred Image Spatial Quality Evaluator, without a reference Image Spatial Quality Evaluator); the feature vector may be a multi-dimensional feature vector(s), such as a 36-dimensional (36) feature vector.
Specifically, the terminal selects a picture to be detected from a video to be detected, inputs the picture to be detected into the feature vector calculation model, and calculates through the feature vector calculation model to obtain the feature vector of the picture to be detected.
Step S103, inputting the characteristic vector to a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected.
In this step, the image quality detection model may be a Support Vector Machine (SVM) model; the image quality detection result may be an image quality evaluation, such as an image quality score (score), of each picture to be detected.
Specifically, the terminal inputs the feature vector of each picture to be detected into a pre-trained image quality detection model, and the image quality detection result of each picture to be detected is obtained through the pre-trained image quality detection model.
And step S104, determining a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
In this step, the video quality detection result may be a video quality evaluation of the video to be detected, such as a video quality score (score).
Specifically, the terminal performs fusion processing (for example, summing and averaging, or weighted summing and averaging) on the image quality detection result of the picture to be detected to obtain a video quality detection result of the video to be detected.
According to the quality detection method of the substation equipment monitoring video, a video to be detected is obtained, the video to be detected comprises a picture to be detected, the picture to be detected is input to a feature vector calculation model to obtain a feature vector of the picture to be detected, the feature vector is input to a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected, and the video quality detection result of the video to be detected is determined according to the image quality detection result of the picture to be detected. According to the scheme, the video is obtained, the characteristic vector of the picture in the video is calculated, so that the judgment basis of the picture quality is quickly obtained, the image quality detection result of the picture is determined according to the characteristic vector, so that the quality detection result of each picture in the video is accurately obtained, the video quality detection result of the video is automatically determined by combining the quality detection result of each picture in the video, and therefore the accuracy and the efficiency of determining the video quality detection result of the video are improved.
In an embodiment, the method may further include the following steps to obtain a low-resolution picture of the picture to be detected, specifically including: performing resolution reduction processing on the picture to be detected to obtain a low-resolution picture of the picture to be detected; the step S102 of inputting the picture to be detected into the feature vector calculation model to obtain the feature vector of the picture to be detected specifically includes: inputting the picture to be detected into a feature vector calculation model to obtain a first feature vector of the picture to be detected; inputting the low-resolution picture into the feature vector calculation model to obtain a second feature vector of the low-resolution picture; and determining the characteristic vector of the picture to be detected according to the first characteristic vector and the second characteristic vector.
In this embodiment, the first feature vector may be a feature vector of the to-be-detected picture calculated by the feature vector calculation model; the second feature vector may be a feature vector of a low-resolution picture calculated by the feature vector calculation model.
Specifically, the terminal performs resolution reduction processing on the picture to be detected to obtain a picture (low-resolution picture) under low resolution, inputs the picture to be detected to the feature vector calculation model to obtain a first feature vector of the picture to be detected, inputs the low-resolution picture to the feature vector calculation model, obtains a second feature vector of the low-resolution picture by using the same principle, and uses the first feature vector and the second feature vector as feature vectors of the picture to be detected (for example, combines the first feature vector and the second feature vector to obtain feature vectors of the picture to be detected).
According to the technical scheme of the embodiment, the feature vector of the picture to be detected is determined according to the first feature vector of the picture to be detected and the second feature vector of the picture with low resolution, so that the feature vector of the picture to be detected can be obtained more conveniently, the features of the picture to be detected can be enhanced, and the accuracy of the video quality detection result of the video can be improved subsequently.
In an embodiment, the step S103 of inputting the feature vector into a pre-trained image quality detection model to obtain an image quality detection result of the to-be-detected picture specifically includes: inputting the characteristic vectors into a pre-trained image quality detection model, and obtaining an image quality detection sub-result of each picture to be detected through the pre-trained image quality detection model; and determining the image quality detection result of the picture to be detected according to the image quality detection sub-result of each picture to be detected.
In this embodiment, the image quality detection sub-result may be a detection result of image quality of each to-be-detected picture, which is obtained by detecting a pre-trained image quality detection model.
Specifically, the terminal inputs the feature vector into a pre-trained image quality detection model, obtains an image quality detection sub-result of each picture to be detected through the pre-trained image quality detection model, and uses the image quality detection sub-result of each picture to be detected as an image quality detection result of the picture to be detected.
