CN108833940A - Video type determines method, device and equipment - Google Patents
Video type determines method, device and equipment Download PDFInfo
- Publication number
- CN108833940A CN108833940A CN201810694958.3A CN201810694958A CN108833940A CN 108833940 A CN108833940 A CN 108833940A CN 201810694958 A CN201810694958 A CN 201810694958A CN 108833940 A CN108833940 A CN 108833940A
- Authority
- CN
- China
- Prior art keywords
- type
- image
- gray level
- level image
- video
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
Abstract
The embodiment of the present invention provides a kind of video type and determines that method, device and equipment, this method include：Multiple key frames of video are obtained, the data in the key frame are used to indicate the complete image of a width；Partial decoding of h is carried out to each key frame, obtains the gray level image of each key frame, the partial decoding of h includes entropy decoding, inverse quantization, inverse transformation and prediction processing；The type of the video is determined according to the gray level image of each key frame.For improving the determination efficiency of video type.
Description
Technical field
The present embodiments relate to video technique fields more particularly to a kind of video type to determine method, device and equipment.
Background technique
Currently, needing to classify to video according to the content of video under plurality of application scenes, for example, the class of video
Type may include personage's type, landscape type, type of animal etc..
In the prior art, when classifying to video, video solution first usually is carried out to each video frame in video
Code obtains the corresponding image of each video frame in video, then identifies to the corresponding image of each video frame, every to determine
The type of the corresponding image of a video frame, and determine according to the type of the corresponding image of each video frame the type of video.However,
In above process, each video frame in video is decoded and the corresponding image of each video frame is identified
It needs to consume more duration, and then causes to determine that the efficiency of video type is lower.
Summary of the invention
The embodiment of the present invention provides a kind of video type and determines method, device and equipment, improves the determination of video type
Efficiency.
In a first aspect, the embodiment of the present invention, which provides a kind of video type, determines method, including：
Multiple key frames of video are obtained, the data in the key frame are used to indicate the complete image of a width；
Partial decoding of h is carried out to each key frame, obtains the gray level image of each key frame, the partial decoding of h includes entropy
Decoding, inverse quantization, inverse transformation and prediction processing；
The type of the video is determined according to the gray level image of each key frame.
In a kind of possible embodiment, the multiple key frames for obtaining video, including：
Dissection process is carried out to the video, obtains multiple data blocks of the video, includes multiple in each data block
Video frame；
To there is the video frame of default mark to be determined as the key frame, alternatively, by each data block in each data block
Video frame on middle predeterminated position is determined as the key frame.
In alternatively possible embodiment, the gray level image according to each key frame determines the class of the video
Type, including：
Obtain the image type of each gray level image；
According to the image type of each gray level image, the corresponding gray level image number of each image type is determined；
According to the corresponding gray level image number of each image type, the type of the video is determined.
In alternatively possible embodiment, for any one first grayscale image in the multiple gray level image
Picture obtains the image type of first gray level image, including：
Obtain the fisrt feature information of first gray level image；
It is corresponding with each first pre-set image type in multiple first pre-set image types to obtain the fisrt feature information
Characteristic information similarity；
By in the multiple first pre-set image type, corresponding characteristic information and the fisrt feature information similarity are most
High pre-set image type is determined as the image type of first gray level image.
In alternatively possible embodiment, the gray level image according to each key frame determines the class of the video
Type, including：
The type information of each gray level image is obtained, the type information of a gray level image includes that the gray level image is corresponding
At least one image type and the gray level image are the probability of each image type；
According to the type information of each gray level image, the type of the video is determined.
In alternatively possible embodiment, for any one second grayscale image in the multiple gray level image
Picture obtains the type information of second gray level image, including：
Obtain the second feature information of second gray level image；
Obtain the similarity of second feature information characteristic information corresponding with multiple second pre-set image types；
According to the similarity of second feature information characteristic information corresponding with each second pre-set image type, determine
The image type of second gray level image is the probability of each second pre-set image type；
It is each second default figure according to the image type of each second pre-set image type and second gray level image
As the probability of type, the type information of second gray level image is determined.
In alternatively possible embodiment, the type information according to each gray level image determines the video
Type, including：
According to the type information of each gray level image, the sum of corresponding probability of each image type, an image class are determined
The sum of corresponding probability of type is that the image type of each gray level image is the sum of probability of the image type；
According to the sum of corresponding probability of each image type, the type of the video is determined.
Second aspect, the embodiment of the present invention provide a kind of video type determining device, including obtain module, decoder module and
Determining module, wherein
The acquisition module is used for, and obtains multiple key frames of video, and the data in the key frame are used to indicate a width
Complete image；
The decoder module is used for, and is carried out partial decoding of h to each key frame, is obtained the gray level image of each key frame, institute
Stating partial decoding of h includes entropy decoding, inverse quantization, inverse transformation and prediction processing；
The determining module is used for, and the type of the video is determined according to the gray level image of each key frame.
