CN112699780A - Object identification method, device, equipment and storage medium - Google Patents
Object identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112699780A CN112699780A CN202011594494.2A CN202011594494A CN112699780A CN 112699780 A CN112699780 A CN 112699780A CN 202011594494 A CN202011594494 A CN 202011594494A CN 112699780 A CN112699780 A CN 112699780A
- Authority
- CN
- China
- Prior art keywords
- handwriting
- sequence
- writing
- sample
- pen
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/333—Preprocessing; Feature extraction
- G06V30/347—Sampling; Contour coding; Stroke extraction
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- User Interface Of Digital Computer (AREA)
- Character Discrimination (AREA)
Abstract
The application discloses an object identification method, device, equipment and storage medium, and belongs to the technical field of input methods. The method comprises the following steps: acquiring a handwriting sequence corresponding to handwriting written on an input method interface by a user; inputting the handwriting sequence into a pre-trained handwriting recognition model, and recognizing the probability that each handwriting is a starting pen; identifying at least one object corresponding to the handwriting sequence according to the probability that each handwriting is a starting pen; the handwriting recognition model is obtained by training based on the label information of each writing in the handwriting sample set and the handwriting sample set, and the label information of each writing is the information of whether the writing is the initial pen or not. According to the embodiment of the application, the problem of low mode accuracy of the initial pen of the handwriting input mode identification object can be solved.
Description
Technical Field
The application belongs to the technical field of input methods, and particularly relates to an object identification method, device, equipment and storage medium.
Background
With the widespread use of electronic devices, the electronic devices have become indispensable tools for users to work, study, and the like. In the process of working, studying and the like, a user usually needs to install an input method application program, a plug-in and the like on the electronic equipment to complete typing. At present, the input methods include a handwriting input method, a five-stroke input method, and the like.
In the related art, the handwriting input method detects that a user writes on a handwriting keyboard of an electronic device, and the user can continuously write a plurality of objects, wherein the plurality of objects are overlapped. It is often necessary to determine the demarcation point between each object by the degree of overlap between the different objects, i.e. to identify the starting pen for each object. However, when the user has a low degree of overlap between objects during handwriting or no overlap occurs between objects, the initial pen of each object may not be accurately recognized.
Therefore, in the handwriting input method, the accuracy of the method of recognizing the start pen of the object in the related art is low.
Disclosure of Invention
An object of the embodiments of the present application is to provide an object recognition method, apparatus, device and storage medium, which can solve the problem of low accuracy of a manner of recognizing a start pen of an object in a handwriting input manner in the related art.
In order to solve the technical problem, the present application is implemented as follows:
in a first aspect, an embodiment of the present application provides an object identification method, where the method includes:
acquiring a handwriting sequence corresponding to handwriting written on an input method interface by a user;
inputting the handwriting sequence into a pre-trained handwriting recognition model, and recognizing the probability that each handwriting is a starting pen;
identifying at least one object corresponding to the handwriting sequence according to the probability that each handwriting is a starting pen;
the handwriting recognition model is obtained by training based on the label information of each writing in the handwriting sample set and the handwriting sample set, and the label information of each writing is the information of whether the writing is the initial pen or not.
In a second aspect, an embodiment of the present application provides an object recognition apparatus, including:
the acquisition module is used for acquiring a handwriting sequence corresponding to handwriting written on the input method interface by a user;
the first recognition module is used for inputting the handwriting sequence into a pre-trained handwriting recognition model and recognizing the probability that each handwriting is a starting pen;
the second recognition module is also used for recognizing at least one object corresponding to the handwriting sequence according to the probability that each handwriting is the initial pen;
the handwriting recognition model is obtained by training based on the label information of each writing in the handwriting sample set and the handwriting sample set, and the label information of each writing is the information of whether the writing is the initial pen or not.
In a third aspect, embodiments of the present application provide an electronic device, which includes a processor, a memory, and a program or instructions stored on the memory and executable on the processor, where the program or instructions, when executed by the processor, implement the steps of the method according to the first aspect.
In a fourth aspect, embodiments of the present application provide a readable storage medium on which a program or instructions are stored, which when executed by a processor, implement the steps of the method according to the first aspect.
In the embodiment of the application, the handwriting recognition model recognizes the handwriting sequence corresponding to the handwriting written by the user on the input method interface, and the handwriting recognition model is obtained by training the label information of each handwriting in the handwriting sample set and the handwriting sample set, so that the handwriting recognition model can accurately recognize the probability of the initial pen of each handwriting, and recognizes at least one object corresponding to the handwriting sequence based on the probability of each handwriting being the initial pen.
