CN113064679B - User-set character size adaptive matching system - Google Patents
User-set character size adaptive matching system Download PDFInfo
- Publication number
- CN113064679B CN113064679B CN202011409220.1A CN202011409220A CN113064679B CN 113064679 B CN113064679 B CN 113064679B CN 202011409220 A CN202011409220 A CN 202011409220A CN 113064679 B CN113064679 B CN 113064679B
- Authority
- CN
- China
- Prior art keywords
- character
- deduplication
- neural network
- convolutional neural
- user
- 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.)
- Expired - Fee Related
Links
- 230000003044 adaptive effect Effects 0.000 title claims description 24
- 230000007246 mechanism Effects 0.000 claims abstract description 59
- 238000013527 convolutional neural network Methods 0.000 claims description 50
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 13
- 230000010354 integration Effects 0.000 claims description 7
- 230000008030 elimination Effects 0.000 claims description 6
- 238000003379 elimination reaction Methods 0.000 claims description 6
- 238000012015 optical character recognition Methods 0.000 claims description 6
- 238000013500 data storage Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 description 17
- 238000013529 biological neural network Methods 0.000 description 5
- 210000004556 brain Anatomy 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 241000257303 Hymenoptera Species 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000003796 beauty Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000016776 visual perception Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000003710 cerebral cortex Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 210000000225 synapse Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- 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
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/2187—Live feed
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Human Computer Interaction (AREA)
- Databases & Information Systems (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
- Character Discrimination (AREA)
Abstract
The invention relates to a self-adaptive matching system for setting character size by a user, which comprises: the distance measuring mechanism is arranged at a main broadcasting client end of the live broadcasting platform, is positioned on a shell of a camera for shooting main broadcasting pictures, and is used for measuring the real-time distance from a main coding part to the camera; the error notification equipment is realized by adopting a display mechanism or a voice playing mechanism and is used for carrying out corresponding error notification operation when the difference value between the received real-time distance and the optimal distance from the main broadcasting to the camera at the current moment exceeds the limit; and the content acquisition equipment is used for acquiring the current live broadcast picture of the anchor client. The self-adaptive matching system for the character size set by the user is effective in identification and timely in control. The broadcasting distance of the anchor can be adaptively adjusted based on the display character size set by each live user in the current live broadcast picture, so that the sizes of the characters of the live broadcast picture and the anchor image are matched.
Description
Technical Field
The invention relates to the field of neural networks, in particular to a self-adaptive matching system for a character size set by a user.
Background
Artificial Neural Networks (ans), also referred to as Neural Networks (NNs) or Connection models (Connection models), are algorithmic mathematical models that Model animal Neural network behavior characteristics and perform distributed parallel information processing. The network achieves the aim of processing information by adjusting the mutual connection relationship among a large number of nodes in the network depending on the complexity of the system.
The biological neural network mainly refers to the neural network of human brain, which is the technical prototype of artificial neural network. The human brain is the material basis for human thinking, whose function is localized in the cerebral cortex, which contains about 10^11 neurons, each of which is connected to about 103 other neurons through a neural synapse, forming a highly complex, highly flexible dynamic network. As a subject, the biological neural network mainly studies the structure, function and working mechanism of the human brain neural network, and aims to explore the law of human brain thinking and intelligent activities.
The artificial neural network is a technological recurrence of biological neural network under a certain simplification meaning, and as a subject, the artificial neural network is mainly used for building a practical artificial neural network model according to the principle of the biological neural network and the requirement of practical application, designing a corresponding learning algorithm, simulating certain intelligent activity of human brain, and then technically realizing the artificial neural network for solving the practical problem. Therefore, biological neural networks mainly study the mechanism of intelligence; the artificial neural network mainly researches the realization of an intelligent mechanism, and the two supplement each other.
The artificial neural network can be applied to various application fields requiring artificial intelligence. In the prior art, due to the rapid development of live broadcast application, the design of software and hardware of a live broadcast platform tends to be perfect. However, since the pursuit of the live broadcast picture is not only the clarity of the picture but also the beauty and harmony of the picture, for example, the size of the characters of the subtitles set by each viewer in the live broadcast picture is different in the prior art, so that it is difficult to match the size of the anchor character with the size of the characters displayed on the picture at each broadcast time.
Disclosure of Invention
The invention has at least the following two important points:
(1) the intelligent matching of the anchor face size and each character size set by each live audience is realized by adopting a convolutional neural network, wherein the imaging area grades of a plurality of characters obtained after screening each character appearing in a current live screen are used as input data of an input layer of the convolutional neural network, and the optimal distance from the anchor to a camera is used as output data of an output layer of the convolutional neural network;
(2) in a specific screening operation of each character appearing in a current live broadcast screen, performing deduplication processing on each character appearing in a current live broadcast picture to obtain a deduplication character set, and performing the following processing on each deduplication character in the deduplication character set: the duplication removing characters correspond to more than one image block in the current live broadcast picture, and the image block closest to the center position of the current live broadcast picture is used as a representative image block of the duplication removing characters; and taking each area grade corresponding to each representative image block of each de-duplicated character in the de-duplicated character set as input data of an input layer of the convolutional neural network.
According to an aspect of the present invention, there is provided a user-set character size adaptive matching system, the system comprising:
the distance measuring mechanism is arranged at a main broadcasting client end of the live broadcasting platform, is positioned on a shell of a camera for shooting main broadcasting pictures, and is used for measuring the real-time distance from a main coding part to the camera;
the error notification equipment is realized by adopting a display mechanism or a voice playing mechanism, is respectively connected with the network operation mechanism and the distance measurement mechanism, and is used for carrying out corresponding error notification operation when the difference value between the received real-time distance and the optimal distance from the main broadcasting to the camera at the current moment exceeds the limit;
the content acquisition equipment is arranged at a main broadcast client end of a live broadcast platform, is connected with the main broadcast client end and is used for acquiring a current live broadcast picture of the main broadcast client end;
the first analysis mechanism is connected with the content acquisition equipment and used for identifying each character in the current live broadcast picture based on OCR (optical character recognition), so as to obtain an image block of each character in the current live broadcast picture, wherein the image block is a rectangular image area only comprising corresponding characters;
the second analysis mechanism is connected with the first analysis mechanism and is used for executing the following processing on the image blocks of each character in the current live broadcast picture: analyzing an area grade of the image patch based on a length of the image patch and a height of the image patch;
the frequency identification equipment is respectively connected with the first analysis mechanism and the second analysis mechanism and is used for carrying out duplication elimination processing on each character appearing in the current live broadcast picture so as to obtain a duplication elimination character set;
a block selection device, connected to the frequency identification device, for performing the following processing for each of the deduplication characters in the deduplication character set: the duplication removing characters correspond to more than one image block in the current live broadcast picture, and the image block closest to the center position of the current live broadcast picture is used as a representative image block of the duplication removing characters;
the data integration equipment is respectively connected with the second analysis mechanism and the block selection equipment and is used for taking each area grade corresponding to each representative image block with a set number as input data of an input layer of the convolutional neural network;
the network operation mechanism is used for operating the convolutional neural network to obtain the optimal distance from the anchor to the camera at the current moment, and the optimal distance from the anchor to the camera is the output data of the convolutional neural network output layer;
the network training equipment is connected with the network operation mechanism and used for training the convolutional neural network by taking each area grade as input data of an input layer of the convolutional neural network and simultaneously taking the optimal distance from the main player to the camera as output data of an output layer of the convolutional neural network;
wherein, taking each area grade corresponding to each representative image block with a set number as the input data of the input layer of the convolutional neural network comprises: and when the number of the deduplication characters in the deduplication character set is larger than the set number, taking the area grades corresponding to the set number of the representative image blocks closest to the center position of the current live broadcast picture as input data of an input layer of a convolutional neural network.
The self-adaptive matching system for the character size set by the user is effective in identification and timely in control. The broadcasting distance of the anchor can be adaptively adjusted based on the display character size set by each live user in the current live broadcast picture, so that the sizes of the characters of the live broadcast picture and the anchor image are matched.
Detailed Description
An embodiment of the user-set character size adaptive matching system of the present invention will be described in detail below.
Convolutional Neural Networks (CNN) are a class of feed forward Neural Networks (fed forward Neural Networks) that contain convolution computations and have a deep structure, and are one of the representative algorithms for deep learning (deep learning). Convolutional Neural Networks have a feature learning (rendering) capability, and can perform Shift-Invariant classification (Shift-Invariant classification) on input information according to a hierarchical structure thereof, and are therefore also called "Shift-Invariant Artificial Neural Networks (SIANN)".
The study of convolutional neural networks began in the 80 to 90 s of the twentieth century, with time delay networks and LeNet-5 being the earliest convolutional neural networks that emerged; after the twenty-first century, with the introduction of deep learning theory and the improvement of numerical computing equipment, convolutional neural networks have been rapidly developed and applied to the fields of computer vision, natural language processing, and the like.
The convolutional neural network is constructed by imitating a visual perception (visual perception) mechanism of a living being, can perform supervised learning and unsupervised learning, and has the advantages that the convolutional neural network can learn grid-like topologic features such as pixels and audio with small calculation amount, has stable effect and has no additional feature engineering (feature engineering) requirement on data due to the fact that convolutional kernel parameter sharing in an implicit layer and sparsity of connection between layers.
In the prior art, due to the rapid development of live broadcast application, the design of software and hardware of a live broadcast platform tends to be perfect. However, since the pursuit of the live broadcast picture is not only the clarity of the picture but also the beauty and harmony of the picture, for example, the size of the characters of the subtitles set by each viewer in the live broadcast picture is different in the prior art, so that it is difficult to match the size of the anchor character with the size of the characters displayed on the picture at each broadcast time.
In order to overcome the defects, the invention builds a self-adaptive matching system for the character size set by the user, and can effectively solve the corresponding technical problem.
The adaptive matching system for the user-set character size shown according to the embodiment of the invention comprises:
the distance measuring mechanism is arranged at a main broadcasting client end of the live broadcasting platform, is positioned on a shell of a camera for shooting main broadcasting pictures, and is used for measuring the real-time distance from a main coding part to the camera;
the error notification equipment is realized by adopting a display mechanism or a voice playing mechanism, is respectively connected with the network operation mechanism and the distance measurement mechanism, and is used for carrying out corresponding error notification operation when the difference value between the received real-time distance and the optimal distance from the main broadcasting to the camera at the current moment exceeds the limit;
the content acquisition equipment is arranged at a main broadcast client end of a live broadcast platform, is connected with the main broadcast client end and is used for acquiring a current live broadcast picture of the main broadcast client end;
the first analysis mechanism is connected with the content acquisition equipment and used for identifying each character in the current live broadcast picture based on OCR (optical character recognition), so as to obtain an image block of each character in the current live broadcast picture, wherein the image block is a rectangular image area only comprising corresponding characters;
the second analysis mechanism is connected with the first analysis mechanism and is used for executing the following processing on the image blocks of each character in the current live broadcast picture: analyzing an area grade of the image patch based on a length of the image patch and a height of the image patch;
the frequency identification equipment is respectively connected with the first analysis mechanism and the second analysis mechanism and is used for carrying out duplication elimination processing on each character appearing in the current live broadcast picture so as to obtain a duplication elimination character set;
a block selection device, connected to the frequency identification device, for performing the following processing for each of the deduplication characters in the deduplication character set: the duplication removing characters correspond to more than one image block in the current live broadcast picture, and the image block closest to the center position of the current live broadcast picture is used as a representative image block of the duplication removing characters;
the data integration equipment is respectively connected with the second analysis mechanism and the block selection equipment and is used for taking each area grade corresponding to each representative image block with a set number as input data of an input layer of the convolutional neural network;
the network operation mechanism is used for operating the convolutional neural network to obtain the optimal distance from the anchor to the camera at the current moment, and the optimal distance from the anchor to the camera is the output data of the convolutional neural network output layer;
the network training equipment is connected with the network operation mechanism and used for training the convolutional neural network by taking each area grade as input data of an input layer of the convolutional neural network and simultaneously taking the optimal distance from the main player to the camera as output data of an output layer of the convolutional neural network;
wherein, taking each area grade corresponding to each representative image block with a set number as the input data of the input layer of the convolutional neural network comprises: and when the number of the deduplication characters in the deduplication character set is larger than the set number, taking the area grades corresponding to the set number of the representative image blocks closest to the center position of the current live broadcast picture as input data of an input layer of a convolutional neural network.
Next, the specific configuration of the user-set character size adaptive matching system of the present invention will be further described.
In the user-set character size adaptive matching system:
and the error notification equipment is also used for interrupting the corresponding error notification operation when the difference value between the received real-time distance and the optimal distance from the anchor to the camera at the current moment is not over the limit.
In the user-set character size adaptive matching system:
the method for using the area grades respectively corresponding to the set number of the representative image blocks as the input data of the input layer of the convolutional neural network further comprises the following steps: when the number of the deduplication characters in the deduplication character set is smaller than the set number, all area grades respectively corresponding to each representative image block of each deduplication character in the deduplication character set are used as partial input data of an input layer of the convolutional neural network, the number of the deduplication characters in the deduplication character set is subtracted from the set number to obtain a difference value, and the area grade with the value of the difference value being zero is used as residual input data of the input layer of the convolutional neural network.
In the user-set character size adaptive matching system:
the method for using the area grades respectively corresponding to the set number of the representative image blocks as the input data of the input layer of the convolutional neural network further comprises the following steps: and when the number of the deduplication characters in the deduplication character set is equal to the set number, taking each area grade corresponding to each representative image block of each deduplication character in the deduplication character set as input data of an input layer of the convolutional neural network.
In the user-set character size adaptive matching system:
analyzing the area level of the image patch based on the length of the image patch and the height of the image patch comprises: and the length of the image block and the height of the image block are supported to obtain the area of the image block, and then an area grade which is in direct proportion to the area is obtained based on the area.
In the user-set character size adaptive matching system:
performing deduplication processing on each character appearing in the current live view to obtain a deduplication character set includes: each deduplication character in the set of deduplication characters is different.
In the user-set character size adaptive matching system, the method further comprises:
and the data storage chip is respectively connected with the block selection equipment, the data integration equipment, the network training equipment and the network operation mechanism.
In the user-set character size adaptive matching system:
the data storage chip is used for respectively storing temporary storage data of the block selection device, the data integration device, the network training device and the network operation mechanism.
In the user-set character size adaptive matching system, the method further comprises:
and the load detection mechanism is respectively connected with the network training equipment and the network operation mechanism and is used for respectively detecting the current operation loads of the network training equipment and the network operation mechanism.
In addition, in the user-set character size adaptive matching system, the method further includes: and the ZIGBEE communication mechanism and the load detection mechanism unit are used for sending the current operation loads of the network training equipment and the network operation mechanism. ZIGBEE is a low power consumption local area network protocol based on the IEEE802.15.4 standard. According to international standards, ZIGBEE technology is a short-range, low-power wireless communication technology. This name (also called the purple bee protocol) is derived from the dance of the eight characters of bees, since bees (bee) communicate the orientation information of pollen with partners by flying and "waving" (ZIG) flapping wings, "i.e. bees form a communication network in the community by this way. Its advantages are short distance, low complexity, self-organization, low power consumption and low data rate. The device is mainly suitable for the fields of automatic control and remote control, and can be embedded into various devices. In short, ZIGBEE is an inexpensive and low-power-consumption short-range wireless networking communication technology. ZIGBEE is a wireless network protocol for low-speed short-range transmission. The ZIGBEE protocol is, from bottom to top, a physical layer (PHY), a media access control layer (MAC), a Transport Layer (TL), a network layer (NWK), an application layer (APL), and the like. Wherein the physical layer and the medium access control layer comply with the provisions of the IEEE802.15.4 standard.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (9)
1. A user-set character size adaptive matching system, the system comprising:
the distance measuring mechanism is arranged at a main broadcasting client end of the live broadcasting platform, is positioned on a shell of a camera for shooting main broadcasting pictures, and is used for measuring the real-time distance from a main broadcasting face to the camera;
the error notification equipment is realized by adopting a display mechanism or a voice playing mechanism, is respectively connected with the network operation mechanism and the distance measurement mechanism, and is used for carrying out corresponding error notification operation when the difference value between the received real-time distance and the optimal distance from the main broadcasting to the camera at the current moment exceeds the limit;
the content acquisition equipment is arranged at a main broadcast client end of a live broadcast platform, is connected with the main broadcast client end and is used for acquiring a current live broadcast picture of the main broadcast client end;
the first analysis mechanism is connected with the content acquisition equipment and used for identifying each character in the current live broadcast picture based on OCR (optical character recognition), so as to obtain an image block of each character in the current live broadcast picture, wherein the image block is a rectangular image area only comprising corresponding characters;
the second analysis mechanism is connected with the first analysis mechanism and is used for executing the following processing on the image blocks of each character in the current live broadcast picture: analyzing an area grade of the image patch based on a length of the image patch and a height of the image patch;
the frequency identification equipment is respectively connected with the first analysis mechanism and the second analysis mechanism and is used for carrying out duplication elimination processing on each character appearing in the current live broadcast picture so as to obtain a duplication elimination character set;
a block selection device, connected to the frequency identification device, for performing the following processing for each of the deduplication characters in the deduplication character set: the duplication removing characters correspond to more than one image block in the current live broadcast picture, and the image block closest to the center position of the current live broadcast picture is used as a representative image block of the duplication removing characters;
the data integration equipment is respectively connected with the second analysis mechanism and the block selection equipment and is used for taking each area grade corresponding to each representative image block with a set number as input data of an input layer of the convolutional neural network;
the network operation mechanism is used for operating the convolutional neural network to obtain the optimal distance from the anchor to the camera at the current moment, and the optimal distance from the anchor to the camera is the output data of the convolutional neural network output layer;
the network training equipment is connected with the network operation mechanism and used for training the convolutional neural network by taking each area grade as input data of an input layer of the convolutional neural network and simultaneously taking the optimal distance from the main player to the camera as output data of an output layer of the convolutional neural network;
wherein, taking each area grade corresponding to each representative image block with a set number as the input data of the input layer of the convolutional neural network comprises: and when the number of the deduplication characters in the deduplication character set is larger than the set number, taking the area grades corresponding to the set number of the representative image blocks closest to the center position of the current live broadcast picture as input data of an input layer of a convolutional neural network.
2. The adaptive matching system for user-set character sizes of claim 1, wherein:
and the error notification equipment is also used for interrupting the corresponding error notification operation when the difference value between the received real-time distance and the optimal distance from the anchor to the camera at the current moment is not over the limit.
3. The adaptive matching system for user-set character sizes of claim 2, wherein:
the method for using the area grades respectively corresponding to the set number of the representative image blocks as the input data of the input layer of the convolutional neural network further comprises the following steps: when the number of the deduplication characters in the deduplication character set is smaller than the set number, all area grades respectively corresponding to each representative image block of each deduplication character in the deduplication character set are used as partial input data of an input layer of the convolutional neural network, the number of the deduplication characters in the deduplication character set is subtracted from the set number to obtain a difference value, and the area grade with the value of the difference value being zero is used as residual input data of the input layer of the convolutional neural network.
4. The adaptive matching system for user-set character sizes of claim 3, wherein:
the method for using the area grades respectively corresponding to the set number of the representative image blocks as the input data of the input layer of the convolutional neural network further comprises the following steps: and when the number of the deduplication characters in the deduplication character set is equal to the set number, taking each area grade corresponding to each representative image block of each deduplication character in the deduplication character set as input data of an input layer of the convolutional neural network.
5. The adaptive matching system for user-set character sizes of claim 4, wherein:
analyzing the area level of the image patch based on the length of the image patch and the height of the image patch comprises: multiplying the length of the image block and the height of the image block to obtain the area of the image block, and further obtaining an area grade proportional to the area based on the area.
6. The adaptive matching system for user-set character sizes of claim 5, wherein:
performing deduplication processing on each character appearing in the current live view to obtain a deduplication character set includes: each deduplication character in the set of deduplication characters is different.
7. The adaptive matching system for user-set character sizes of claim 6, wherein the system further comprises:
and the data storage chip is respectively connected with the block selection equipment, the data integration equipment, the network training equipment and the network operation mechanism.
8. The adaptive matching system for user-set character sizes of claim 7, wherein:
the data storage chip is used for respectively storing temporary storage data of the block selection device, the data integration device, the network training device and the network operation mechanism.
9. The adaptive matching system for user-set character sizes of claim 8, wherein the system further comprises:
and the load detection mechanism is respectively connected with the network training equipment and the network operation mechanism and is used for respectively detecting the current operation loads of the network training equipment and the network operation mechanism.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011409220.1A CN113064679B (en) | 2020-12-06 | 2020-12-06 | User-set character size adaptive matching system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011409220.1A CN113064679B (en) | 2020-12-06 | 2020-12-06 | User-set character size adaptive matching system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113064679A CN113064679A (en) | 2021-07-02 |
| CN113064679B true CN113064679B (en) | 2021-12-31 |
Family
ID=76558691
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011409220.1A Expired - Fee Related CN113064679B (en) | 2020-12-06 | 2020-12-06 | User-set character size adaptive matching system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113064679B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110598011A (en) * | 2019-09-27 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Data processing method, data processing device, computer equipment and readable storage medium |
| CN110826355A (en) * | 2018-08-07 | 2020-02-21 | 腾讯数码（天津）有限公司 | Image recognition method, device and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052937B (en) * | 2017-12-28 | 2019-05-31 | 百度在线网络技术（北京）有限公司 | Based on Weakly supervised character machining device training method, device, system and medium |
-
2020
- 2020-12-06 CN CN202011409220.1A patent/CN113064679B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826355A (en) * | 2018-08-07 | 2020-02-21 | 腾讯数码（天津）有限公司 | Image recognition method, device and storage medium |
| CN110598011A (en) * | 2019-09-27 | 2019-12-20 | 腾讯科技（深圳）有限公司 | Data processing method, data processing device, computer equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113064679A (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108898579B (en) | Image definition recognition method and device and storage medium | |
| US10482329B2 (en) | Systems and methods for identifying activities and/or events in media contents based on object data and scene data | |
| CN110532970B (en) | Age and gender attribute analysis method, system, equipment and medium for 2D images of human faces | |
| CN108399362A (en) | A kind of rapid pedestrian detection method and device | |
| US20190244362A1 (en) | Differentiable Jaccard Loss Approximation for Training an Artificial Neural Network | |
| CN110490252B (en) | Indoor people number detection method and system based on deep learning | |
| CN108154102A (en) | A kind of traffic sign recognition method | |
| CN107909008A (en) | Video target tracking method based on multichannel convolutive neutral net and particle filter | |
| CN109559362B (en) | Image subject face replacing method and device | |
| WO2021063056A1 (en) | Facial attribute recognition method and apparatus, and electronic device and storage medium | |
| CN110263654A (en) | A kind of flame detecting method, device and embedded device | |
| CN113052150B (en) | Living body detection method, living body detection device, electronic apparatus, and computer-readable storage medium | |
| CN111753805A (en) | Method and device for detecting wearing of safety helmet | |
| CN105718932A (en) | Colorful image classification method based on fruit fly optimization algorithm and smooth twinborn support vector machine and system thereof | |
| US20200327709A1 (en) | System and method for rendering an image | |
| CN113449806A (en) | Two-stage forestry pest identification and detection system and method based on hierarchical structure | |
| WO2022183805A1 (en) | Video classification method, apparatus, and device | |
| CN110826380A (en) | Abnormal signature identification method and system | |
| Tiwari | A deep learning approach to recognizing bees in video analysis of bee traffic | |
| Yuan et al. | Sensitivity examination of YOLOv4 regarding test image distortion and training dataset attribute for apple flower bud classification | |
| CN113064679B (en) | User-set character size adaptive matching system | |
| CN111612090B (en) | Image emotion classification method based on content color cross correlation | |
| CN114022811A (en) | Water surface floater monitoring method and system based on continuous learning | |
| Li et al. | Saliency consistency-based image re-colorization for color blindness | |
| CN112507770B (en) | Rice disease and insect pest identification method and system |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Qu Jianbo Inventor after: Zhao Jian Inventor after: Yang Yang Inventor before: Yang Yang |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211213 Address after: 273500 Department of Neurosurgery, Zoucheng people's Hospital, No. 5677, Chongyi Road, Zoucheng City, Jining City, Shandong Province Applicant after: Qu Jianbo Applicant after: Zhao Jian Address before: 225300 601a, building 1, No. 98, Yingchun East Road, hailing Industrial Park, Taizhou City, Jiangsu Province Applicant before: Taizhou langjiaxin Network Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20211231 |