In a kind of possible embodiment, the acquisition module is specifically used for：
Dissection process is carried out to the video, obtains multiple data blocks of the video, includes multiple in each data block
Video frame；
To there is the video frame of default mark to be determined as the key frame, alternatively, by each data block in each data block
Video frame on middle predeterminated position is determined as the key frame.
In alternatively possible embodiment, the determining module is specifically used for：
Obtain the image type of each gray level image；
According to the image type of each gray level image, the corresponding gray level image number of each image type is determined；
According to the corresponding gray level image number of each image type, the type of the video is determined.
In alternatively possible embodiment, for any one first grayscale image in the multiple gray level image
Picture, the determining module are specifically used for：
Obtain the fisrt feature information of first gray level image；
It is corresponding with each first pre-set image type in multiple first pre-set image types to obtain the fisrt feature information
Characteristic information similarity；
By in the multiple first pre-set image type, corresponding characteristic information and the fisrt feature information similarity are most
High pre-set image type is determined as the image type of first gray level image.
In alternatively possible embodiment, the determining module is specifically used for：
The type information of each gray level image is obtained, the type information of a gray level image includes that the gray level image is corresponding
At least one image type and the gray level image are the probability of each image type；
According to the type information of each gray level image, the type of the video is determined.
In alternatively possible embodiment, for any one second grayscale image in the multiple gray level image
Picture, the determining module are specifically used for：
Obtain the second feature information of second gray level image；
Obtain the similarity of second feature information characteristic information corresponding with multiple second pre-set image types；
According to the similarity of second feature information characteristic information corresponding with each second pre-set image type, determine
The image type of second gray level image is the probability of each second pre-set image type；
It is each second default figure according to the image type of each second pre-set image type and second gray level image
As the probability of type, the type information of second gray level image is determined.
In alternatively possible embodiment, the determining module is specifically used for：
According to the type information of each gray level image, the sum of corresponding probability of each image type, an image class are determined
The sum of corresponding probability of type is that the image type of each gray level image is the sum of probability of the image type；
According to the sum of corresponding probability of each image type, the type of the video is determined.
The third aspect, the embodiment of the present invention provide a kind of terminal device, including：Processor, the processor and memory
Coupling；
The memory is used for, and stores computer program；
The processor is used for, and executes the computer program stored in the memory, so that the terminal device is held
The above-mentioned described in any item methods of first aspect of row.
Fourth aspect, the embodiment of the present invention provide a kind of readable storage medium storing program for executing, including program or instruction, when described program or
When instruction is run on computers, the method as described in above-mentioned first aspect any one is performed
Video type provided in an embodiment of the present invention determines method, device and equipment, when needing to obtain the type of video,
The multiple key frames for first obtaining video carry out partial decoding of h to each key frame, obtain the gray level image of each key frame, part
Decoding includes entropy decoding, inverse quantization, inverse transformation and prediction processing, and the class of video is determined according to the gray level image of each key frame
Type.In above process, key frame is a part in video frame, and the application is only decoded key frame, reduces decoding
Workload further during being decoded to key frame, only carries out partial decoding of h to key frame, is further reduced
Decoding workload, and then improve the determination efficiency of video type.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the system architecture diagram that video type provided in an embodiment of the present invention determines method；
Fig. 2 is the flow diagram that a kind of video type provided in an embodiment of the present invention determines method；
Fig. 3 is the flow diagram one of the type method provided in an embodiment of the present invention that video is determined according to gray level image；
Fig. 4 is the flow diagram two of the type method provided in an embodiment of the present invention that video is determined according to gray level image；
Fig. 5 is the structural schematic diagram of video type determining device provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 is the system architecture diagram that video type provided in an embodiment of the present invention determines method.Referring to Figure 1, terminal is set
It is standby to extract the key frame in video after acquiring video, and to key frame progress partial decoding of h, obtain key frame
Corresponding gray level image carries out image recognition to the corresponding gray level image of key frame, obtains the image type of each gray level image,
And the type of video is determined according to the image type of each gray level image.
In this application, key frame is a part in video frame, and the application is only decoded key frame, reduces and understands
Code workload further during being decoded to key frame, only carries out partial decoding of h to key frame, further subtracts
Lack decoding workload, and then improves the determination efficiency of video type.
In the following, technical solution shown in the application is described in detail by specific embodiment.Under it should be noted that
The several specific embodiments in face can be combined with each other, and for the same or similar content, no longer carry out weight in various embodiments
Multiple explanation.
Fig. 2 is the flow diagram that a kind of video type provided in an embodiment of the present invention determines method.Fig. 2 is referred to, it should
Method may include：
S201, the multiple key frames for obtaining video, the data in key frame are used to indicate the complete image of a width.