Drawings
Fig. 1 is a schematic flowchart of an object identification method according to an embodiment of the present application;
FIG. 2 is one of schematic diagrams of an interface of an electronic device provided by an embodiment of the present application;
fig. 3 is a second schematic diagram of an interface of an electronic device according to an embodiment of the present disclosure;
FIG. 4 is a flowchart illustrating a method for training a handwriting recognition model according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an object recognition apparatus according to an embodiment of the present application;
fig. 6 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present application.
Detailed Description
Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended to be illustrative only and are not intended to be limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present application by illustrating examples thereof.
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 … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
With the widespread use of electronic devices, the electronic devices have become indispensable tools for users to work, study, and the like. In the process of working, studying and the like, a user usually needs to install an input method application program, a plug-in and the like on the electronic equipment to complete typing. At present, the input methods include a handwriting input method, a five-stroke input method, and the like.
In the related art, the handwriting input method detects that a user writes on a handwriting keyboard of an electronic device, and the user can continuously write a plurality of objects, wherein the plurality of objects are overlapped. It is often necessary to determine the demarcation point between each object by the degree of overlap between the different objects, i.e. to identify the starting pen for each object. However, when the user has a low degree of overlap between objects during handwriting or no overlap occurs between objects, the initial pen of each object may not be accurately recognized.
Therefore, in the handwriting input method, the accuracy of the method of recognizing the start pen of the object in the related art is low.
In order to solve the prior art problems, embodiments of the present application provide an object identification method, an apparatus, a device, and a computer storage medium. The execution main body of the object identification method provided by the embodiment of the application can be an electronic device, such as a mobile phone, a tablet computer and the like, and can also be a server. The object identification method provided by the embodiment of the present application is described in detail below with an execution subject as an electronic device.
Fig. 1 shows a flowchart of an object identification method 100 provided in an embodiment of the present application.
As shown in fig. 1, an object identification method 100 provided in an embodiment of the present application may include the following steps: S101-S103.
S101: and acquiring a handwriting sequence corresponding to the handwriting written on the input method interface by the user.
S102: inputting the handwriting sequence into a pre-trained handwriting recognition model, and recognizing the probability that each writing handwriting is an initial pen.
S103: and identifying at least one object corresponding to the handwriting sequence according to the probability that each handwriting is the initial pen.
The above-mentioned S101 to S103 will be described in detail.
First, S101 is described.
In the case where the interface of the electronic device displays an input method interface, the user may input an object through the input method interface, such as: text, characters, and the like. When the input method is a handwriting method, as shown in fig. 2, the user can continuously write a plurality of objects on the handwriting keyboard. When a user writes a plurality of objects continuously, the electronic equipment can acquire the handwriting written by the user on the input method interface, so that a handwriting sequence corresponding to the handwriting can be generated. For example, the objects that the user writes consecutively are: what, then the handwriting sequence is { ノㄧーㄧノㄥヽ}.
Next, S102 is introduced.
The handwriting recognition model is pre-trained. Wherein the handwriting recognition model is used for recognizing the probability that each handwriting is the initial pen in the handwriting sequence.
Inputting the sequence of scripts into the handwriting recognition model, the probability that each written script is the initial pen can be determined. For example, the sequence of handwriting { ノㄧーㄧノㄥヽ} is input into a handwriting recognition model, and the probability that each of the handwriting in the sequence of handwriting { ノㄧーㄧノㄥヽ} is the starting pen is determined.
Here, the handwriting recognition model is obtained by training based on the label information of each writing in the handwriting sample set and the handwriting sample set, and the label information of each writing is information of whether the writing is the initial pen.
Specifically, in some embodiments of the present application, the handwriting recognition model includes a first neural network and a second neural network. Step S102 may identify the probability that each writing trace is the initial pen by: firstly, inputting a handwriting sequence into a first neural network, and extracting a first feature vector of each handwriting. The first feature vector is then input into a second neural network, identifying the probability that each writing script is a start pen.
The first feature vector of each writing is used for representing the features of the writing and the handwriting features corresponding to the writing before the writing. Inputting the handwriting sequence into a first neural network, wherein the first neural network can determine a first feature vector corresponding to each handwriting, and the first neural network can be a trained convolutional neural network. And inputting the first feature vector corresponding to each writing script into the second neural network, so as to determine the probability that each writing script is the initial pen. Wherein the second neural network may be a trained long-short term memory network.
