CN116579990A - Video mosaic detection method, system, equipment and medium - Google Patents
Video mosaic detection method, system, equipment and medium Download PDFInfo
- Publication number
- CN116579990A CN116579990A CN202310442540.4A CN202310442540A CN116579990A CN 116579990 A CN116579990 A CN 116579990A CN 202310442540 A CN202310442540 A CN 202310442540A CN 116579990 A CN116579990 A CN 116579990A
- Authority
- CN
- China
- Prior art keywords
- mosaic
- image
- video
- detection
- detection result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 263
- 238000012549 training Methods 0.000 claims abstract description 40
- 238000013528 artificial neural network Methods 0.000 claims abstract description 23
- 238000000034 method Methods 0.000 claims abstract description 21
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 2
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 2
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a video mosaic detection method, a system, equipment and a medium, belonging to the technical field of video image processing, wherein the method comprises the following steps: step 1, decoding a video to be detected according to frames to obtain a decoded single image; step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image; repeating the step 1 and the step 2 until the video decoding to be detected is finished; determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not. The invention reduces the false detection rate of video mosaic detection.
Description
Technical Field
The invention relates to the technical field of video image processing, in particular to a video mosaic detection method, a video mosaic detection system, video mosaic detection equipment and video mosaic detection media.
Background
The appearance of a frame picture of video in a video program is covered by a series of small rectangular blocks, i.e. mosaic blocks, which appear such that we cannot see the original face of the picture, typically produced in video coding transmissions. When a live video sees a relevant video program, a mosaic in a picture is often seen, and the video viewing experience is seriously affected.
The existing video mosaic detection method mainly comprises a traditional detection method and a detection method based on a neural network. Conventional detection methods include detection methods based on edge and template matching and detection methods based on region analysis. The detection method based on the neural network model training is to train the neural network model by using pictures marked with mosaic areas as training samples, so as to achieve the purpose of mosaic detection.
The detection method based on edge and template matching has strong dependence on the size of the image block and the reference point, and has very high false detection rate although the omission rate is very low, especially for the scene similar to mosaic, such as electronic screen, chessboard, net, block, full screen text and the like, the false detection is easy to occur. The detection method based on neural network model training depends on sample collection conditions, can influence the detection accuracy, and the detection speed can be influenced due to poor model selection. In the prior art, the problems of high false detection rate and low detection efficiency exist.
Disclosure of Invention
The invention provides a video mosaic detection method, a system, equipment and a medium, which are used for solving the defect of low detection accuracy of the video mosaic in the prior art, realizing the improvement of the detection accuracy of the video mosaic, reducing the false detection rate of the mosaic detection and expanding the application scene.
The invention provides a video mosaic detection method, which comprises the following steps:
step 1, decoding a video to be detected according to frames to obtain a decoded single image;
step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
According to the video mosaic detection method provided by the invention, the mosaic image set comprises a plurality of historical mosaic images;
the conventional image set comprises a plurality of historical normal images and historical misjudgment mosaic images;
wherein the mosaic image set is the same as the number of images in the regular image set.
According to the video mosaic detection method provided by the invention, the determining of the second detection result corresponding to the video to be detected based on the first detection result corresponding to each image comprises the following steps:
determining that a second detection result corresponding to the video to be detected indicates that the video to be detected contains mosaic under the condition that the number of images, which indicate that the images are mosaic images, in the first detection result corresponding to each image is greater than or equal to a preset threshold value;
or determining that the second detection result corresponding to the video to be detected indicates that the video to be detected does not contain mosaic under the condition that the number of images, which indicate that the images are mosaic images, in the first detection result corresponding to each image is smaller than a preset threshold value.
According to the video mosaic detection method provided by the invention, after the image is input into the mosaic image detection model to obtain the first detection result corresponding to the image output by the mosaic image detection model, the method further comprises the following steps:
adding one to the value of the judgment value under the condition that the image is indicated to be a mosaic image in the first detection result corresponding to the image;
correspondingly, the determining the second detection result corresponding to the video to be detected based on the first detection result corresponding to each image includes:
determining that a second detection result corresponding to the video to be detected indicates that the video to be detected contains mosaic when the value of the determination value is larger than or equal to a preset threshold value;
or determining that the second detection result corresponding to the video to be detected indicates that the video to be detected does not contain mosaic under the condition that the value of the judgment value is smaller than a preset threshold value.
According to the video mosaic detection method provided by the invention, the method further comprises the following steps:
and determining the duration of the mosaic in the video to be detected based on the number of images indicating that the images are mosaic images in the first detection results corresponding to the images.