According to the technical scheme of the embodiment, the image quality detection result of the picture to be detected is determined according to the image quality detection sub-result of each picture to be detected, so that the image quality detection sub-results of each picture to be detected can be accurately and quickly obtained, and the image quality detection result of the picture to be detected can be accurately determined according to the accurate image quality detection sub-results, and the accuracy and the efficiency of determining the video quality detection result of the video can be improved.
In an embodiment, the method may further obtain the video to be detected through the following steps, specifically including: acquiring an original video; inputting an original video into a pre-trained quality abnormity identification model, deleting a quality abnormity picture in an original picture through the pre-trained quality abnormity identification model, and obtaining the original video with the quality abnormity picture deleted; and taking the original video with the quality abnormal picture deleted as the video to be detected.
In this embodiment, the original video includes an original picture, and the original picture may be a video frame in the original video; the pre-trained quality anomaly identification model can be a pre-trained model for identifying pictures with quality problems, such as a pre-trained deep learning-based classification model, and can be an image binary classification model trained by using a Resnet50 algorithm (a residual neural network model with 50 convolutional layers in the whole network); the picture with abnormal quality may be a picture with quality problems, for example, a picture with serious quality problems such as a serious black screen, a serious snowflake or a serious stripe caused in the process of processing the video information, or a picture in which the equipment target of the substation cannot be seen.
Specifically, before the terminal acquires the video to be detected, the terminal acquires an original video, inputs the original video to a pre-trained quality abnormality recognition model, deletes a quality abnormality picture in an original picture in the original video through the pre-trained quality abnormality recognition model to obtain the original video with the quality abnormality picture deleted, and uses the original video with the quality abnormality picture deleted as the video to be detected.
According to the technical scheme, the quality abnormal picture in the original picture is deleted through the pre-trained quality abnormal recognition model, and the original video with the quality abnormal picture deleted is used as the video to be detected, so that the finally obtained image quality detection result of the picture to be detected and the video quality detection result of the video to be detected are prevented from being greatly influenced due to the picture with serious quality problem in the video to be detected, and the accuracy of the video quality detection result of the video to be detected is favorably improved.
In one embodiment, the pre-trained quality anomaly recognition model is obtained by training in the following way, specifically including: acquiring a normal-quality sample picture and an abnormal-quality sample picture; and training the quality abnormity identification model to be trained by utilizing the sample quality normal picture and the sample quality abnormity picture to obtain a pre-trained quality abnormity identification model.
In this embodiment, the sample picture with normal quality may be a sample picture with normal quality; the sample quality abnormality picture may be a quality abnormality sample picture.
Specifically, the terminal acquires a sample quality normal picture and a sample quality abnormal picture from a training set, inputs the sample quality normal picture and the sample quality abnormal picture into a quality abnormal recognition model to be trained for repeated training until a recognition result of the quality abnormal recognition model to be trained for recognizing the sample quality normal picture and the sample quality abnormal picture (a classification result of two classifications of the sample quality normal picture and the sample quality abnormal picture) meets a preset recognition accuracy condition, stops training, and takes the trained quality abnormal recognition model as a pre-trained quality abnormal recognition model.
According to the technical scheme, the quality abnormity identification model to be trained is trained by utilizing the sample quality normal picture and the sample quality abnormity picture, so that the more accurate and efficient pre-trained quality abnormity identification model can be obtained, and the accuracy and the efficiency of the video quality detection result of the video can be improved subsequently.
In one embodiment, the pre-trained image quality inspection model is obtained by training in the following way, specifically including: acquiring a sample characteristic vector and a real image quality detection result of the sample characteristic vector; and training the image quality detection model to be trained by using the sample characteristic vector and the real image quality detection result to obtain a pre-trained image quality detection model.
In this embodiment, the sample feature vector may be a feature vector obtained by performing feature vector calculation on the sample picture, and the real image quality detection result of the sample feature vector may be a real image quality detection result of the sample picture.
Specifically, the terminal obtains a sample feature vector and a real image quality detection result corresponding to the sample feature vector from another training set, inputs the sample feature vector to an image quality detection model to be trained, and repeatedly trains the sample feature vector until a matching degree between the image quality detection result of the sample feature vector obtained by detecting the sample feature vector by the image quality detection model to be trained and the real image quality detection result corresponding to the sample feature vector reaches a preset matching degree condition (which is equivalent to the preset accuracy condition that the accuracy of the image quality detection result obtained by detecting the sample feature vector by the image quality detection model to be trained reaches), stops training, and takes the trained image quality detection model as a pre-trained image quality detection model.