For example, the handwriting sequence is { ノ pieces ー pieces ノ ㄥ ヽ }, wherein the first feature vector of the writing script "ー" can not only characterize the handwriting characteristics of the writing script "ー" itself, but also characterize the handwriting characteristics of the writing scripts "ノ" and "piece".
Inputting the first feature vector corresponding to each writing stroke into the second neural network, and determining that the probability that the writing stroke "ノ" is the initial pen is 99.99%, the probability that the writing stroke "I" is the initial pen is 6.25%, the probability that the writing stroke "ー" is the initial pen is 46.3%, the probability that the writing stroke "I" is the initial pen is 5.32%, the probability that the writing stroke "ノ" is the initial pen is 96.98%, the probability that the writing stroke "ㄥ" is the initial pen is 8.36%, and the probability that the writing stroke "ヽ" is the initial pen is 8.99%.
Finally, S103 is introduced.
Based on the probability that each writing script is the initial pen, at least one object corresponding to the sequence of notes may be identified. The handwriting sequence can correspond to at least one object, e.g., what the two words of the handwriting sequence ノ, ー and ノ ㄥ ヽ correspond to.
Specifically, in some embodiments of the present application, the target writing script may be determined to be the initial pen when the probability that the initial pen of the target writing script is greater than the preset threshold in the writing scripts may be determined by comparing the probability that each writing script is the initial pen with the preset threshold. And dividing the handwriting sequence by the target handwriting to obtain a plurality of sub-handwriting sequences.
For example, the handwriting sequence is { ノ pieces ー pieces ノ ㄥ ヽ }, wherein the first feature vector of the writing script "ー" can not only characterize the handwriting characteristics of the writing script "ー" itself, but also characterize the handwriting characteristics of the writing scripts "ノ" and "piece". Inputting the first feature vector corresponding to each writing script into the second neural network, determining that the probability that the writing script "ノ" is the starting pen is 99.99%, the probability that the writing script "i" is the starting pen is 6.25%, the probability that the writing script "ー" is the starting pen is 46.3%, the probability that the writing script "i" is the starting pen is 5.32%, the probability that the writing script "ノ" is the starting pen is 96.98%, the probability that the writing script "ㄥ" is the starting pen is 8.36%, the probability that the writing script "ヽ" is the starting pen is 8.99%, and the preset threshold value is 60%, determining that the writing scripts ノ and ノ are both starting pens, and thus dividing the sequence of { ノ i | ー i | ノ ㄥ ヽ } into { ノ i ー } and { ノ ㄥ ヽ }.
And combining the writing scripts in each sub-handwriting sequence to generate a character corresponding to each sub-handwriting sequence, and obtaining at least one object corresponding to the handwriting sequence. For example, "what" may be "obtained by combining { ノ I ー I } and" what "may be" obtained by combining { ノ ㄥ ヽ } so as to obtain an object corresponding to the sequence of handwriting { ノㄧーㄧノㄥヽ}.
In some embodiments of the application, in order to improve the accuracy of object recognition, a first writing script corresponding to an nth writing script before a target writing script, a second writing script corresponding to an mth writing script after the target writing script, and the target writing script may be used as starting pens, and the handwriting sequences are respectively divided to obtain a plurality of sub-handwriting sequences, where M and N are positive integers.
For example, if the target writing script is "I", it may be determined that a first stroke "I" of the target writing script "I" is a first writing script, and a second stroke "ㄥ" after the target writing script "I" is a second writing script, so that the sequence of scripts { ノㄧーㄧノㄥヽ} may be divided into a sequence of sub-scripts: { ノ I ー } and { ノ ㄥ ヽ }, { ノ I } and { ー I ノ ㄥ ヽ }, { ノ I ー I ノ } and { ㄥ ヽ }.
The following steps may be specifically included in S103: and combining the writing scripts in each sub-handwriting sequence, so as to obtain at least one object corresponding to each sub-handwriting sequence.
For example, objects corresponding to the sub-script sequences { ノ i ー } and { ノ ㄥ ヽ } are "xi", "ma", objects corresponding to the sub-script sequences { ノ } and { ー i ノ ㄥ ヽ } are "xi", "ma", respectively, objects corresponding to the sub-script sequences { ノ i ー i ノ } and { ㄥ ヽ } are "xi", "yizhu", respectively.
In some embodiments of the present application, in order to improve the accuracy of object recognition, at least one object corresponding to each sub-handwriting sequence may be further combined into word information. And then respectively inputting the word information corresponding to each sub-handwriting sequence into a preset word handwriting recognition model, and determining target word information. For example, the word information corresponding to the sub-handwriting sequence is: when the words are input to the preset model, the word information can be determined as "what". And then displaying the target word information in an information input box of the electronic equipment interface. For example, as shown in fig. 3, "what" is displayed in the information input box 30.