According to the video mosaic detection method provided by the invention, the method further comprises the following steps:
acquiring the mosaic image set and the conventional image set;
training images in the mosaic image set and the conventional image set based on the neural network to obtain at least one candidate model and the accuracy of each candidate model;
and determining the mosaic image detection model from the candidate models based on the accuracy of the candidate models, wherein the accuracy of the mosaic image detection model is highest.
According to the video mosaic detection method provided by the invention, the neural network comprises a ResNet network.
The invention also provides a video mosaic detection system, which comprises:
the detection module is used for executing the following steps:
step 1, decoding a video to be detected according to frames to obtain a decoded single image;
step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
the determining module is used for determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the video mosaic detection method according to any one of the above when executing the program.
The invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a video mosaic detection method as described in any one of the above.
In the video mosaic detection method provided by the invention, the single image after the video to be detected is decoded is detected through the mosaic detection model, the first detection result of whether the corresponding single image contains the mosaic is output, the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network, and then based on the first detection result corresponding to each image, whether the video to be detected contains the mosaic can be determined, so that the false detection rate of the detection model on the image mosaic detection is reduced, the accuracy of the video mosaic detection is improved, and the application scene is expanded.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a video mosaic detection method provided by the present invention;
FIG. 2 is one of the judging flowcharts of the video mosaic detecting method provided by the present invention;
FIG. 3 is a second flowchart of a video mosaic detection method according to the present invention
Fig. 4 is a schematic structural diagram of a video mosaic detection system provided by the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. 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.
The video mosaic detection method, system, device and medium of the present invention are described below with reference to fig. 1-4.
Fig. 1 is a flow chart of a video mosaic detection method provided by the present invention, as shown in fig. 1, the method includes:
s1, decoding a video to be detected according to frames to obtain a decoded single image;
in this step, the video to be detected may be an on-demand video or a continuous live video. When the continuous live video is processed, the continuous live video can be regarded as a plurality of video-on-demand, and the video-on-demand is decoded according to frames to obtain a single image.
S2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model;
in the step, a single image obtained after the video to be detected is decoded according to frames is obtained, the single image is input into a mosaic image detection model, and a detection result corresponding to the single image is output.
The mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
in specific implementation, the method comprises the steps of firstly decoding the video to be detected, detecting a single image, then decoding the next frame of the video to be detected after detection is completed, and continuing to detect until the video decoding is finished, wherein the acquired single image is empty.
S3, determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
In a specific implementation, after all decoded single images of the video to be detected are detected, a second detection result corresponding to the video to be detected is obtained, and the second detection result can represent whether a mosaic appears in the video to be detected.
According to the video mosaic detection method provided by the embodiment, the single image after the video to be detected is decoded is detected through the mosaic detection model, whether the corresponding single image contains a first detection result of the mosaic is output, the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network, and then whether the video to be detected contains the mosaic can be determined based on the first detection result corresponding to each image, so that the false detection rate of the detection model on the image mosaic is reduced, the accuracy of video mosaic detection is improved, and the application scene is expanded. According to the invention, the training data is not required to be preprocessed, and the training sample is divided into the mosaic image and the conventional image for training through the neural network by carrying out image two-classification, so that the false detection rate of the detection model on the image mosaic detection is reduced.
In an optional embodiment, in the video mosaic detection method, the mosaic image set includes a plurality of historical mosaic images;
the conventional image set comprises a plurality of historical normal images and historical misjudgment mosaic images;
wherein the mosaic image set is the same as the number of images in the regular image set.
Training samples based on neural network training are divided into two main categories, including mosaic image sets and conventional image sets. Mosaic images are randomly collected, the resolution of the images is not limited, and the mosaic images are placed in a 0_mosaic_imgs folder, namely a mosaic image set. And randomly selecting some historical normal images, collecting some normal images which are easy to misdetect as mosaics in the prior art, namely, the historical misjudged mosaic images, wherein the resolution of the images is not limited, and putting the images into a 1_good_imgs folder, namely, a conventional image set. The number of images in folders 0_mosaic_imgs and 1_good_imgs is the same. In a specific implementation, the selection of the training samples can be optimized for multiple times, so as to realize the maximum improvement of the training effect of the detection model.
According to the video mosaic detection method provided by the embodiment, the mosaic image is used as a large class, the normal image and the image which is easy to be misdetected as the mosaic are used as a large class, and the images are used as samples for model training, so that the misdetection rate of the final training finished detection model on the mosaic image detection is reduced.