According to the technical scheme, the image quality detection model to be trained is trained by utilizing the sample characteristic vector and the real image quality detection result, so that the more accurate and efficient pre-trained image quality detection model can be obtained, and the accuracy and the efficiency of determining the video quality detection result of the video can be improved subsequently.
In an embodiment, the method may further include determining a preview picture of the video to be detected by the following steps, specifically including: selecting a picture to be detected, of which the image quality detection result meets a preset image quality detection condition, from the pictures to be detected; and taking the picture to be detected with the image quality detection result meeting the preset image quality detection condition as a preview picture of the video to be detected.
In this embodiment, the preset image quality detection condition may be a preset image quality scoring threshold condition; the preview picture may be a picture (e.g., a clear picture) with good image quality contained in the video to be detected.
Specifically, the terminal determines an image quality detection result of each picture to be detected according to the image quality detection result, determines an image quality score of each picture to be detected according to the image quality detection result of each picture to be detected (for example, the image quality score ranges from 0 to 100, and the score is lower, indicating that the image quality is better, and the score is higher, indicating that the image quality is worse), selects a picture to be detected (for example, a picture to be detected with an image quality detection result meeting a preset image quality detection condition) from each picture to be detected, and takes the picture to be detected with the image quality detection result meeting the preset image quality detection condition as a preview picture of the video to be detected (for example, the preview picture can be displayed when a user selects to browse the content of the video to be detected).
According to the technical scheme, the picture to be detected with the image quality detection result meeting the preset image quality detection condition is taken as the preview picture of the video to be detected, so that the picture (clear picture) with good image quality can be selected from the video to be detected and taken as the preview picture of the video, and the accuracy and the efficiency of determining the picture with the image quality meeting the requirements in the video can be improved.
The quality detection method for the monitoring video of the substation equipment provided by the application is described in an embodiment, and the embodiment is exemplified by applying the method to a terminal, and mainly comprises the following steps:
firstly, a terminal acquires a picture with normal sample quality and a picture with abnormal sample quality.
And secondly, training the quality abnormity identification model to be trained by the terminal by using the sample quality normal picture and the sample quality abnormity picture to obtain a pre-trained quality abnormity identification model.
And thirdly, the terminal acquires the sample characteristic vector and a real image quality detection result of the sample characteristic vector.
And fourthly, training the image quality detection model to be trained by the terminal by using the sample characteristic vector and the real image quality detection result to obtain a pre-trained image quality detection model.
And fifthly, the terminal acquires an original video.
And sixthly, inputting the original video to a pre-trained quality abnormity identification model by the terminal, and deleting a quality abnormity picture in the original picture through the pre-trained quality abnormity identification model to obtain the original video with the quality abnormity picture deleted.
And seventhly, the terminal takes the original video with the quality abnormal pictures deleted as the video to be detected.
And eighthly, selecting the picture to be detected with the image quality detection result meeting the preset image quality detection condition from the pictures to be detected by the terminal.
And step nine, the terminal takes the picture to be detected with the image quality detection result meeting the preset image quality detection condition as a preview picture of the video to be detected.
And step ten, the terminal performs resolution reduction processing on the picture to be detected to obtain a low-resolution picture of the picture to be detected.
And step eleven, the terminal inputs the picture to be detected into the feature vector calculation model to obtain a first feature vector of the picture to be detected.
And step ten, the terminal inputs the low-resolution picture into the feature vector calculation model to obtain a second feature vector of the low-resolution picture.
And thirteenth, the terminal determines the feature vector of the picture to be detected according to the first feature vector and the second feature vector.
And fourteenth, inputting the characteristic vector into a pre-trained image quality detection model by the terminal, and obtaining the image quality detection sub-result of each picture to be detected through the pre-trained image quality detection model.
And fifteenth, the terminal determines the image quality detection result of the picture to be detected according to the image quality detection sub-result of each picture to be detected.
Sixthly, the terminal determines the video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
Wherein, the original video comprises an original picture; the video to be detected comprises a picture to be detected.