In some embodiments of the present application, before S101, a preset neural network model needs to be trained, so as to obtain a handwriting recognition model. As shown in fig. 4, the handwriting recognition model training method provided by the embodiment of the present application includes the following steps S401 to S404.
S401: a handwriting sample set is obtained, and the handwriting sample set comprises a plurality of handwriting training samples.
Each handwriting training sample comprises a handwriting sequence sample and label information corresponding to each handwriting sample in the handwriting sequence sample, and the label information of each handwriting sample is information of whether the handwriting sample is a starting pen.
For example, if the sample of the handwriting sequence is { ノㄧーㄧノㄥヽ}, the tag information of the handwriting sample "ノ" is "original pen", and the tag information of "I" is "non-original pen".
For each handwriting sequence sample, respectively executing the following steps:
s402: inputting the handwriting sequence samples into a preset neural network model, and determining the classification result of each handwriting sample.
Specifically, the preset neural network model comprises a convolutional neural network and a long-term and short-term memory network.
And inputting the handwriting sequence samples into a convolutional neural network, and extracting a second feature vector of each handwriting sample in the handwriting sequence samples. And inputting the second feature vector into the long-term and short-term memory network, and determining the classification result of each handwriting sample.
Here, the second feature vector is used to represent the handwriting sample and the features of the handwriting sample preceding the handwriting sample. For example, the second feature vector of the writing sample "ー" can not only characterize the handwriting features of the writing "ー" itself, but also characterize the handwriting features of the writing "ノ" and "i".
The classification result may be "1" or "0". Wherein, the classification result "1" indicates that the handwriting sample is the initial pen, and "0" indicates that the handwriting sample is not the initial pen.
S403: and determining a loss function value according to the classification result of each handwriting sample and the label information corresponding to each handwriting sample.
S404: and training a preset neural network model according to the loss function value to obtain a handwriting recognition model.
In the process of training the preset neural network model, parameters of the convolutional neural network and the long-term and short-term memory network can be adjusted according to the loss function value.
Therefore, the preset neural network model is trained through the handwriting sample set and the label information, the handwriting recognition model can be obtained, whether the handwriting is the initial pen or not is recognized through the handwriting recognition model, and the initial pen recognition efficiency is improved.
According to the object recognition method provided by the embodiment of the application, the handwriting sequence corresponding to the handwriting written on the input method interface by the user is recognized through the handwriting recognition model, and the handwriting recognition model is obtained through training of label information of each handwriting in the handwriting sample set and the handwriting sample set, so that the handwriting recognition model can accurately recognize the probability of the initial pen of each handwriting, and accordingly at least one object corresponding to the handwriting sequence is recognized based on the probability that each handwriting is the initial pen.
It should be noted that, in the embodiment of the present application, an electronic device executes an object identification method as an example, and the object identification method provided in the embodiment of the present application is described. However, in the object recognition method provided in the embodiment of the present application, the execution subject may be an electronic device, or may also be an object recognition apparatus or a control module in the object recognition apparatus for executing the object recognition method.
Based on the object identification method provided by the application, correspondingly, the application provides an object identification device of an embodiment. Next, in the embodiments of the present application, an object recognition method executed by an object recognition apparatus is taken as an example, and the object recognition apparatus provided in the embodiments of the present application is described.
Fig. 5 is a schematic structural diagram of an object recognition apparatus 500 according to the present application.
As shown in fig. 5, the object recognition apparatus 500 provided by the present application may include: the system comprises an acquisition module 501, a first identification module 502 and a second identification module 503.
An obtaining module 501, configured to obtain a handwriting sequence corresponding to a handwriting written on an input method interface by a user;
a first recognition module 502, configured to input the handwriting sequence into a pre-trained handwriting recognition model, and recognize a probability that each handwriting is a start pen;
the second recognition module 503 is further configured to recognize at least one object corresponding to the handwriting sequence according to the probability that each handwriting is the initial pen;
the handwriting recognition model is obtained by training based on the label information of each writing in the handwriting sample set and the handwriting sample set, and the label information of each writing is the information of whether the writing is the initial pen or not.
In some embodiments of the present application, the handwriting recognition model includes a first neural network and a second neural network, and the first recognition module 502 specifically includes:
the extraction unit is used for inputting the handwriting sequence into a first neural network and extracting a first feature vector of each handwriting; the first feature vector of each handwriting is used for representing the features of the handwriting and the handwriting features corresponding to the handwriting before the handwriting;
and the recognition unit is used for inputting the first feature vector into the second neural network and recognizing the probability that each writing script is the initial pen.
In some embodiments of the present application, the obtaining module 501 is configured to obtain a handwriting sample set, where the handwriting sample set includes a plurality of handwriting training samples, each of the handwriting training samples includes a handwriting sequence sample and label information corresponding to each of the handwriting samples in the handwriting sequence sample, and the label information of each of the handwriting samples is information of whether the handwriting sample is a start pen;
the apparatus 500 further comprises:
the first determining module is used for inputting the handwriting sequence samples into a preset neural network model and determining the classification result of each handwriting sample;
the second determining module is used for determining a loss function value according to the classification result of each handwriting sample and the label information corresponding to each handwriting sample;
and the training module is used for training a preset neural network model according to the loss function value to obtain a handwriting recognition model.
In some embodiments of the present application, the predetermined neural network model includes a convolutional neural network and a long-short term memory network;
the first determining module specifically includes:
the extracting unit is used for inputting the handwriting sequence samples into the convolutional neural network, extracting a second feature vector of each handwriting sample in the handwriting sequence samples, wherein the second feature vector is used for representing the handwriting sample and the features of the handwriting sample before the handwriting sample;
and the determining unit is used for inputting the second feature vector into the long-term and short-term memory network and determining the classification result of each handwriting sample.
In some embodiments of the present application, the generating module specifically includes:
the determining unit is used for determining the target writing handwriting as the initial pen under the condition that the probability that the target writing handwriting is the initial pen is greater than a preset threshold value; wherein the writing handwriting comprises a target writing handwriting;
the dividing unit is used for dividing the handwriting sequence by the target handwriting to obtain a plurality of sub-handwriting sequences;
and the combining unit is used for combining the writing scripts in each sub-handwriting sequence to obtain at least one object corresponding to the handwriting sequence.
In some embodiments of the application, the dividing unit is specifically configured to divide the handwriting sequence to obtain a plurality of sub-handwriting sequences, with a first writing handwriting corresponding to an nth stroke before the target writing, a second writing handwriting corresponding to an mth stroke after the target writing, and the target writing as starting pens, respectively, where M and N are positive integers;
and the combining unit is specifically used for combining each writing script in each sub-handwriting sequence to obtain at least one object corresponding to each sub-handwriting sequence.
In some embodiments of the present application, the apparatus 500 further comprises:
the combination module is used for combining at least one object corresponding to each sub-handwriting sequence into word information;
the determining module is used for respectively inputting the word information corresponding to each sub-handwriting sequence into a preset word handwriting recognition model and determining target word information;
and the sending module is used for displaying the target word information in an information input box of the electronic equipment interface.
Each module/unit in the apparatus shown in fig. 5 has a function of implementing each step in fig. 1, and can achieve the corresponding technical effect, and for brevity, the description is not repeated here.
The object recognition device provided by the embodiment of the application recognizes the handwriting sequence corresponding to the handwriting written by the user on the input method interface through the handwriting recognition model, and the handwriting recognition model is obtained by training the label information of each handwriting in the handwriting sample set, so that the handwriting recognition model can accurately recognize the probability of the initial pen of each handwriting, and recognizes at least one object corresponding to the handwriting sequence based on the probability of the initial pen of each handwriting.
Fig. 6 shows a hardware structure diagram of an electronic device provided in an embodiment of the present application.
As shown in fig. 6, the electronic device may include a processor 601 and a memory 602 storing computer program instructions.
Specifically, the processor 601 may include a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement the embodiments of the present Application.
In one example, the Memory 602 may be a Read Only Memory (ROM). In one example, the ROM may be mask programmed ROM, programmable ROM (prom), erasable prom (eprom), electrically erasable prom (eeprom), electrically rewritable ROM (earom), or flash memory, or a combination of two or more of these.
The memory 602 may include Read Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors), it is operable to perform operations described with reference to the methods according to an aspect of the present disclosure.
The processor 601 reads and executes the computer program instructions stored in the memory 602 to implement the method/steps in the embodiment shown in fig. 2, and achieve the corresponding technical effect achieved by the embodiment shown in fig. 2 executing the method/steps, which is not described herein again for brevity.
In one example, the electronic device may also include a communication interface 603 and a bus 610. As shown in fig. 6, the processor 601, the memory 602, and the communication interface 603 are connected via a bus 610 to complete communication therebetween.
The communication interface 603 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiments of the present application.