In an optional embodiment, in the video mosaic detection method, determining, based on the first detection result corresponding to each image, a second detection result corresponding to the video to be detected includes:
determining that a second detection result corresponding to the video to be detected indicates that the video to be detected contains mosaic under the condition that the number of images, which indicate that the images are mosaic images, in the first detection result corresponding to each image is greater than or equal to a preset threshold value;
or determining that the second detection result corresponding to the video to be detected indicates that the video to be detected does not contain mosaic under the condition that the number of images, which indicate that the images are mosaic images, in the first detection result corresponding to each image is smaller than a preset threshold value.
In a specific implementation, after the video decoding is completed, a plurality of first detection results obtained after all the single images of one video are detected are summarized, and if the number of images indicating that the images are mosaic images in the first detection results corresponding to the images is greater than or equal to a preset threshold value, whether the mosaic occurs in the video can be judged.
According to the video mosaic detection method, through the trained detection model, judgment of the instantaneous mosaic, namely the image mosaic, can be achieved, judgment of the video mosaic can be achieved through threshold judgment, and application scenes are expanded.
Fig. 2 is one of the judging flowcharts of the video mosaic detection method provided by the present invention, as shown in fig. 2, after the image is input into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model, the method further includes:
adding one to the value of the judgment value under the condition that the image is indicated to be a mosaic image in the first detection result corresponding to the image;
correspondingly, the determining the second detection result corresponding to the video to be detected based on the first detection result corresponding to each image includes:
determining that a second detection result corresponding to the video to be detected indicates that the video to be detected contains mosaic when the value of the determination value is larger than or equal to a preset threshold value;
or determining that the second detection result corresponding to the video to be detected indicates that the video to be detected does not contain mosaic under the condition that the value of the judgment value is smaller than a preset threshold value.
In a specific implementation, a decision value for indicating that a single image contains a mosaic is initialized to 0, and when a first detection result of a current single image indicates that the single image contains the mosaic, the single image is detected by decoding a next frame of a video to be detected, wherein the count is equal to 1. After the video to be detected is decoded, a final count value is subjected to threshold judgment, if the count value is greater than or equal to a preset threshold value, the video to be detected is determined to contain mosaics, in a specific implementation, the preset threshold value can be set to 10, namely the count value records the number of times that a video clip appears, and when the count value is greater than or equal to 10 times, the video clip can be determined to contain the mosaics.
According to the video mosaic detection method, the threshold judgment mode is added, so that mosaic detection judgment of a video can be realized, the detection process is simple, the speed is high, and the detection efficiency of the video mosaic is improved to a certain extent.
In an optional embodiment, the video mosaic detection method further includes:
and determining the duration of the mosaic in the video to be detected based on the number of images indicating that the images are mosaic images in the first detection results corresponding to the images.
FIG. 3 is a second flowchart of a video mosaic detection method according to the present invention;
in a specific implementation, the detection method of the embodiment may also be applied to live video. For live video, continuous on-demand video can be considered as continuous mosaic detection. The purpose of the detection is whether a continuous mosaic appears in the live video, and a specific judgment method is shown in fig. 3.
In the specific implementation, adding the determination of the num value, initializing the num value to 0, and recording the number of live video fragments with continuous mosaics.
Obtaining a current video to be detected, performing mosaic detection on the decoded single image by using a trained mosaic detection model, if a first detection result indicates that the current single image contains a mosaic, then carrying out single image decoding on the current video to be detected and detecting that the current video decoding is finished if the first detection result indicates that the current single image does not contain the mosaic, and carrying out threshold judgment on a final count value, and if the final count value is greater than or equal to a preset threshold, recording num+1, and performing decoding and detection on the next video to be detected of the live video;
if the count value of the first video is smaller than a preset threshold value until decoding is completed and the final count values of three continuous videos to be detected in the live video are smaller than the preset threshold value, resetting the num value to 0;
continuing to decode and detect all videos to be detected in the live video until the decoding and detection of all the videos to be detected in the live video are finished, judging whether the num value is greater than or equal to a num set threshold value or not, and if so, judging that continuous mosaics appear in the live video; if not, judging that the continuous mosaic does not appear in the live video. In an alternative embodiment, the num set threshold is 4, and if the num value is greater than or equal to 4, the live video is considered to have a continuous mosaic, and the num value may calculate the duration of the occurrence of the mosaic. In a specific implementation, the total length of the whole live video is 50s, the live video is divided into 5 video segments of 10s, detection is carried out, the final count value after detection is 4, then the live video is considered to have continuous mosaic, and the duration of the mosaic is confirmed to be 40s. According to the video mosaic detection method provided by the embodiment, mosaic detection of live video can be realized through the increased num threshold judgment method, and the application scene of the method of the embodiment is expanded.