According to the technical scheme, the judgment basis of the picture quality is quickly obtained by acquiring the video and calculating the characteristic vector of the picture in the video, the image quality detection result of the picture is determined according to the characteristic vector, the quality detection result of each picture in the video is accurately obtained, the video quality detection result of the video is automatically determined by combining the quality detection result of each picture in the video, and therefore the accuracy and the efficiency of the video quality detection result of the video are improved.
The quality detection method for the substation equipment monitoring video provided by the application is described as an application example, the application example is illustrated by applying the method to a terminal, and as shown in fig. 2, the method mainly comprises the following steps:
firstly, a terminal collects a large number of pictures with serious quality problems such as black screens, snowflakes and stripes (the pictures are characterized in that equipment objects of a transformer substation cannot be seen), the pictures with the problems and normal video monitoring pictures are divided into two folders to be stored, an image classification model is trained by using a Resnet50 algorithm, and a Resnet50 algorithm model based on deep learning is obtained.
And secondly, the terminal intercepts pictures (to-be-detected pictures) from the monitoring video (to-be-detected video), primary classification of the video monitoring image (to-be-detected video) is realized based on a Resnet50 algorithm model (image two classification model) of deep learning, and pictures with serious quality problems such as black screens, snowflakes, stripes and the like caused in the video information processing process are screened out (if the detected image is abnormal, equipment is abnormal to give an alarm, the Resnet50 algorithm model based on the deep learning is used for processing, and if the detected image is not abnormal, the next step is carried out).
Thirdly, the terminal calculates the image feature vector based on a BRISQUE algorithm: the terminal calculates the image (picture to be detected) MSCN (Mean minus Contrast Normalized) coefficient (Mean minus Contrast Normalized value of calculated image) according to the following formula:
where I ∈ {1,2, \8230;, M }, j ∈ {1,2, \8230;, N }, M, N are the height and width of the image, respectively, I (I, j) is the pixel value of the grayscale image, the value range is [0, 255], C =1 guarantees that the denominator is not zero, and the variables μ (I, j) and σ (I, j) are defined as follows:
wherein w is a 2-dimensional Gaussian weight function, and the values of K and L are usually 3; fitting MSCN value into GGD (Generalized Gaussian distribution) according to the following formula to obtain 2 GGD eigenvectors (alpha, sigma) 2 ) Where α controls the shape, σ 2 Controlling the variance:
in order to add correlation information between connected pixelsThen, the terminal calculates the product of adjacent MSCN coefficients in four directions of horizontal (H), vertical (V), main diagonal (D1) and secondary diagonal (D2), fits the product of adjacent coefficients into AGGD (Asymmetric Generalized Gaussian Distribution), and obtains 4 AGGD eigenvectorsThe total of 16 features in four directions and 2 feature vectors extracted by the GGD are added to obtain 18-dimensional features, the resolution of an original image (picture to be detected) is reduced to obtain an image (low-resolution picture) under low scale (low resolution), the 18-dimensional features are calculated according to the above, and the 36-dimensional features (calculated image features) are obtained, for example,
Fourthly, the terminal collects clear substation monitoring video pictures and pictures which are not clear enough due to various reasons, the number of the pictures is possibly small, means for assisting image enhancement such as image atomization and noise adding are carried out, 36-dimensional features are obtained by the algorithm of the last step on the clear pictures and the pictures which are not clear enough, a subjective quality evaluation score is given to each picture, the image features and the quality scores are arranged into a training set, a support vector machine is trained to obtain a prediction model, the test image features are extracted and then sent to the prediction model (input to the support vector machine for prediction), the image quality score of the test image is obtained through prediction (for example, the image quality score range is 0-100, the score is lower, the image quality is better, and the score is higher, the image quality is worse), and the video quality detection result of the video to be detected is determined according to the image quality score (the image quality detection result) of each test image (the picture to be predicted).
According to the technical scheme of the application example, quality evaluation aiming at the video monitoring image of the transformer substation is achieved, the quality evaluation of the image can be achieved without relying on manual parameter adjustment and reference images, the video quality detection result is obtained, the judgment basis is provided for the follow-up computer vision algorithm, and the method and the device have wide application value in the field of video monitoring.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a quality detection device of the substation equipment monitoring video, which is used for realizing the quality detection method of the substation equipment monitoring video. The implementation scheme for solving the problem provided by the apparatus is similar to the implementation scheme described in the method, so the specific limitations in the following embodiments of the quality detection apparatus for one or more transformer substation equipment monitoring videos may refer to the limitations on the quality detection method for the transformer substation equipment monitoring videos, which are not described herein again.
In one embodiment, as shown in fig. 3, there is provided a quality detection apparatus for substation equipment monitoring video, where the apparatus 300 may include:
the video acquisition module 301 is used for acquiring a video to be detected; the video to be detected comprises a picture to be detected;
the picture input module 302 is configured to input the picture to be detected to a feature vector calculation model to obtain a feature vector of the picture to be detected;
a vector input module 303, configured to input the feature vector to a pre-trained image quality detection model to obtain an image quality detection result of the to-be-detected picture;
a result determining module 304, configured to determine a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
In one embodiment, the apparatus 300 further comprises: a low resolution picture obtaining module, configured to perform resolution reduction processing on the picture to be detected to obtain a low resolution picture of the picture to be detected; the picture input module 302 is further configured to input the picture to be detected to the feature vector calculation model to obtain a first feature vector of the picture to be detected; inputting the low-resolution picture into the feature vector calculation model to obtain a second feature vector of the low-resolution picture; and determining the characteristic vector of the picture to be detected according to the first characteristic vector and the second characteristic vector.
In one embodiment, the vector input module 303 is further configured to input the feature vector to the pre-trained image quality detection model, and obtain an image quality detection sub-result of each picture to be detected through the pre-trained image quality detection model; and determining the image quality detection result of the picture to be detected according to the image quality detection sub-result of each picture to be detected.
In one embodiment, the apparatus 300 further comprises: the video is used as a module for obtaining an original video; the original video comprises an original picture; inputting the original video into a pre-trained quality abnormity identification model, deleting a quality abnormity picture in the original picture through the pre-trained quality abnormity identification model, and obtaining the original video with the quality abnormity picture deleted; and taking the original video with the quality abnormal picture deleted as the video to be detected.
In one embodiment, the apparatus 300 further comprises: the quality anomaly identification model training module is used for acquiring a sample quality normal picture and a sample quality anomaly picture; and training a quality abnormity identification model to be trained by utilizing the sample quality normal picture and the sample quality abnormity picture to obtain the pre-trained quality abnormity identification model.
In one embodiment, the apparatus 300 further comprises: the image quality detection model training module is used for acquiring a sample feature vector and a real image quality detection result of the sample feature vector; and training an image quality detection model to be trained by using the sample feature vector and the real image quality detection result to obtain the pre-trained image quality detection model.
In one embodiment, the apparatus 300 further comprises: the picture serving as a module is used for selecting the picture to be detected, of which the image quality detection result meets the preset image quality detection condition, from the pictures to be detected; and taking the picture to be detected with the image quality detection result meeting the preset image quality detection condition as a preview picture of the video to be detected.
All modules in the quality detection device of the substation equipment monitoring video can be completely or partially realized through software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 4. The computer apparatus includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input device. The processor, the memory and the input/output interface are connected by a system bus, and the communication interface, the display unit and the input device are connected by the input/output interface to the system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The input/output interface of the computer device is used for exchanging information between the processor and an external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to realize a quality detection method for the substation equipment monitoring video. The display unit of the computer device is used for forming a visual visible picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the configuration shown in fig. 4 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, carries out the steps in the method embodiments described above.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the relevant laws and regulations and standards of the relevant country and region.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), magnetic Random Access Memory (MRAM), ferroelectric Random Access Memory (FRAM), phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), for example. The databases involved in the embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A quality detection method for a monitoring video of substation equipment is characterized by comprising the following steps:
acquiring a video to be detected; the video to be detected comprises a picture to be detected;
inputting the picture to be detected into a feature vector calculation model to obtain a feature vector of the picture to be detected;
inputting the characteristic vector to a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected;
and determining a video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
2. The method according to claim 1, before inputting the picture to be detected into a feature vector calculation model to obtain the feature vector of the picture to be detected, further comprising:
performing resolution reduction processing on the picture to be detected to obtain a low-resolution picture of the picture to be detected;
the inputting the picture to be detected into a feature vector calculation model to obtain the feature vector of the picture to be detected comprises the following steps:
inputting the picture to be detected into the feature vector calculation model to obtain a first feature vector of the picture to be detected;
inputting the low-resolution picture into the feature vector calculation model to obtain a second feature vector of the low-resolution picture;
and determining the characteristic vector of the picture to be detected according to the first characteristic vector and the second characteristic vector.
3. The method according to claim 1, wherein the inputting the feature vector to a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected comprises:
inputting the feature vectors into the pre-trained image quality detection model, and obtaining image quality detection sub-results of the pictures to be detected through the pre-trained image quality detection model;
and determining the image quality detection result of the picture to be detected according to the image quality detection sub-result of each picture to be detected.
4. The method of claim 1, further comprising, prior to acquiring the video to be detected:
acquiring an original video; the original video comprises an original picture;
inputting the original video into a pre-trained quality abnormity identification model, deleting a quality abnormity picture in the original picture through the pre-trained quality abnormity identification model, and obtaining the original video with the quality abnormity picture deleted;
and taking the original video with the quality abnormal picture deleted as the video to be detected.
5. The method of claim 4, wherein the pre-trained quality anomaly recognition model is trained by:
acquiring a normal-quality sample picture and an abnormal-quality sample picture;
and training a quality abnormity identification model to be trained by utilizing the sample quality normal picture and the sample quality abnormity picture to obtain the pre-trained quality abnormity identification model.
6. The method of claim 1, wherein the pre-trained image quality detection model is trained by:
acquiring a sample characteristic vector and a real image quality detection result of the sample characteristic vector;
and training an image quality detection model to be trained by using the sample characteristic vector and the real image quality detection result to obtain the pre-trained image quality detection model.
7. The method of claim 1, further comprising:
selecting the picture to be detected with the image quality detection result meeting the preset image quality detection condition from the pictures to be detected;
and taking the picture to be detected with the image quality detection result meeting the preset image quality detection condition as a preview picture of the video to be detected.
8. A quality detection device for substation equipment monitoring video is characterized by comprising:
the video acquisition module is used for acquiring a video to be detected; the video to be detected comprises a picture to be detected;
the image input module is used for inputting the image to be detected into the feature vector calculation model to obtain the feature vector of the image to be detected;
the vector input module is used for inputting the characteristic vector to a pre-trained image quality detection model to obtain an image quality detection result of the picture to be detected;
and the result determining module is used for determining the video quality detection result of the video to be detected according to the image quality detection result of the picture to be detected.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211461253.XA CN115580721B (en) | 2022-11-18 | 2022-11-18 | Quality detection method, device, equipment and medium for monitoring video of substation equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211461253.XA CN115580721B (en) | 2022-11-18 | 2022-11-18 | Quality detection method, device, equipment and medium for monitoring video of substation equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115580721A true CN115580721A (en) | 2023-01-06 |
CN115580721B CN115580721B (en) | 2023-03-31 |
Family
ID=84588182
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211461253.XA Active CN115580721B (en) | 2022-11-18 | 2022-11-18 | Quality detection method, device, equipment and medium for monitoring video of substation equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115580721B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105610642A (en) * | 2015-12-22 | 2016-05-25 | 海南电网有限责任公司 | Intelligent scheduling and fault result displaying method for video quality diagnosis of transformer substation |
CN109740480A (en) * | 2018-12-26 | 2019-05-10 | 浙江捷尚视觉科技股份有限公司 | A kind of identified again based on non-motor vehicle pedestrian retrieval method under monitoring scene |
CN109840528A (en) * | 2019-01-31 | 2019-06-04 | 北京字节跳动网络技术有限公司 | The method and apparatus for extracting the characteristic pattern of image |
CN111738048A (en) * | 2020-03-10 | 2020-10-02 | 重庆大学 | Pedestrian re-identification method |
CN112418009A (en) * | 2020-11-06 | 2021-02-26 | 中保车服科技服务股份有限公司 | Image quality detection method, terminal device and storage medium |
CN112597864A (en) * | 2020-12-16 | 2021-04-02 | 佳都新太科技股份有限公司 | Monitoring video abnormity detection method and device |
CN113887373A (en) * | 2021-09-27 | 2022-01-04 | 中关村科学城城市大脑股份有限公司 | Attitude identification method and system based on urban intelligent sports parallel fusion network |
CN114612821A (en) * | 2022-03-04 | 2022-06-10 | 广州方硅信息技术有限公司 | Live front cover image quality detection method, device, equipment and storage medium |
CN114666634A (en) * | 2022-03-21 | 2022-06-24 | 北京达佳互联信息技术有限公司 | Image quality detection result display method, device, equipment and storage medium |
CN115330661A (en) * | 2021-04-26 | 2022-11-11 | 北京奇虎科技有限公司 | Video image quality detection method, device and equipment and readable storage medium |
-
2022
- 2022-11-18 CN CN202211461253.XA patent/CN115580721B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105610642A (en) * | 2015-12-22 | 2016-05-25 | 海南电网有限责任公司 | Intelligent scheduling and fault result displaying method for video quality diagnosis of transformer substation |
CN109740480A (en) * | 2018-12-26 | 2019-05-10 | 浙江捷尚视觉科技股份有限公司 | A kind of identified again based on non-motor vehicle pedestrian retrieval method under monitoring scene |
CN109840528A (en) * | 2019-01-31 | 2019-06-04 | 北京字节跳动网络技术有限公司 | The method and apparatus for extracting the characteristic pattern of image |
CN111738048A (en) * | 2020-03-10 | 2020-10-02 | 重庆大学 | Pedestrian re-identification method |
CN112418009A (en) * | 2020-11-06 | 2021-02-26 | 中保车服科技服务股份有限公司 | Image quality detection method, terminal device and storage medium |
CN112597864A (en) * | 2020-12-16 | 2021-04-02 | 佳都新太科技股份有限公司 | Monitoring video abnormity detection method and device |
CN115330661A (en) * | 2021-04-26 | 2022-11-11 | 北京奇虎科技有限公司 | Video image quality detection method, device and equipment and readable storage medium |
CN113887373A (en) * | 2021-09-27 | 2022-01-04 | 中关村科学城城市大脑股份有限公司 | Attitude identification method and system based on urban intelligent sports parallel fusion network |
CN114612821A (en) * | 2022-03-04 | 2022-06-10 | 广州方硅信息技术有限公司 | Live front cover image quality detection method, device, equipment and storage medium |
CN114666634A (en) * | 2022-03-21 | 2022-06-24 | 北京达佳互联信息技术有限公司 | Image quality detection result display method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN115580721B (en) | 2023-03-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111080620B (en) | Road disease detection method based on deep learning | |
CN112132093B (en) | High-resolution remote sensing image target detection method and device and computer equipment | |
KR20200087297A (en) | Defect inspection method and apparatus using image segmentation based on artificial neural network | |
CN104008538A (en) | Super-resolution method based on single image | |
CN111209858B (en) | Real-time license plate detection method based on deep convolutional neural network | |
CN107886082B (en) | Method and device for detecting mathematical formulas in images, computer equipment and storage medium | |
CN107784288A (en) | A kind of iteration positioning formula method for detecting human face based on deep neural network | |
CN105260995B (en) | A kind of image repair and denoising method and system | |
CN112906794A (en) | Target detection method, device, storage medium and terminal | |
CN111199523A (en) | Power equipment identification method and device, computer equipment and storage medium | |
CN110766027A (en) | Image area positioning method and training method of target area positioning model | |
CN113111722A (en) | Automatic driving target identification method based on improved Mask R-CNN | |
CN112419202A (en) | Wild animal image automatic identification system based on big data and deep learning | |
CN110807463B (en) | Image segmentation method and device, computer equipment and storage medium | |
CN117576724A (en) | Unmanned plane bird detection method, system, equipment and medium | |
CN111738164A (en) | Pedestrian detection method based on deep learning | |
WO2022219402A1 (en) | Semantically accurate super-resolution generative adversarial networks | |
CN114332602A (en) | Commodity identification method of intelligent container | |
CN111582057B (en) | Face verification method based on local receptive field | |
CN117037244A (en) | Face security detection method, device, computer equipment and storage medium | |
CN115580721B (en) | Quality detection method, device, equipment and medium for monitoring video of substation equipment | |
CN116543333A (en) | Target recognition method, training method, device, equipment and medium of power system | |
CN116310832A (en) | Remote sensing image processing method, device, equipment, medium and product | |
CN112232102A (en) | Building target identification method and system based on deep neural network and multitask learning | |
CN115937095A (en) | Printing defect detection method and system integrating image processing algorithm and deep learning |
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 |