The bus 610 includes hardware, software, or both to couple the components of the object recognition device to each other. By way of example, and not limitation, a Bus may include an Accelerated Graphics Port (AGP) or other Graphics Bus, an Enhanced Industry Standard Architecture (EISA) Bus, a Front-Side Bus (Front Side Bus, FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) Bus, an infiniband interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a Micro Channel Architecture (MCA) Bus, a Peripheral Component Interconnect (PCI) Bus, a PCI-Express (PCI-X) Bus, a Serial Advanced Technology Attachment (SATA) Bus, a video electronics standards association local (VLB) Bus, or other suitable Bus or a combination of two or more of these. Bus 610 may include one or more buses, where appropriate. Although specific buses are described and shown in the embodiments of the application, any suitable buses or interconnects are contemplated by the application.
The electronic device can execute the object identification method in the embodiment of the application based on the currently intercepted spam messages and the messages reported by the user, thereby realizing the object identification method and the object identification device described in conjunction with fig. 1, fig. 4 and fig. 5.
In addition, in combination with the object identification method in the foregoing embodiments, the embodiments of the present application may provide a computer storage medium to implement. The computer storage medium having computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement any of the object recognition methods in the above embodiments.
It is to be understood that the present application is not limited to the particular arrangements and instrumentality described above and shown in the attached drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions or change the order between the steps after comprehending the spirit of the present application.
The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic Circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the present application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware for performing the specified functions or acts, or combinations of special purpose hardware and computer instructions.
As described above, only the specific embodiments of the present application are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present application, and these modifications or substitutions should be covered within the scope of the present application.
Claims (10)
1. An object recognition method, comprising:
acquiring a handwriting sequence corresponding to handwriting written on an input method interface by a user;
inputting the handwriting sequence into a pre-trained handwriting recognition model, and recognizing the probability that each handwriting is a starting pen;
identifying at least one object corresponding to the handwriting sequence according to the probability that each handwriting is a starting pen;
the handwriting recognition model is obtained by training label information of each handwriting in a handwriting sample set based on the handwriting sample set, wherein the label information of each handwriting is information of whether the handwriting is a starting pen or not.
2. A method as claimed in claim 1, wherein the handwriting recognition model includes a first neural network and a second neural network, the inputting the sequence of handwriting into a pre-trained handwriting recognition model, recognizing the probability that each writing handwriting is an originating pen, comprises:
inputting the handwriting sequence into the first neural network, and extracting a first feature vector of each handwriting; the first feature vector of each writing is used for representing the features of the writing and the handwriting features corresponding to the writing before the writing;
inputting the first feature vector into the second neural network, and identifying a probability that each writing script is an initial pen.
3. The method of claim 2, wherein before the obtaining of the handwriting sequence corresponding to the handwriting written by the user on the input method interface, the method further comprises:
acquiring a handwriting sample set, wherein the handwriting sample set comprises a plurality of handwriting training samples, each handwriting training sample comprises a handwriting sequence sample and label information corresponding to each handwriting sample in the handwriting sequence sample, and the label information of each handwriting sample is information of whether the handwriting sample is a starting pen;
for each handwriting sequence sample, respectively executing the following steps:
inputting the handwriting sequence samples into a preset neural network model, and determining the classification result of each handwriting sample;
determining a loss function value according to the classification result of each handwriting sample and the label information corresponding to each handwriting sample;
and training the preset neural network model according to the loss function value to obtain the handwriting recognition model.
4. The method of claim 3, wherein the pre-defined neural network model comprises a convolutional neural network and a long-short term memory network;
inputting the handwriting sequence samples into a preset neural network model, and determining a classification result of each handwriting sample, specifically comprising:
inputting the handwriting sequence samples into the convolutional neural network, and extracting a second feature vector of each handwriting sample in the handwriting sequence samples, wherein the second feature vector is used for representing the handwriting sample and the features of the handwriting sample before the handwriting sample;
and inputting the second feature vector into a long-term and short-term memory network, and determining the classification result of each handwriting sample.
5. A method as claimed in claim 1, wherein said identifying at least one object corresponding to said sequence of writings according to the probability that said each writing is a start pen comprises:
under the condition that the probability that the target writing handwriting is the initial pen is greater than a preset threshold value, determining that the target writing handwriting is the initial pen; wherein the writing handwriting comprises the target writing handwriting;
dividing the handwriting sequence by the target writing handwriting to obtain a plurality of sub-handwriting sequences;
and combining the writing scripts in each sub-handwriting sequence to obtain at least one object corresponding to the handwriting sequence.
6. The method according to claim 5, wherein the dividing the handwriting sequence by the target writing handwriting to obtain a plurality of sub-handwriting sequences comprises:
respectively dividing the handwriting sequence by taking a first writing handwriting corresponding to an Nth handwriting before the target writing handwriting, a second writing handwriting corresponding to an Mth handwriting after the target writing handwriting and the target writing handwriting as initial pens to obtain a plurality of sub-handwriting sequences, wherein M and N are positive integers;
combining the handwriting in each sub-handwriting sequence to generate a word corresponding to each sub-handwriting sequence, and obtaining at least one object corresponding to the handwriting sequence, including:
and combining the writing scripts in each sub-script sequence to obtain at least one object corresponding to each sub-script sequence.
7. The method as claimed in claim 6, wherein after obtaining at least one object corresponding to each of the sub-handwriting sequences according to the plurality of sub-note sequences, the method further comprises:
combining at least one object corresponding to each sub-handwriting sequence into word information;
respectively inputting word information corresponding to each sub-handwriting sequence into a preset word handwriting recognition model, and determining target word information;
and displaying the target word information in an information input box of an electronic equipment interface.
8. An object recognition apparatus, comprising:
the acquisition module is used for acquiring a handwriting sequence corresponding to handwriting written on the input method interface by a user;
the first recognition module is used for inputting the handwriting sequence into a pre-trained handwriting recognition model and recognizing the probability that each handwriting is a starting pen;
the second recognition module is further used for recognizing at least one object corresponding to the handwriting sequence according to the probability that each handwriting is the initial pen;
the handwriting recognition model is obtained by training label information of each handwriting in a handwriting sample set based on the handwriting sample set, wherein the label information of each handwriting is information of whether the handwriting is a starting pen or not.
9. An electronic device, characterized in that the device comprises: a processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the object recognition method of any one of claims 1-7.
10. A computer storage medium having computer program instructions stored thereon which, when executed by a processor, implement the object recognition method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011594494.2A CN112699780A (en) | 2020-12-29 | 2020-12-29 | Object identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011594494.2A CN112699780A (en) | 2020-12-29 | 2020-12-29 | Object identification method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112699780A true CN112699780A (en) | 2021-04-23 |
Family
ID=75511845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011594494.2A Pending CN112699780A (en) | 2020-12-29 | 2020-12-29 | Object identification method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112699780A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114399772A (en) * | 2021-12-20 | 2022-04-26 | 北京百度网讯科技有限公司 | Sample generation, model training and trajectory recognition methods, devices, equipment and medium |
| CN115328383A (en) * | 2022-08-22 | 2022-11-11 | 珠海凯芯微电子科技有限公司 | Identification method and system for display equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1260524A (en) * | 1999-01-13 | 2000-07-19 | 国际商业机器公司 | Method for automatically cutting and identiying hand written Chinese characters and system for using said method |
| CN102156609A (en) * | 2010-12-10 | 2011-08-17 | 上海合合信息科技发展有限公司 | Overlap handwriting input method |
| CN102750552A (en) * | 2012-06-18 | 2012-10-24 | 北京捷通华声语音技术有限公司 | Handwriting recognition method and system as well as handwriting recognition terminal |
| CN106339726A (en) * | 2015-07-17 | 2017-01-18 | 佳能株式会社 | Method and device for handwriting recognition |
| CN106407874A (en) * | 2016-03-25 | 2017-02-15 | 东南大学 | Handwriting recognition method based on handwriting coordinate sequence |
| CN106570518A (en) * | 2016-10-14 | 2017-04-19 | 上海新同惠自动化系统有限公司 | Chinese and Japanese handwritten text identification method |
| CN110968246A (en) * | 2018-09-28 | 2020-04-07 | 北京搜狗科技发展有限公司 | Intelligent Chinese handwriting input recognition method and device |
| CN110992441A (en) * | 2019-12-03 | 2020-04-10 | 上海眼控科技股份有限公司 | Writing track processing method and device |
| CN111401099A (en) * | 2018-12-28 | 2020-07-10 | 中国电信股份有限公司 | Text recognition method, device and storage medium |
| CN111814743A (en) * | 2020-07-30 | 2020-10-23 | 深圳壹账通智能科技有限公司 | Handwriting recognition method and device and computer readable storage medium |
-
2020
- 2020-12-29 CN CN202011594494.2A patent/CN112699780A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1260524A (en) * | 1999-01-13 | 2000-07-19 | 国际商业机器公司 | Method for automatically cutting and identiying hand written Chinese characters and system for using said method |
| CN102156609A (en) * | 2010-12-10 | 2011-08-17 | 上海合合信息科技发展有限公司 | Overlap handwriting input method |
| CN102750552A (en) * | 2012-06-18 | 2012-10-24 | 北京捷通华声语音技术有限公司 | Handwriting recognition method and system as well as handwriting recognition terminal |
| CN106339726A (en) * | 2015-07-17 | 2017-01-18 | 佳能株式会社 | Method and device for handwriting recognition |
| CN106407874A (en) * | 2016-03-25 | 2017-02-15 | 东南大学 | Handwriting recognition method based on handwriting coordinate sequence |
| CN106570518A (en) * | 2016-10-14 | 2017-04-19 | 上海新同惠自动化系统有限公司 | Chinese and Japanese handwritten text identification method |
| CN110968246A (en) * | 2018-09-28 | 2020-04-07 | 北京搜狗科技发展有限公司 | Intelligent Chinese handwriting input recognition method and device |
| CN111401099A (en) * | 2018-12-28 | 2020-07-10 | 中国电信股份有限公司 | Text recognition method, device and storage medium |
| CN110992441A (en) * | 2019-12-03 | 2020-04-10 | 上海眼控科技股份有限公司 | Writing track processing method and device |
| CN111814743A (en) * | 2020-07-30 | 2020-10-23 | 深圳壹账通智能科技有限公司 | Handwriting recognition method and device and computer readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| S. MAPARI, N. CHAUDHARY, S. NAIK ， P. METKEWAR: "Usage of fuzzy rule and SOM based model to identify a handwritten chemical symbol or structures", 《2017 SECOND INTERNATIONAL CONFERENCE ON ELECTRICAL, COMPUTER AND COMMUNICATION TECHNOLOGIES (ICECCT)》 * |
| 刘曼飞: "基于深度学习的联机手写汉字分析与识别", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114399772A (en) * | 2021-12-20 | 2022-04-26 | 北京百度网讯科技有限公司 | Sample generation, model training and trajectory recognition methods, devices, equipment and medium |
| CN114399772B (en) * | 2021-12-20 | 2024-02-27 | 北京百度网讯科技有限公司 | Sample generation, model training and track recognition methods, devices, equipment and media |
| CN115328383A (en) * | 2022-08-22 | 2022-11-11 | 珠海凯芯微电子科技有限公司 | Identification method and system for display equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107330471B (en) | Method and device for problem location of feedback content, computer equipment and storage medium | |
| CN112883968B (en) | Image character recognition method, device, medium and electronic equipment | |
| CN112699780A (en) | Object identification method, device, equipment and storage medium | |
| CN107341143A (en) | A kind of sentence continuity determination methods and device and electronic equipment | |
| CN112052005A (en) | Interface processing method, device, equipment and storage medium | |
| CN112883966A (en) | Image character recognition method, device, medium and electronic equipment | |
| CN114241471B (en) | Video text recognition method and device, electronic equipment and readable storage medium | |
| CN117333889A (en) | Training method and device for document detection model and electronic equipment | |
| CN112416141A (en) | Processing method, device, equipment and medium for input character string | |
| CN116092066A (en) | Text recognition method, device, equipment, medium and product | |
| CN113064497A (en) | Statement identification method, device, equipment and computer storage medium | |
| CN115909376A (en) | Text recognition method, text recognition model training device and storage medium | |
| CN115565186A (en) | Method and device for training character recognition model, electronic equipment and storage medium | |
| CN111339776B (en) | Resume parsing method and device, electronic equipment and computer-readable storage medium | |
| CN115273103A (en) | Text recognition method and device, electronic equipment and storage medium | |
| CN108008879A (en) | Application program searching method and device and electronic equipment | |
| CN115294581A (en) | Method and device for identifying error characters, electronic equipment and storage medium | |
| CN114724144A (en) | Text recognition method, model training method, device, equipment and medium | |
| CN114117031A (en) | List screening method, device, equipment and computer storage medium | |
| CN112417959A (en) | Picture generation method and device, electronic equipment and computer storage medium | |
| CN111753548B (en) | Information acquisition method and device, computer storage medium and electronic equipment | |
| CN114415931A (en) | Electronic whiteboard display method and device, electronic whiteboard and storage medium | |
| CN114049686A (en) | Signature recognition model training method and device and electronic equipment | |
| CN114254650A (en) | Information processing method, device, equipment and medium | |
| CN112861471A (en) | Object display method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210423 |
|
| WD01 | Invention patent application deemed withdrawn after publication |