In an optional embodiment, the video mosaic detection method further includes:
acquiring the mosaic image set and the conventional image set;
training images in the mosaic image set and the conventional image set based on the neural network to obtain at least one candidate model and the accuracy of each candidate model;
and determining the mosaic image detection model from the candidate models based on the accuracy of the candidate models, wherein the accuracy of the mosaic image detection model is highest.
In specific implementation, the detection model is trained and tested for multiple times based on the neural network by optimizing the selection of training samples, and the detection model with the highest detection accuracy is selected as the final output.
In an alternative embodiment, the above video mosaic detection method, the neural network includes a res net network.
In a specific implementation, the present embodiment trains the detection model using a residual network (res net network), which is characterized by easy optimization and can improve accuracy by increasing a considerable depth. The residual blocks inside the deep neural network are connected in a jumping mode, and the gradient disappearance problem caused by depth increase in the deep neural network is relieved.
According to the video mosaic detection method provided by the embodiment, the ResNet network is adopted for training the detection model, the detection process is simple, the algorithm complexity is low, the detection speed is high, the detection time of one frame 1920 x 1080 image on the NVIDIA GeForce GTX 1060+Intel Core i7-7700HQ CPU is about 40ms, on the basis of ensuring the detection accuracy, the training sample is not required to be preset, the detection efficiency is improved, the detection timeliness is ensured, and the application scene is expanded.
The video mosaic detection system provided by the invention is described below, and the video mosaic detection system described below and the video mosaic detection method described above can be referred to correspondingly.
Fig. 4 is a schematic structural diagram of a video mosaic detection system provided by the present invention, as shown in fig. 4, the system includes:
the detection module 41 is configured to perform the following steps:
step 1, decoding a video to be detected according to frames to obtain a decoded single image;
step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
a determining module 42, configured to determine a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
According to the video mosaic detection system, through the mutual coordination among the modules, mosaic detection is carried out on a single image after video decoding through the detection model in the detection module, and the detection judgment of the video mosaic is realized through the threshold judgment in the determination module, so that the improvement of the detection accuracy of the video mosaic is realized, the structure is simple, and the detection speed is high.
Fig. 5 illustrates a physical schematic diagram of an electronic device, as shown in fig. 5, which may include: processor 510, communication interface (Communications Interface) 520, memory 630, and communication bus 540, wherein processor 510, communication interface 520, and memory 530 communicate with each other via communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a video mosaic detection method comprising:
step 1, decoding a video to be detected according to frames to obtain a decoded single image; step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
Further, the logic instructions in the memory 530 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the video mosaic detection method provided by the above methods, the method comprising:
step 1, decoding a video to be detected according to frames to obtain a decoded single image; step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A video mosaic detection method, comprising:
step 1, decoding a video to be detected according to frames to obtain a decoded single image;
step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
2. The video mosaic detection method according to claim 1, wherein said mosaic image set comprises a plurality of historical mosaic images;
the conventional image set comprises a plurality of historical normal images and historical misjudgment mosaic images;
wherein the mosaic image set is the same as the number of images in the regular image set.
3. The method according to claim 1, wherein the determining the second detection result corresponding to the video to be detected based on the first detection result corresponding to each image includes:
determining that a second detection result corresponding to the video to be detected indicates that the video to be detected contains mosaic under the condition that the number of images, which indicate that the images are mosaic images, in the first detection result corresponding to each image is greater than or equal to a preset threshold value;
or determining that the second detection result corresponding to the video to be detected indicates that the video to be detected does not contain mosaic under the condition that the number of images, which indicate that the images are mosaic images, in the first detection result corresponding to each image is smaller than a preset threshold value.
4. The video mosaic detection method according to claim 1, wherein after said inputting the image to a mosaic image detection model, obtaining a first detection result corresponding to the image output by the mosaic image detection model, the method further comprises:
adding one to the value of the judgment value under the condition that the image is indicated to be a mosaic image in the first detection result corresponding to the image;
correspondingly, the determining the second detection result corresponding to the video to be detected based on the first detection result corresponding to each image includes:
determining that a second detection result corresponding to the video to be detected indicates that the video to be detected contains mosaic when the value of the determination value is larger than or equal to a preset threshold value;
or determining that the second detection result corresponding to the video to be detected indicates that the video to be detected does not contain mosaic under the condition that the value of the judgment value is smaller than a preset threshold value.