The executing subject of the embodiment of the present invention is terminal device, or the video type being arranged in terminal device is true
Determine device.
Optionally, terminal device can be mobile phone, apparatus such as computer.
Optionally, video type determining device can be by software realization, or being implemented in combination with by software and hardware.
Optionally, multiple key frames of video can be obtained by following feasible implementation：Video is parsed
Processing obtains multiple data blocks of video, includes multiple video frames in each data block；To there is pre- bidding in each data block
The video frame of knowledge is determined as key frame, alternatively, the video frame in each data block on predeterminated position is determined as key frame.
It include key frame and non-key frame in data block, the data in key frame are used to indicate the complete image of a width, non-
Data in key frame can not indicate the complete image of a width.
For example, in a video, it is assumed that first frame is key frame, and the second frame is non-key frame, and data are in the second frame
The difference of second frame and first frame in video.
It in actual application, include one or more key frames in a data block, key frame can be located at data
Predeterminated position in block, can will be pre- in data block in the case of this kind for example, key frame can be the first frame in data block
If the video frame on position is determined as key frame.Key frame in data block can also have default mark, correspondingly, can incite somebody to action
There is the video frame of default mark to be determined as key frame in database.
S202, partial decoding of h is carried out to each key frame, obtains the gray level image of each key frame, partial decoding of h includes entropy
Decoding, inverse quantization, inverse transformation and prediction processing.
Conventional video decoding (for example, h264 is decoded) generally includes following steps：It is entropy decoding, inverse quantization, inverse transformation, pre-
Survey processing and deblocking filtering, wherein deblocking filtering step is a complicated treatment process, needs the long period that could complete deblocking
Filtering.
And partial decoding of h shown in the application includes entropy decoding, inverse quantization, inverse transformation and prediction processing, that is, the application institute
The step of partial decoding of h shown only includes the part steps in conventional video decoding, simplifies video decoding, and then improve
The decoded efficiency of video.
For any one key frame, entropy decoding, inverse quantization, inverse transformation and prediction processing are successively carried out to key frame
Afterwards, the gray level image of available each key frame.Wherein, the primary picture of key frame is contained in the gray level image of key frame
Content.
S203, the type that video is determined according to the gray level image of each key frame.
Optionally, the type of video may include personage's type, landscape type, type of animal etc..
Certainly, in actual application, it can be set according to actual needs the type of video, the embodiment of the present invention is to this
It is not especially limited.
Optionally, image recognition can be carried out by the gray level image to each key frame respectively, with each gray level image of determination
Type determine the type of video and according to the type of each gray level image.
It should be noted that determining video to according to the gray level image of key frame in Fig. 3-embodiment shown in Fig. 4
The process of type is described in detail, and is no longer repeated herein.
Video type provided in an embodiment of the present invention determines method, when needing to obtain the type of video, first obtains video
Multiple key frames, to each key frame carry out partial decoding of h, obtain the gray level image of each key frame, partial decoding of h includes entropy
Decoding, inverse quantization, inverse transformation and prediction processing, the type of video is determined according to the gray level image of each key frame.In above-mentioned mistake
Cheng Zhong, key frame are a part in video frame, and the application is only decoded key frame, reduce decoding effort amount, into one
Step, during being decoded to key frame, partial decoding of h only is carried out to key frame, further reduces decoding effort
Amount, and then improve the determination efficiency of video type.
It, optionally, can be by following feasible implementation according to every on the basis of any one above-mentioned embodiment
The gray level image of a key frame determines the type of video, specifically, referring to Fig. 3-embodiment shown in Fig. 4.
Fig. 3 is the flow diagram one of the type method provided in an embodiment of the present invention that video is determined according to gray level image.
Fig. 3 is referred to, this method may include：
S301, the image type for obtaining each gray level image.
Optionally, the image type of gray level image may include personage's type, landscape type, type of animal etc..
Certainly, in actual application, it can be set according to actual needs the image type of gray level image, the present invention is real
It applies example and this is not especially limited.
Optionally, the process for obtaining the image type of each gray level image is identical, in the following, to obtain multiple gray level images
In any one the first gray level image image type for, the process for the image type for obtaining gray level image is carried out detailed
Illustrate, may include steps of A- step D：
Step A, the fisrt feature information of the first gray level image is obtained.
It should be noted that the of the first gray level image can be extracted by a variety of feasible implementations in the prior art
One characteristic information, the present invention is not especially limit this.
Step B, the corresponding characteristic information of multiple first pre-set image types is obtained.
Optionally, the first pre-set image type may include personage's type, landscape type, type of animal etc..
Certainly, in actual application, it can be set according to actual needs the first pre-set image type, the present invention is implemented
Example is not especially limited this.
Optionally, each corresponding characteristic information of the first pre-set image type can be pre-generated, and in default storage
Position stores each corresponding characteristic information of the first pre-set image type.Correspondingly, when needing using the first pre-set image class
When the corresponding characteristic information of type, the corresponding characteristic information of the first pre-set image type directly is obtained in preset memory locations.
In actual application, the process for generating each corresponding characteristic information of the first pre-set image type is identical,
In the following, being illustrated by taking the process for the characteristic information for generating any one the first pre-set image type as an example：
The corresponding multiple sample gray level images of the first pre-set image type are obtained, it is corresponding to extract each sample gray level image
Characteristic information generates the corresponding characteristic information of the first pre-set image type according to the characteristic information of each sample gray level image.
For example, the corresponding characteristic information of multiple sample gray level images can be learnt by deep neural network, with
Obtain the corresponding characteristic information of the first pre-set image type.
Step C, the similarity of fisrt feature information characteristic information corresponding with each first pre-set image type is obtained.
It should be noted that fisrt feature information and every can be obtained by a variety of feasible implementations in the prior art
The similarity of the corresponding characteristic information of one the first pre-set image type, the present invention is not especially limit this.
Step D, by multiple first pre-set image types, corresponding characteristic information and fisrt feature information similarity highest
Pre-set image type be determined as the image type of the first gray level image.
S302, according to the image type of each gray level image, determine the corresponding gray level image number of each image type.
Optionally, the image type of each gray level image can be counted, it is corresponding with each image type of determination
Gray level image number.
S303, according to the corresponding gray level image number of each image type, determine the type of video.
Optionally, the most multiple image types of gray level image number can be determined as to the type of video.
In the following, technical solution shown in Fig. 3 embodiment is described in detail by specific example.
Illustratively, it is assumed that it extracts obtain 10 key frames in video, correspondingly, the number of gray level image is 10,
It is denoted as gray level image 1- gray level image 10 respectively, it is assumed that the image type of determining gray level image 1- gray level image 10 such as 1 institute of table
Show：
Table 1
| The mark of gray level image | The image type of gray level image |
| Gray level image 1 | Image type 1 |
| Gray level image 2 | Image type 2 |
| Gray level image 3 | Image type 3 |
| Gray level image 4 | Image type 1 |
| Gray level image 5 | Image type 2 |
| Gray level image 6 | Image type 1 |
| Gray level image 7 | Image type 2 |
| Gray level image 8 | Image type 1 |
| Gray level image 9 | Image type 4 |
| Gray level image 10 | Image type 1 |
The image type of each gray level image shown in table 1 is counted, the corresponding gray scale of each image type is obtained
Image number is as shown in table 2：
Table 2
| The image type of gray level image | Gray level image number |
| Image type 1 | 5 |
| Image type 2 | 3 |
| Image type 3 | 1 |
| Image type 4 | 1 |
As shown in Table 2, the corresponding gray level image number of image type 1 is most, therefore, image type 1 can be determined as
The type of video.
For example, it is assumed that image type 1 is figure kind's type, then the type of video is also figure kind's type.
In the embodiment shown in fig. 3, the image type for first determining each gray level image, according to the figure of each gray level image
As type determines the corresponding gray level image number of each image type, and the largest number of image types of gray level image are true
It is set to the type of video.That is, determining that obtained video type is the image type of most images in video.
Fig. 4 is the flow diagram two of the type method provided in an embodiment of the present invention that video is determined according to gray level image.
Fig. 4 is referred to, this method may include：
S401, the type information for obtaining each gray level image.
Wherein, the type information of a gray level image includes at least one corresponding image type of the gray level image and the ash
Spend the probability that image is each image type.
In actual application, the process for obtaining the type information of each gray level image is identical, in the following, more to obtain
For the process of the type information of any one the second gray level image in a gray level image, the type for obtaining gray level image is believed
The process of breath is described in detail, and may include steps of A- step D：
Step A, the second feature information of the second gray level image is obtained.
It should be noted that the of the first gray level image can be extracted by a variety of feasible implementations in the prior art
One characteristic information, the present invention is not especially limit this.
Step B, the corresponding characteristic information of multiple second pre-set image types is obtained.
Optionally, the second pre-set image type may include personage's type, landscape type, type of animal etc..
Certainly, in actual application, it can be set according to actual needs the first pre-set image type, the present invention is implemented
Example is not especially limited this.
Optionally, each corresponding characteristic information of the second pre-set image type can be pre-generated, and in default storage
Position stores each corresponding characteristic information of the second pre-set image type.Correspondingly, when needing using the second pre-set image class
When the corresponding characteristic information of type, the corresponding characteristic information of the second pre-set image type directly is obtained in preset memory locations.
In actual application, the process for generating each corresponding characteristic information of the second pre-set image type is identical,
In the following, being illustrated by taking the process for the characteristic information for generating any one the second pre-set image type as an example：
The corresponding multiple sample gray level images of the second pre-set image type are obtained, it is corresponding to extract each sample gray level image
Characteristic information generates the corresponding characteristic information of the second pre-set image type according to the characteristic information of each sample gray level image.
For example, the corresponding characteristic information of multiple sample gray level images can be learnt by deep neural network, with
Obtain the corresponding characteristic information of the second pre-set image type.
Step C, the similarity of second feature information characteristic information corresponding with each second pre-set image type is obtained.
It should be noted that second feature information and every can be obtained by a variety of feasible implementations in the prior art
The similarity of the corresponding characteristic information of one the second pre-set image type, the present invention is not especially limit this.
Step D, according to the similarity of second feature information characteristic information corresponding with each second pre-set image type, really
The image type of fixed second gray level image is the probability of each second pre-set image type.
Optionally, the similarity of second feature information characteristic information corresponding with the second pre-set image type is higher, and second
The image type of gray level image is that the probability of the second pre-set image type is bigger.
It step E, is each second default according to the image type of each second pre-set image type and the second gray level image
The probability of image type determines the type information of the second gray level image.
For example, it is assumed that the characteristic information of the second gray level image is characterized information 1, there are 5 the second pre-set image types, divide
It is not denoted as pre-set image Class1-pre-set image type 5, the corresponding characteristic information of this 5 the second pre-set image types is denoted as respectively
Default characteristic information 1- default characteristic information 5, it is further assumed that between characteristic information 1 and default characteristic information 1- default characteristic information 5
Similarity it is as shown in table 3：
Table 3
| The mark of characteristic information | The mark of default characteristic information | Similarity |
| Characteristic information 1 | Default characteristic information 1 | 0.2 |
| Characteristic information 1 | Default characteristic information 2 | 0.3 |
| Characteristic information 1 | Default characteristic information 3 | 0.9 |
| Characteristic information 1 | Default characteristic information 4 | 0.1 |
| Characteristic information 1 | Default characteristic information 5 | 0.15 |
As shown in Table 3, the second gray level image is that pre-set image Class1-pre-set image type 5 probability is as shown in table 4：
Table 4
| The mark of pre-set image type | Second gray level image is the probability of the pre-set image type |
| Pre-set image Class1 | 0.2 |
| Pre-set image type 2 | 0.3 |
| Pre-set image type 3 | 0.9 |
| Pre-set image type 4 | 0.1 |
| Pre-set image type 5 | 0.15 |
As shown in Table 4, the type information of the second gray level image includes：Pre-set image Class1,0.2；Pre-set image type 2,
0.3；Pre-set image type 3,0.9；Pre-set image type 4,0.1；Pre-set image type 5,0.15；
S402, according to the type information of each gray level image, determine the type of video.
Optionally, the type of video can be determined by following feasible implementation：According to the class of each gray level image
Type information determines the sum of corresponding probability of each image type, according to the sum of corresponding probability of each image type, determines video
Type.Wherein, it is the image type that the sum of corresponding probability of an image type, which is the image type of each gray level image,
The sum of probability；
Optionally, the highest image type of the sum of probability can be determined as to the type of video.
For example, it is assumed that there are 5 gray level images, it is denoted as gray level image 1- gray level image 5 respectively, it is assumed that the figure of gray level image 1
As the probability that type is image type 1 is 0.1, the image type of gray level image 2 is that the probability of image type 1 is 0.15, gray scale
The image type of image 3 is that the probability of image type 1 is 0.1, and the image type of gray level image 4 is that the probability of image type 1 is
0.2, the image type of gray level image 5 is that the probability of image type 1 is 0.3, then the sum of corresponding probability of image type 1 is：0.1
+ 0.15+0.1+0.2+0.3=0.85.
In the following, method shown in Fig. 4 embodiment is described in detail by specific example.
Illustratively, it is assumed that the number of gray level image is 5, is denoted as gray level image 1- gray level image 5 respectively, and second is default
The number of image type is 3, respectively personage's type, landscape type and type of animal, the class of gray level image 1- gray level image 5
Type information is as shown in table 5：
Table 5
According to the type information of gray level image 1- gray level image 5 shown in table 5, it can determine that pre-set image Class1-is default
The sum of probability of image type 3 is as shown in table 6：
Table 6
| Pre-set image type | The sum of probability |
| Personage's type | 0.2+0.1+0.4+0.15+0.25=1.1 |
| Landscape type | 0.8+0.9+0.7+0.85+0.7=3.95 |
| Type of animal | 0.1+0.15+0.2+0.1+0.1=0.65 |
The sum of corresponding probability of each pre-set image type according to shown in table 6 can determine that the type of video is landscape
Type.
In the embodiment shown in fig. 4, first obtaining each gray level image is that (image type can be each image type
Pre-set image type) probability, then count the sum of corresponding probability of each image type, and corresponding according to each image type
The sum of probability, determine the type of video, avoid the image type for approximately determining gray level image, avoid deviation accumulation, into
And it can to determine that the accuracy of obtained video type is higher.
Fig. 5 is the structural schematic diagram of video type determining device provided in an embodiment of the present invention.Fig. 5 is referred to, the device
Including obtaining module 11, decoder module 12 and determining module 13, wherein
The acquisition module 11 is used for, and obtains multiple key frames of video, and the data in the key frame are used to indicate one
The complete image of width；
The decoder module 12 is used for, and is carried out partial decoding of h to each key frame, is obtained the gray level image of each key frame,
The partial decoding of h includes entropy decoding, inverse quantization, inverse transformation and prediction processing；
The determining module 13 is used for, and the type of the video is determined according to the gray level image of each key frame.
Video type determining device provided in an embodiment of the present invention can execute technical side shown in above method embodiment
Case, realization principle and beneficial effect are similar, are no longer repeated herein.
In a kind of possible embodiment, the acquisition module 11 is specifically used for：
Dissection process is carried out to the video, obtains multiple data blocks of the video, includes multiple in each data block
Video frame；
To there is the video frame of default mark to be determined as the key frame, alternatively, by each data block in each data block
Video frame on middle predeterminated position is determined as the key frame.
In alternatively possible embodiment, the determining module 13 is specifically used for：
Obtain the image type of each gray level image；
According to the image type of each gray level image, the corresponding gray level image number of each image type is determined；
According to the corresponding gray level image number of each image type, the type of the video is determined.
In alternatively possible embodiment, for any one first grayscale image in the multiple gray level image
Picture, the determining module 13 are specifically used for：
Obtain the fisrt feature information of first gray level image；
It is corresponding with each first pre-set image type in multiple first pre-set image types to obtain the fisrt feature information
Characteristic information similarity；
By in the multiple first pre-set image type, corresponding characteristic information and the fisrt feature information similarity are most
High pre-set image type is determined as the image type of first gray level image.
In alternatively possible embodiment, the determining module 13 is specifically used for：
The type information of each gray level image is obtained, the type information of a gray level image includes that the gray level image is corresponding
At least one image type and the gray level image are the probability of each image type；
According to the type information of each gray level image, the type of the video is determined.
In alternatively possible embodiment, for any one second grayscale image in the multiple gray level image
Picture, the determining module 13 are specifically used for：
Obtain the second feature information of second gray level image；
Obtain the similarity of second feature information characteristic information corresponding with multiple second pre-set image types；
According to the similarity of second feature information characteristic information corresponding with each second pre-set image type, determine
The image type of second gray level image is the probability of each second pre-set image type；
It is each second default figure according to the image type of each second pre-set image type and second gray level image
As the probability of type, the type information of second gray level image is determined.
In alternatively possible embodiment, the determining module 13 is specifically used for：
According to the type information of each gray level image, the sum of corresponding probability of each image type, an image class are determined
The sum of corresponding probability of type is that the image type of each gray level image is the sum of probability of the image type；
According to the sum of corresponding probability of each image type, the type of the video is determined.
Video type determining device provided in an embodiment of the present invention can execute technical side shown in above method embodiment
Case, realization principle and beneficial effect are similar, are no longer repeated herein.
The embodiment of the present invention also provides a kind of terminal device, including：Processor and memory, the processor are deposited with described
Reservoir coupling；
The memory is used for, and stores computer program；
The processor is used for, and executes the computer program stored in the memory, so that the terminal device is held
Technical solution shown in any one above-mentioned embodiment of the method for row.
The embodiment of the present invention also provides a kind of readable storage medium storing program for executing, including program or instruction, when described program or instruction exist
When running on computer, technical solution shown in any one above-mentioned embodiment of the method is performed.
Those of ordinary skill in the art will appreciate that：Realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence；And storage medium above-mentioned includes：ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally it should be noted that：The above various embodiments is only to illustrate the technical solution of the embodiment of the present invention, rather than to it
Limitation；Although the embodiment of the present invention is described in detail referring to foregoing embodiments, those skilled in the art
It should be understood that：It is still possible to modify the technical solutions described in the foregoing embodiments, either to part of or
All technical features are equivalently replaced；And these are modified or replaceed, it does not separate the essence of the corresponding technical solution this hair
The range of bright example scheme.
Claims (16)
1. a kind of video type determines method, which is characterized in that including：
Multiple key frames of video are obtained, the data in the key frame are used to indicate the complete image of a width；
Partial decoding of h is carried out to each key frame, obtains the gray level image of each key frame, the partial decoding of h include entropy decoding,
Inverse quantization, inverse transformation and prediction processing；
The type of the video is determined according to the gray level image of each key frame.
2. the method according to claim 1, wherein it is described obtain video multiple key frames, including：
Dissection process is carried out to the video, obtains multiple data blocks of the video, includes multiple videos in each data block
Frame；
To there is the video frame of default mark to be determined as the key frame, alternatively, by pre- in each data block in each data block
If the video frame on position is determined as the key frame.
3. the method according to claim 1, wherein described according to the determination of the gray level image of each key frame
The type of video, including：
Obtain the image type of each gray level image；
According to the image type of each gray level image, the corresponding gray level image number of each image type is determined；
According to the corresponding gray level image number of each image type, the type of the video is determined.
4. according to the method described in claim 3, it is characterized in that, in the multiple gray level image any one first
Gray level image obtains the image type of first gray level image, including：
Obtain the fisrt feature information of first gray level image；
Obtain fisrt feature information spy corresponding with the first pre-set image type each in multiple first pre-set image types
The similarity of reference breath；
By in the multiple first pre-set image type, corresponding characteristic information and the fisrt feature information similarity are highest
Pre-set image type is determined as the image type of first gray level image.
5. the method according to claim 1, wherein described according to the determination of the gray level image of each key frame
The type of video, including：
The type information of each gray level image is obtained, the type information of a gray level image includes that the gray level image is corresponding at least
One image type and the gray level image are the probability of each image type；
According to the type information of each gray level image, the type of the video is determined.
6. according to the method described in claim 5, it is characterized in that, in the multiple gray level image any one second
Gray level image obtains the type information of second gray level image, including：
Obtain the second feature information of second gray level image；
Obtain the similarity of second feature information characteristic information corresponding with multiple second pre-set image types；
According to the similarity of second feature information characteristic information corresponding with each second pre-set image type, determine described in
The image type of second gray level image is the probability of each second pre-set image type；
It is each second pre-set image class according to the image type of each second pre-set image type and second gray level image
The probability of type determines the type information of second gray level image.
7. method according to claim 5 or 6, which is characterized in that the type information according to each gray level image, really
The type of the fixed video, including：
According to the type information of each gray level image, the sum of corresponding probability of each image type, an image type pair are determined
The sum of probability answered is that the image type of each gray level image is the sum of probability of the image type；
According to the sum of corresponding probability of each image type, the type of the video is determined.
8. a kind of video type determining device, which is characterized in that including obtaining module, decoder module and determining module, wherein
The acquisition module is used for, and obtains multiple key frames of video, it is complete that the data in the key frame are used to indicate a width
Image；
The decoder module is used for, and is carried out partial decoding of h to each key frame, is obtained the gray level image of each key frame, the portion
Decomposing code includes entropy decoding, inverse quantization, inverse transformation and prediction processing；
The determining module is used for, and the type of the video is determined according to the gray level image of each key frame.
9. device according to claim 8, which is characterized in that the acquisition module is specifically used for：
Dissection process is carried out to the video, obtains multiple data blocks of the video, includes multiple videos in each data block
Frame；
To there is the video frame of default mark to be determined as the key frame, alternatively, by pre- in each data block in each data block
If the video frame on position is determined as the key frame.
10. device according to claim 8, which is characterized in that the determining module is specifically used for：
Obtain the image type of each gray level image；
According to the image type of each gray level image, the corresponding gray level image number of each image type is determined；
According to the corresponding gray level image number of each image type, the type of the video is determined.
11. device according to claim 10, which is characterized in that for any one in the multiple gray level image
One gray level image, the determining module are specifically used for：
Obtain the fisrt feature information of first gray level image；
Obtain fisrt feature information spy corresponding with the first pre-set image type each in multiple first pre-set image types
The similarity of reference breath；
By in the multiple first pre-set image type, corresponding characteristic information and the fisrt feature information similarity are highest
Pre-set image type is determined as the image type of first gray level image.
12. device according to claim 8, which is characterized in that the determining module is specifically used for：
The type information of each gray level image is obtained, the type information of a gray level image includes that the gray level image is corresponding at least
One image type and the gray level image are the probability of each image type；
According to the type information of each gray level image, the type of the video is determined.
13. device according to claim 12, which is characterized in that for any one in the multiple gray level image
Two gray level images, the determining module are specifically used for：
Obtain the second feature information of second gray level image；
Obtain the similarity of second feature information characteristic information corresponding with multiple second pre-set image types；
According to the similarity of second feature information characteristic information corresponding with each second pre-set image type, determine described in
The image type of second gray level image is the probability of each second pre-set image type；
It is each second pre-set image class according to the image type of each second pre-set image type and second gray level image
The probability of type determines the type information of second gray level image.
14. device according to claim 12 or 13, which is characterized in that the determining module is specifically used for：
According to the type information of each gray level image, the sum of corresponding probability of each image type, an image type pair are determined
The sum of probability answered is that the image type of each gray level image is the sum of probability of the image type；
According to the sum of corresponding probability of each image type, the type of the video is determined.
15. a kind of terminal device, which is characterized in that including：Processor and memory, the processor and the memory coupling
It closes；
The memory is used for, and stores computer program；
The processor is used for, and executes the computer program stored in the memory, so that the terminal device right of execution
Benefit requires the described in any item methods of 1-7.
16. a kind of readable storage medium storing program for executing, which is characterized in that including program or instruction, when described program or instruct on computers
When operation, the described in any item methods of claim 1-7 are performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810694958.3A CN108833940A (en) | 2018-06-29 | 2018-06-29 | Video type determines method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810694958.3A CN108833940A (en) | 2018-06-29 | 2018-06-29 | Video type determines method, device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108833940A true CN108833940A (en) | 2018-11-16 |
Family
ID=64134130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810694958.3A Pending CN108833940A (en) | 2018-06-29 | 2018-06-29 | Video type determines method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108833940A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111782842A (en) * | 2020-06-23 | 2020-10-16 | 珠海格力智能装备有限公司 | Image retrieval method and device and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020097272A1 (en) * | 2001-01-24 | 2002-07-25 | Mitsumasa Tanaka | Image editing apparatus and image editing method |
| CN101790097A (en) * | 2010-03-05 | 2010-07-28 | 天津大学 | Method for detecting multiple times of compression and coding of digital video |
| CN105718861A (en) * | 2016-01-15 | 2016-06-29 | 北京市博汇科技股份有限公司 | Method and device for identifying video streaming data category |
| CN106162235A (en) * | 2016-08-17 | 2016-11-23 | 北京百度网讯科技有限公司 | Method and apparatus for Switch Video stream |
| CN107180074A (en) * | 2017-03-31 | 2017-09-19 | 北京奇艺世纪科技有限公司 | A kind of video classification methods and device |
-
2018
- 2018-06-29 CN CN201810694958.3A patent/CN108833940A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020097272A1 (en) * | 2001-01-24 | 2002-07-25 | Mitsumasa Tanaka | Image editing apparatus and image editing method |
| CN101790097A (en) * | 2010-03-05 | 2010-07-28 | 天津大学 | Method for detecting multiple times of compression and coding of digital video |
| CN105718861A (en) * | 2016-01-15 | 2016-06-29 | 北京市博汇科技股份有限公司 | Method and device for identifying video streaming data category |
| CN106162235A (en) * | 2016-08-17 | 2016-11-23 | 北京百度网讯科技有限公司 | Method and apparatus for Switch Video stream |
| CN107180074A (en) * | 2017-03-31 | 2017-09-19 | 北京奇艺世纪科技有限公司 | A kind of video classification methods and device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111782842A (en) * | 2020-06-23 | 2020-10-16 | 珠海格力智能装备有限公司 | Image retrieval method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ostrovski et al. | Autoregressive quantile networks for generative modeling | |
| US10586312B2 (en) | Method for image processing and video compression with sparse zone salient features | |
| CN108229314B (en) | Target person searching method and device and electronic equipment | |
| CN111670580B (en) | Progressive compressed domain computer vision and deep learning system | |
| US9105119B2 (en) | Anonymization of facial expressions | |
| CN109214337B (en) | Crowd counting method, device, equipment and computer readable storage medium | |
| CN110866958A (en) | Method for text to image | |
| CN110796512A (en) | Product information recommendation method and device, storage medium and computer equipment | |
| CN111445424B (en) | Image processing method, device, equipment and medium for processing mobile terminal video | |
| Gunasekar et al. | Face detection on distorted images augmented by perceptual quality-aware features | |
| CN112686193B (en) | Action recognition method and device based on compressed video and computer equipment | |
| Duan et al. | Optimizing JPEG quantization table for low bit rate mobile visual search | |
| Arora et al. | On knowledge distillation from complex networks for response prediction | |
| CN111179201A (en) | Video denoising method and electronic equipment | |
| CN105979283A (en) | Video transcoding method and device | |
| CN108833940A (en) | Video type determines method, device and equipment | |
| Fu et al. | Texture segmentation based video compression using convolutional neural networks | |
| CN111626178B (en) | Compressed domain video motion recognition method and system based on new spatio-temporal feature stream | |
| US20220417540A1 (en) | Encoding Device and Method for Utility-Driven Video Compression | |
| WO2023122132A2 (en) | Video and feature coding for multi-task machine learning | |
| JP2012257125A (en) | Image processing apparatus and image processing method | |
| CN111539263B (en) | Video face recognition method based on aggregation countermeasure network | |
| Hu et al. | Improved color image coding schemes based on single bit map block truncation coding | |
| CN116760983B (en) | Loop filtering method and device for video coding | |
| WO2023122149A2 (en) | Systems and methods for video coding of features using subpictures |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181116 |
|
| RJ01 | Rejection of invention patent application after publication |