5. The video mosaic detection method according to claim 1, wherein said method further comprises:
and determining the duration of the mosaic in the video to be detected based on the number of images indicating that the images are mosaic images in the first detection results corresponding to the images.
6. The video mosaic detection method according to claim 1, wherein said method further comprises:
acquiring the mosaic image set and the conventional image set;
training images in the mosaic image set and the conventional image set based on the neural network to obtain at least one candidate model and the accuracy of each candidate model;
and determining the mosaic image detection model from the candidate models based on the accuracy of the candidate models, wherein the accuracy of the mosaic image detection model is highest.
7. The video mosaic detection method according to claim 1, wherein said neural network comprises a res net network.
8. A video mosaic detection system, comprising:
the detection module is used for executing the following steps:
step 1, decoding a video to be detected according to frames to obtain a decoded single image;
step 2, inputting the image into a mosaic image detection model to obtain a first detection result corresponding to the image output by the mosaic image detection model; the mosaic image detection model is obtained by taking a mosaic image set and a conventional image set as training samples and training based on a neural network; the first detection result is used for indicating that the image is a mosaic image or a conventional image;
repeating the step 1 and the step 2 until the video to be detected is decoded;
the determining module is used for determining a second detection result corresponding to the video to be detected based on the first detection result corresponding to each image; the second detection result is used for indicating whether the video to be detected contains mosaic or not.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the video mosaic detection method according to any one of claims 1 to 7 when executing the program.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the video mosaic detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310442540.4A CN116579990A (en) | 2023-04-23 | 2023-04-23 | Video mosaic detection method, system, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310442540.4A CN116579990A (en) | 2023-04-23 | 2023-04-23 | Video mosaic detection method, system, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116579990A true CN116579990A (en) | 2023-08-11 |
Family
ID=87533206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310442540.4A Pending CN116579990A (en) | 2023-04-23 | 2023-04-23 | Video mosaic detection method, system, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116579990A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117893952A (en) * | 2024-03-15 | 2024-04-16 | 视睿（杭州）信息科技有限公司 | Video mosaic defect detection method based on deep learning |
-
2023
- 2023-04-23 CN CN202310442540.4A patent/CN116579990A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117893952A (en) * | 2024-03-15 | 2024-04-16 | 视睿（杭州）信息科技有限公司 | Video mosaic defect detection method based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109756746A (en) | Video reviewing method, device, server and storage medium | |
| CN103945214A (en) | Terminal side time-domain video quality evaluation method and apparatus | |
| CN103281473A (en) | General video steganalysis method based on video pixel space-time relevance | |
| CN107886518B (en) | Picture detection method and device, electronic equipment and readable storage medium | |
| CN116579990A (en) | Video mosaic detection method, system, equipment and medium | |
| JP2012058787A (en) | Information processor and processing method thereof | |
| CN103067713B (en) | Method and system of bitmap joint photographic experts group (JPEG) compression detection | |
| CN114842524B (en) | Face false distinguishing method based on irregular significant pixel cluster | |
| KR20160107734A (en) | Method for classifying objectionable movies using duration information and apparatus for the same | |
| CN110555863A (en) | moving object detection method and device and computer readable storage medium | |
| US9268791B2 (en) | Method and apparatus for image processing and computer readable medium | |
| CN107251053A (en) | A kind of method and device for the compression artefacts for reducing lossy compression method image | |
| CN110121109A (en) | Towards the real-time source tracing method of monitoring system digital video, city video monitoring system | |
| CN112489103B (en) | High-resolution depth map acquisition method and system | |
| CN107180419B (en) | Median filtering detection method based on PCA (principal component analysis) network | |
| Le Moan et al. | Exploiting change blindness for image compression | |
| CN117333795A (en) | River surface flow velocity measurement method and system based on screening post-treatment | |
| WO2017113735A1 (en) | Video format distinguishing method and system | |
| CN103647963A (en) | Video quality evaluation method based on Gop scene complexity | |
| CN113542780B (en) | Method and device for removing compression artifacts of live webcast video | |
| CN116091862A (en) | Picture quality identification method, device, equipment, storage medium and product | |
| Wan et al. | A video forensic technique for detecting frame integrity using human visual system-inspired measure | |
| CN113628121B (en) | Method and device for processing and training multimedia data | |
| Hu et al. | SDM: Semantic distortion measurement for video encryption | |
| CN113554685A (en) | Method and device for detecting moving target of remote sensing satellite, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |