CN111814743A - Handwriting recognition method and device and computer readable storage medium - Google Patents
Handwriting recognition method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN111814743A CN111814743A CN202010752375.9A CN202010752375A CN111814743A CN 111814743 A CN111814743 A CN 111814743A CN 202010752375 A CN202010752375 A CN 202010752375A CN 111814743 A CN111814743 A CN 111814743A
- Authority
- CN
- China
- Prior art keywords
- handwriting
- training
- model
- picture
- basic stroke
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000012549 training Methods 0.000 claims abstract description 144
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 48
- 238000012360 testing method Methods 0.000 claims abstract description 46
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 238000011176 pooling Methods 0.000 claims description 60
- 230000006870 function Effects 0.000 claims description 41
- 230000009467 reduction Effects 0.000 claims description 30
- 230000004913 activation Effects 0.000 claims description 16
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims 1
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000007726 management method Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/30—Writer recognition; Reading and verifying signatures
- G06V40/33—Writer recognition; Reading and verifying signatures based only on signature image, e.g. static signature recognition
-
- 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
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Character Discrimination (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to artificial intelligence, and discloses a handwriting identification method, which comprises the following steps: training two pre-constructed convolutional neural network models by utilizing a handwriting original picture set to obtain a basic stroke characteristic graph model and a radical characteristic graph model; performing feature extraction on the test handwriting picture set by using the basic stroke feature picture model and the radical feature picture model to obtain a basic stroke feature picture set and a radical feature picture set; training by using the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a handwriting recognition model; and identifying the picture to be detected by using the handwriting identification model to obtain an identification result. The embodiment of the invention also relates to a block chain technology, and the data of the model training can be stored in the block chain. The invention also provides a handwriting recognition device, electronic equipment and a computer readable storage medium. The invention can improve the accuracy of handwriting recognition.
Description
Technical Field
The present invention relates to the field of artificial intelligence, and in particular, to a handwriting recognition method, apparatus, electronic device, and computer-readable storage medium.
Background
At present, handwriting recognition and identification work on the market is mainly entrusted with an identification organization or professionals to carry out identification, a physical identification method is mainly adopted, the whole identification period is very long, and the efficiency is low.
In addition, the model identification method adopted in the prior art only utilizes handwriting pictures for training, handwriting characteristics are lost more, and model identification accuracy is low, so that a model identification method with high handwriting identification accuracy is required.
Disclosure of Invention
The invention provides a handwriting recognition method, a handwriting recognition device, electronic equipment and a computer-readable storage medium, and mainly aims to improve the accuracy of handwriting recognition.
In order to achieve the above object, the present invention provides a handwriting recognition method, including:
acquiring a handwriting original picture set of a target user, and training a pre-constructed first convolution neural network model by using the handwriting original picture set to obtain a basic stroke characteristic picture model;
acquiring a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set;
training a pre-constructed second convolutional neural network model by using the handwriting original picture set to obtain a component characteristic graph model;
performing radical feature extraction on the test handwriting picture set by using the radical feature picture model to obtain a radical feature picture set;
training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a handwriting recognition model;
and when the picture to be detected is received, identifying the picture to be detected by utilizing the handwriting identification model to obtain an identification result.
Optionally, the training of the first convolutional neural network model pre-constructed by using the handwriting original image set to obtain a basic stroke feature map model includes:
taking the handwriting original picture set as a first training set;
carrying out basic stroke marking on the handwriting original picture set to obtain a first label set;
and training the first convolutional neural network model by using the first training set and the first label set to obtain the basic stroke feature diagram model.
Optionally, the training the first convolutional neural network model using the first training set and the first label set includes:
performing convolution pooling operation on the first training set according to a preset first convolution pooling number to obtain a first dimension reduction data set;
according to a preset first deconvolution frequency, performing deconvolution operation on the first dimensionality reduction data set to obtain a first dimensionality increasing data set;
calculating the first ascending-dimensional data set by using a preset first activation function to obtain a first predicted value, and calculating by using an input parameter of a pre-constructed first loss function according to the first predicted value and a tag value contained in the first tag set to obtain a first loss value;
comparing the first loss value with a preset first loss threshold, if the first loss value is greater than or equal to the first loss threshold, returning to the first convolution pooling times according to a preset first convolution pooling number, and performing convolution pooling operation on the first training set; and if the first loss value is smaller than the first loss threshold value, stopping training to obtain the basic stroke feature graph model.
Optionally, the training a pre-constructed third convolutional neural network model by using the test handwriting picture set, the basic stroke feature map set, and the radical feature map set to obtain a handwriting recognition model includes:
summarizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a third training set;
performing handwriting marking on the third training set to obtain a third label set;
and training the third convolutional neural network model by using the third training set and the third label set to obtain the handwriting recognition model.
Optionally, the training the third convolutional neural network model by using the third training set and the third label set to obtain the handwriting recognition model, including:
according to preset depth separable convolution pooling times, performing depth separable convolution pooling operation on the third training set to obtain a third dimension reduction data set;
calculating the third dimensionality reduction data set by using a preset third activation function to obtain a third predicted value, and calculating by using an input parameter of a pre-constructed third loss function according to the third predicted value and a label value contained in the third label set to obtain a third loss value;
comparing the third loss value with a preset third loss threshold, if the third loss value is greater than or equal to the third loss threshold, returning to the step of performing depth separable convolution pooling according to the preset depth separable convolution pooling times, and performing the depth separable convolution pooling operation on the third training set; and if the third loss value is smaller than the third loss threshold value, stopping training to obtain the handwriting recognition model.
Optionally, the performing a deep separable convolution pooling operation on the third training set to obtain a third reduced-dimension data set includes:
performing packet convolution operation on the third training set to obtain a deep convolution data set;
performing point-by-point convolution operation on the depth convolution data set to obtain a point-by-point convolution data set;
and carrying out average pooling operation on the point-by-point convolution data sets to obtain the third dimension reduction data set.
In order to solve the above problems, the present invention also provides a handwriting recognition apparatus, including:
the basic stroke recognition module is used for acquiring a handwriting original picture set of a target user and training a pre-constructed first convolution neural network model by utilizing the handwriting original picture set to obtain a basic stroke characteristic picture model; acquiring a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set;
the radical identification module is used for training a pre-constructed second convolutional neural network model by utilizing the handwriting original image set to obtain a radical feature map model; performing radical feature extraction on the test handwriting picture set by using the radical feature picture model to obtain a radical feature picture set;
the handwriting recognition module is used for training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model; and when the picture to be detected is received, identifying the picture to be detected by utilizing the handwriting identification model to obtain an identification result.
Optionally, the basic stroke recognition module obtains a basic stroke feature map model by using a first convolutional neural network model which is trained and pre-constructed by the handwriting original image set, and includes:
taking the handwriting original picture set as a first training set;
carrying out basic stroke marking on the handwriting original picture set to obtain a first label set;
and training the first convolutional neural network model by using the first training set and the first label set to obtain the basic stroke feature diagram model.
Optionally, the base stroke recognition module trains the first convolutional neural network model using the first training set and the first label set, including:
performing convolution pooling operation on the first training set according to a preset first convolution pooling number to obtain a first dimension reduction data set;
according to a preset first deconvolution frequency, performing deconvolution operation on the first dimensionality reduction data set to obtain a first dimensionality increasing data set;
calculating the first ascending-dimensional data set by using a preset first activation function to obtain a first predicted value, and calculating by using an input parameter of a pre-constructed first loss function according to the first predicted value and a tag value contained in the first tag set to obtain a first loss value;
comparing the first loss value with a preset first loss threshold, if the first loss value is greater than or equal to the first loss threshold, returning to the first convolution pooling times according to a preset first convolution pooling number, and performing convolution pooling operation on the first training set; and if the first loss value is smaller than the first loss threshold value, stopping training to obtain the basic stroke feature graph model.
Optionally, the handwriting recognition module trains a pre-constructed third convolutional neural network model by using the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model, including:
summarizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a third training set;
performing handwriting marking on the third training set to obtain a third label set;
and training the third convolutional neural network model by using the third training set and the third label set to obtain the handwriting recognition model.
Optionally, the handwriting recognition module trains the third convolutional neural network model by using the third training set and the third label set to obtain the handwriting recognition model, including:
according to preset depth separable convolution pooling times, performing depth separable convolution pooling operation on the third training set to obtain a third dimension reduction data set;
calculating the third dimensionality reduction data set by using a preset third activation function to obtain a third predicted value, and calculating by using an input parameter of a pre-constructed third loss function according to the third predicted value and a label value contained in the third label set to obtain a third loss value;
comparing the third loss value with a preset third loss threshold, if the third loss value is greater than or equal to the third loss threshold, returning to the step of performing depth separable convolution pooling according to the preset depth separable convolution pooling times, and performing the depth separable convolution pooling operation on the third training set; and if the third loss value is smaller than the third loss threshold value, stopping training to obtain the handwriting recognition model.
Optionally, the handwriting recognition module performs a deep separable convolution pooling operation on the third training set to obtain a third reduced-dimension data set, including:
performing packet convolution operation on the third training set to obtain a deep convolution data set;
performing point-by-point convolution operation on the depth convolution data set to obtain a point-by-point convolution data set;
and carrying out average pooling operation on the point-by-point convolution data sets to obtain the third dimension reduction data set.
In order to solve the above problem, the present invention also provides an electronic device, including:
a memory storing at least one instruction; and
and the processor executes the instructions stored in the memory to realize the handwriting recognition method.
In order to solve the above problem, the present invention also provides a computer-readable storage medium including a stored data area and a stored program area, the stored data area storing data created according to the use of blockchain nodes, the stored program area storing a computer program, the computer-readable storage medium having stored therein at least one instruction, the at least one instruction being executed by a processor in an electronic device to implement the handwriting recognition method described above.
In the embodiment of the invention, a basic stroke feature map model is obtained by utilizing the handwriting original picture set to train a pre-constructed first convolution neural network model, basic stroke feature extraction is carried out on the test handwriting picture set by utilizing the basic stroke feature map model, definition and identification of basic strokes are realized through the basic stroke feature map model, and the identification range is reduced; training a pre-constructed second convolutional neural network model by using the handwriting original image set to obtain a radical feature map model, performing radical feature extraction on the test handwriting image set by using the radical feature map model, realizing definition recognition of radicals by using the radical feature map model, and improving the detail feature range of handwriting recognition; training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model, and reducing the range of feature extraction in model training and improving the training precision of the handwriting recognition model by taking the basic stroke feature map set and the radical feature map set as additional training sets; when a picture of the handwriting to be detected is received, the handwriting picture is recognized by the handwriting recognition model to obtain a recognition result, the basic strokes and the radicals are defined and recognized, the range of feature extraction in the training of the handwriting recognition model is narrowed, the training precision of the handwriting recognition model is improved, and therefore the accuracy of handwriting recognition is improved.
Drawings
FIG. 1 is a flow chart of a handwriting recognition method according to an embodiment of the present invention;
FIG. 2 is a block diagram of a handwriting recognition apparatus according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an internal structure of an electronic device implementing a handwriting recognition method according to an embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a handwriting recognition method. Referring to fig. 1, a flow chart of a handwriting recognition method according to an embodiment of the invention is shown. The method may be performed by an apparatus, which may be implemented by software and/or hardware.
In this embodiment, the handwriting recognition method includes:
s1, acquiring a handwriting original picture set of the target user, and training a pre-constructed first convolution neural network model by using the handwriting original picture set to obtain a basic stroke feature picture model.
In the embodiment of the invention, the handwriting original picture set is a set of target user historical handwriting, for example, the historical notes can be in a picture form. The original handwriting picture set of the target user can be obtained from past archives, academic records, hand-written manuscripts and the like of the target user in a picture scanning mode. The target user is a party needing handwriting recognition authentication.
Preferably, in the embodiment of the present invention, the first convolutional neural network model may be a full convolutional neural network model.
In detail, in the embodiment of the present invention, the handwriting original image set is determined as a first training set, and basic strokes are performed on the handwriting original image set to obtain a first label set. Wherein the basic strokes are (left stroke), horizontal stroke, vertical stroke, left stroke, right stroke, left stroke, right stroke
Lifting device
Eight kinds of Chinese character five and hook. Preferably, the embodiment of the present invention can use the Label Me image annotation tool to manually mark the basic strokes.
Further, training the first convolutional neural network model using the first training set and the first label set according to the embodiment of the present invention includes:
step A: performing convolution pooling operation on the first training set according to a preset first convolution pooling number to obtain a first dimension reduction data set;
and B: according to a preset first deconvolution frequency, performing deconvolution operation on the first dimensionality reduction data set to obtain a first dimensionality increasing data set;
and C: and calculating the first ascending-dimensional data set by using a preset first activation function to obtain a first predicted value, and calculating by using a pre-constructed first loss function according to the first predicted value and the label value contained in the first label set to obtain a first loss value.
Step D: comparing the first loss value with a preset first loss threshold value, and if the first loss value is greater than or equal to the first loss threshold value, returning to the step A; and if the first loss value is smaller than the first loss threshold value, stopping training to obtain the basic stroke recognition feature map model.
In detail, the convolution pooling operation includes: convolution operations and pooling operations.
Further, the convolution operation is:
and ω' is a first convolution data set, ω is the first training set, k is the size of a preset convolution kernel, f is the step of a preset convolution operation, and p is a preset data zero padding matrix.
Preferably, in the embodiment of the present invention, the pooling operation is a maximal pooling operation performed on the first convolution data set to obtain the first dimension reduction data set.
Further, in a preferred embodiment of the present invention, the first activation function includes:
wherein, mutRepresenting the first predicted value, s representing data in the first up-dimensioned data set.
In detail, the first loss function according to the preferred embodiment of the present invention includes:
wherein T represents the first loss value, n is the number of data of the first training set, T is a positive integer, ytIs the first label value, mutIs the first predicted value.
S2, obtaining a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set;
in the embodiment of the invention, the test handwriting picture set is a set of handwriting pictures of another part of the target user different from the handwriting original picture set.
The definition recognition of the basic strokes is realized through the basic stroke characteristic graph model, and the training precision of the subsequent handwriting recognition model is improved.
And S3, training a pre-constructed second convolutional neural network model by using the handwriting original picture set to obtain a component characteristic picture model.
In detail, the embodiment of the invention determines the handwriting original image set as a second training set, and performs radical marking on the handwriting original image set to obtain a second label set. Wherein, the radical is as follows: water (at three points) (the side of the rectangle), the handle (the back of the bowl), the four-point bottom (), etc.
Preferably, the embodiment of the invention can use a Label Me image annotation tool to manually mark the radicals.
Further, the training the second convolutional neural network model by using the second training set and the second label set in the embodiment of the present invention includes:
s31: performing convolution pooling operation on the second training set according to a preset second convolution pooling number to obtain a second dimension reduction data set;
s32: according to a preset second deconvolution frequency, performing deconvolution operation on the second dimensionality reduction data set to obtain a second dimensionality increasing data set;
s33: and calculating the second ascending-dimensional data set by using a preset second activation function to obtain a second predicted value, and calculating by using a pre-constructed second loss function according to the second predicted value and the label value contained in the second label set to obtain a second loss value.
S34: and comparing the second loss value with a preset second loss threshold value, and if the second loss value is greater than or equal to the second loss threshold value, returning to the step S31. And if the second loss value is smaller than the second loss threshold value, stopping training to obtain the radical feature map model.
In detail, in the preferred embodiment of the present invention, the second loss function E can be calculated by using the following formula:
wherein b represents a set composed of tag values included in all the second tag sets, and c represents a set composed of all the second predicted values.
S4, extracting the character of the radicals from the test handwriting image set by using the character image model of the radicals to obtain a character image set of the radicals;
the definition recognition of the radicals is realized through the radical feature map model, and the precision of the subsequent handwriting recognition model training is improved.
And S5, training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a handwriting recognition model.
In the embodiment of the invention, the test handwriting picture set, the basic stroke feature map set and the radical feature map set are collected to obtain a third training set, and the third training set is subjected to handwriting marking to obtain a third label set. Preferably, the embodiment of the invention can manually perform handwriting marking by using a Label Me image marking tool.
Preferably, in the preferred embodiment of the present invention, the third neural convolutional network model can be constructed by using a deep separable convolutional network model.
Further, the training the third convolutional neural network model by using the third training set and the third label set in the embodiment of the present invention includes:
s51: according to preset depth separable convolution pooling times, performing depth separable convolution pooling operation on the third training set to obtain a third dimension reduction data set;
s52: and calculating the third dimensionality reduction data set by using a preset third activation function to obtain a third predicted value, and calculating by using a pre-constructed third loss function to obtain a third loss value according to the third predicted value and the label value contained in the third label set.
S53: comparing the third loss value with a preset third loss threshold value, and returning to S51 if the third loss value is greater than or equal to the third loss threshold value; and if the third loss value is smaller than the third loss threshold value, stopping training to obtain the handwriting recognition model.
In detail, the depth separable convolution pooling operation includes: and performing grouping convolution operation on the third training set to obtain a depth convolution data set, performing point-by-point convolution operation on the depth convolution data set to obtain a point-by-point convolution data set, and performing average pooling operation on the point-by-point convolution data set to obtain the third dimension reduction data set.
In the preferred embodiment of the present invention, the third activation function can be calculated by using the following formula:
f(x)＝ax(0,x)
wherein f (x) is the third predicted value, and x is data in the third reduced-dimension dataset.
In the preferred embodiment of the present invention, the third loss function can be calculated by using the following formula:
wherein N is the number of data included in the third training sample, i is a positive integer, and hiFor the tag value, m, contained in the third set of tagsiIs the third predicted value.
The basic stroke feature atlas and the radical feature atlas are determined to be an additional training set, so that the range of feature extraction in model training is narrowed, and the training precision of the handwriting recognition model is improved.
In another embodiment of the present invention, the basic stroke feature map model, the radical feature map model and the data trained by the handwriting recognition model may be stored in a blockchain.
And S6, when the picture to be detected is received, recognizing the picture to be detected by using the handwriting recognition model to obtain a recognition result.
In the embodiment of the invention, the pictures to be detected are pictures of different handwriting. For example: hand-written signature pictures and hand-written article pictures.
Further, the embodiment of the invention utilizes the handwriting recognition model to perform handwriting recognition on the picture to be detected and outputs the handwriting recognition probability; confirming a confidence threshold value of the handwriting recognition by using a johnson index principle; comparing the handwriting recognition probability to the confidence threshold; and when the handwriting recognition probability is greater than or equal to the confidence threshold, judging that the handwriting to be detected is the target user handwriting, and when the handwriting recognition probability is less than or equal to the confidence threshold, judging that the handwriting to be detected is not the target user handwriting.
In the embodiment of the invention, a basic stroke feature map model is obtained by utilizing the handwriting original picture set to train a pre-constructed first convolution neural network model, basic stroke feature extraction is carried out on the test handwriting picture set by utilizing the basic stroke feature map model, definition and identification of basic strokes are realized through the basic stroke feature map model, and the identification range is reduced; training a pre-constructed second convolutional neural network model by using the handwriting original image set to obtain a radical feature map model, performing radical feature extraction on the test handwriting image set by using the radical feature map model, realizing definition recognition of radicals by using the radical feature map model, and improving the detail feature range of handwriting recognition; training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model, and reducing the range of feature extraction in model training and improving the training precision of the handwriting recognition model by taking the basic stroke feature map set and the radical feature map set as additional training sets; when a picture of the handwriting to be detected is received, the handwriting picture is recognized by the handwriting recognition model to obtain a recognition result, the basic strokes and the radicals are defined and recognized, the range of feature extraction in the training of the handwriting recognition model is narrowed, the training precision of the handwriting recognition model is improved, and therefore the accuracy of handwriting recognition is improved.
FIG. 2 is a functional block diagram of the handwriting recognition device of the present invention.
The handwriting recognition apparatus 100 of the present invention can be installed in an electronic device. According to the realized functions, the handwriting recognition device can comprise a basic stroke recognition module 101, a radical recognition module 102 and a handwriting recognition module 103. A module according to the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the basic stroke recognition module 101 is configured to obtain a handwriting original image set of a target user, and train a pre-constructed first convolution neural network model by using the handwriting original image set to obtain a basic stroke feature map model; and acquiring a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set.
In the embodiment of the invention, the handwriting original picture set is a set of target user historical handwriting, for example, the historical notes can be in the form of pictures. The original handwriting picture set of the target user can be obtained from past archives, academic records, hand-written manuscripts and the like of the target user in a picture scanning mode.
Preferably, in the embodiment of the present invention, the first convolutional neural network model may be a full convolutional neural network model.
In detail, in the embodiment of the present invention, the basic stroke recognition module 101 determines the handwriting original image set as a first training set, and performs basic stroke marking on the handwriting original image set to obtain a first label set. Wherein, the basic strokes are eight kinds (stroke), horizontal stroke, vertical stroke, left falling stroke, right falling stroke and hook stroke. Preferably, the embodiment of the present invention can use the Label Me image annotation tool to manually mark the basic strokes.
Further, in the embodiment of the present invention, the basic stroke recognition module 101 trains the first convolutional neural network model by using the following means:
a: performing convolution pooling operation on the first training set according to a preset first convolution pooling number to obtain a first dimension reduction data set;
b: according to a preset first deconvolution frequency, performing deconvolution operation on the first dimensionality reduction data set to obtain a first dimensionality increasing data set;
c: and calculating the first ascending-dimensional data set by using a preset first activation function to obtain a first predicted value, and calculating by using a pre-constructed first loss function according to the first predicted value and the label value contained in the first label set to obtain a first loss value.
D: comparing the first loss value with a preset first loss threshold value, and if the first loss value is greater than or equal to the first loss threshold value, returning to the step A; and if the first loss value is smaller than the first loss threshold value, stopping training to obtain the basic stroke recognition feature map model.
In detail, the convolution pooling operation includes: convolution operations and pooling operations.
Further, the convolution operation is:
wherein ω' is a first convolution data set, ω is the first training set, k is a size of a preset convolution kernel, f is a step of a preset convolution operation, and p is a preset data zero padding matrix;
preferably, the pooling operation in this embodiment of the present invention is that the base stroke recognition module 101 performs a maximum pooling operation on the first convolved data set to obtain the first reduced-dimension data set.
Further, in a preferred embodiment of the present invention, the first activation function includes:
wherein mutRepresenting the first predicted value, s representing data in the first up-dimensioned data set.
In detail, the first loss function according to the preferred embodiment of the present invention includes:
wherein T represents the first loss value, n is the number of data of the first training set, T is a positive integer, ytIs the first label value, mutIs the first predicted value.
In the embodiment of the invention, the test handwriting picture set is a set of handwriting pictures of another part of the target user different from the handwriting original picture set.
The definition recognition of the basic strokes is realized through the basic stroke characteristic graph model, and the training precision of the subsequent handwriting recognition model is improved.
The radical recognition module 102 is configured to train a pre-constructed second convolutional neural network model by using the handwriting original image set to obtain a radical feature map model; and performing the feature extraction of the radicals on the test handwriting picture set by utilizing the radical feature picture model to obtain a radical feature picture set.
In detail, the radical recognition module 102 according to the embodiment of the present invention determines the handwriting original image set as a second training set, and performs radical marking on the handwriting original image set to obtain a second label set. Wherein, the radical is as follows: water (at three points) (the side of the rectangle), the handle (the back of the bowl), the four-point bottom (), etc.
Preferably, the embodiment of the invention can use a Label Me image annotation tool to manually mark the radicals.
Further, the training of the second convolutional neural network model by the radical identification module 102 according to the embodiment of the present invention by using the following means includes:
h: performing convolution pooling operation on the second training set according to a preset second convolution pooling number to obtain a second dimension reduction data set;
i: according to a preset second deconvolution frequency, performing deconvolution operation on the second dimensionality reduction data set to obtain a second dimensionality increasing data set;
j: and calculating the second ascending-dimensional data set by using a preset second activation function to obtain a second predicted value, and calculating by using a pre-constructed second loss function according to the second predicted value and the label value contained in the second label set to obtain a second loss value.
K: and comparing the second loss value with a preset second loss threshold value, and if the second loss value is greater than or equal to the second loss threshold value, returning to H. And if the second loss value is smaller than the second loss threshold value, stopping training to obtain the radical feature map model.
In detail, in the preferred embodiment of the present invention, the second loss function E can be calculated by using the following formula:
wherein b represents a set composed of tag values included in all the second tag sets, and c represents a set composed of all the second predicted values.
The definition recognition of the radicals is realized through the radical feature map model, and the precision of the subsequent handwriting recognition model training is improved.
The handwriting recognition module 103 is configured to train a pre-constructed third convolutional neural network model by using the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model; and when the picture to be detected is received, identifying the picture to be detected by utilizing the handwriting identification model to obtain an identification result.
In the embodiment of the present invention, the handwriting recognition module 103 collects the test handwriting picture set, the basic stroke feature map set, and the radical feature map set to obtain a third training set; and the handwriting recognition module 103 performs handwriting marking on the third training set to obtain a third label set. Preferably, the embodiment of the invention can manually perform handwriting marking by using a Label Me image marking tool.
Preferably, in the preferred embodiment of the present invention, the third neural convolutional network model can be constructed by using a deep separable convolutional network model.
Further, the training of the third convolutional neural network model by the handwriting recognition module 103 according to the embodiment of the present invention includes:
x: according to preset depth separable convolution pooling times, performing depth separable convolution pooling operation on the third training set to obtain a third dimension reduction data set;
y: and calculating the third dimensionality reduction data set by using a preset third activation function to obtain a third predicted value, and calculating by using a pre-constructed third loss function to obtain a third loss value according to the third predicted value and the label value contained in the third label set.
Z: comparing the third loss value with a preset third loss threshold value, and if the third loss value is greater than or equal to the third loss threshold value, returning to X; and if the third loss value is smaller than the third loss threshold value, stopping training to obtain the handwriting recognition model.
In detail, the handwriting recognition module 103 performs a depth separable convolution pooling operation using:
performing packet convolution operation on the third training set to obtain a deep convolution data set;
performing point-by-point convolution operation on the depth convolution data set to obtain a point-by-point convolution data set;
and carrying out average pooling operation on the point-by-point convolution data sets to obtain the third dimension reduction data set.
In the preferred embodiment of the present invention, the third activation function can be calculated by using the following formula:
f(x)＝ax(0,x)
wherein f (x) is the third predicted value and x is data in the third reduced-dimension dataset.
In the preferred embodiment of the present invention, the third loss function can be calculated by using the following formula:
wherein N is the number of data included in the third training sample, i is a positive integer, and hiFor the tag value, m, contained in the third set of tagsiIs the third predicted value.
In another embodiment of the present invention, the basic stroke feature map model, the radical feature map model and the data trained by the handwriting recognition model may be stored in a blockchain.
The basic stroke feature atlas and the radical feature atlas are determined to be an additional training set, so that the range of feature extraction in model training is narrowed, and the training precision of the handwriting recognition model is improved.
In the embodiment of the invention, the pictures to be detected are pictures of different handwriting. For example: hand-written signature pictures and hand-written article pictures.
Further, the handwriting recognition module 103 of the embodiment of the present invention performs handwriting recognition on the to-be-detected picture by using the handwriting recognition model, and outputs a handwriting recognition probability; the handwriting recognition module 103 confirms the confidence threshold of the handwriting recognition by using the johnson index principle; the handwriting recognition module 103 compares the handwriting recognition probability with the confidence threshold; and when the handwriting recognition probability is greater than or equal to the confidence threshold, judging that the handwriting to be detected is the target user handwriting, and when the handwriting recognition probability is less than or equal to the confidence threshold, judging that the handwriting to be detected is not the target user handwriting.
FIG. 3 is a schematic structural diagram of an electronic device for implementing a handwriting recognition method according to the present invention.
The electronic device 1 may comprise a processor 10, a memory 11 and a bus, and may further comprise a computer program, such as a handwriting recognition program, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, which includes flash memory, removable hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used not only to store application software installed in the electronic device 1 and various types of data, such as codes of handwriting recognition programs, etc., but also to temporarily store data that has been output or is to be output.
The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device 1 by running or executing programs or modules (e.g., handwriting recognition programs, etc.) stored in the memory 11 and calling data stored in the memory 11.
The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 11 and at least one processor 10 or the like.
Fig. 3 shows only an electronic device with components, and it will be understood by those skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than those shown, or some components may be combined, or a different arrangement of components.
For example, although not shown, the electronic device 1 may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so as to implement functions of charge management, discharge management, power consumption management, and the like through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device 1 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the electronic device 1 may further include a network interface, and optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a bluetooth interface, etc.), which are generally used for establishing a communication connection between the electronic device 1 and other electronic devices.
Optionally, the electronic device 1 may further comprise a user interface, which may be a Display (Display), an input unit (such as a Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the electronic device 1 and for displaying a visualized user interface, among other things.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The handwriting recognition program 12 stored in the memory 11 of the electronic device 1 is a combination of instructions that, when executed in the processor 10, may implement:
acquiring a handwriting original picture set of a target user, and training a pre-constructed first convolution neural network model by using the handwriting original picture set to obtain a basic stroke characteristic picture model;
acquiring a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set;
training a pre-constructed second convolutional neural network model by using the handwriting original picture set to obtain a component characteristic graph model;
performing radical feature extraction on the test handwriting picture set by using the radical feature picture model to obtain a radical feature picture set;
training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a handwriting recognition model;
and when the picture to be detected is received, identifying the picture to be detected by utilizing the handwriting identification model to obtain an identification result.
Specifically, the specific implementation method of the processor 10 for the instruction may refer to the description of the relevant steps in the embodiment corresponding to fig. 1, which is not described herein again.
Further, the integrated modules/units of the electronic device 1, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. The computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM).
Further, the computer usable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the blockchain node, and the like.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. A method of handwriting recognition, the method comprising:
acquiring a handwriting original picture set of a target user, and training a pre-constructed first convolution neural network model by using the handwriting original picture set to obtain a basic stroke characteristic picture model;
acquiring a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set;
training a pre-constructed second convolutional neural network model by using the handwriting original picture set to obtain a component characteristic graph model;
performing radical feature extraction on the test handwriting picture set by using the radical feature picture model to obtain a radical feature picture set;
training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a handwriting recognition model;
and when the picture to be detected is received, identifying the picture to be detected by utilizing the handwriting identification model to obtain an identification result.
2. The handwriting recognition method according to claim 1, wherein the training of the pre-constructed first convolutional neural network model by using the handwriting original picture set to obtain a basic stroke feature map model comprises:
taking the handwriting original picture set as a first training set;
carrying out basic stroke marking on the handwriting original picture set to obtain a first label set;
and training the first convolutional neural network model by using the first training set and the first label set to obtain the basic stroke feature diagram model.
3. The handwriting recognition method of claim 2, wherein said training the first convolutional neural network model using the first training set and the first set of labels, comprises:
performing convolution pooling operation on the first training set according to a preset first convolution pooling number to obtain a first dimension reduction data set;
according to a preset first deconvolution frequency, performing deconvolution operation on the first dimensionality reduction data set to obtain a first dimensionality increasing data set;
calculating the first ascending-dimensional data set by using a preset first activation function to obtain a first predicted value, and calculating by using an input parameter of a pre-constructed first loss function according to the first predicted value and a tag value contained in the first tag set to obtain a first loss value;
comparing the first loss value with a preset first loss threshold, if the first loss value is greater than or equal to the first loss threshold, returning to the first convolution pooling times according to a preset first convolution pooling number, and performing convolution pooling operation on the first training set; and if the first loss value is smaller than the first loss threshold value, stopping training to obtain the basic stroke feature graph model.
4. The handwriting recognition method according to claim 1, wherein training a pre-constructed third convolutional neural network model using the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model comprises:
summarizing the test handwriting picture set, the basic stroke feature picture set and the radical feature picture set to obtain a third training set;
performing handwriting marking on the third training set to obtain a third label set;
and training the third convolutional neural network model by using the third training set and the third label set to obtain the handwriting recognition model.
5. The handwriting recognition method according to claim 4, wherein said training the third convolutional neural network model using the third training set and the third label set to obtain the handwriting recognition model comprises:
according to preset depth separable convolution pooling times, performing depth separable convolution pooling operation on the third training set to obtain a third dimension reduction data set;
calculating the third dimensionality reduction data set by using a preset third activation function to obtain a third predicted value, and calculating by using an input parameter of a pre-constructed third loss function according to the third predicted value and a label value contained in the third label set to obtain a third loss value;
comparing the third loss value with a preset third loss threshold, if the third loss value is greater than or equal to the third loss threshold, returning to the step of performing depth separable convolution pooling according to the preset depth separable convolution pooling times, and performing the depth separable convolution pooling operation on the third training set; and if the third loss value is smaller than the third loss threshold value, stopping training to obtain the handwriting recognition model.
6. A method for handwriting recognition according to claim 5 and wherein said performing a deep separable convolution pooling operation on said third training set to obtain a third reduced-dimension dataset comprises:
performing packet convolution operation on the third training set to obtain a deep convolution data set;
performing point-by-point convolution operation on the depth convolution data set to obtain a point-by-point convolution data set;
and carrying out average pooling operation on the point-by-point convolution data sets to obtain the third dimension reduction data set.
7. A handwriting recognition apparatus, comprising:
the basic stroke recognition module is used for acquiring a handwriting original picture set of a target user and training a pre-constructed first convolution neural network model by utilizing the handwriting original picture set to obtain a basic stroke characteristic picture model; acquiring a test handwriting picture set, and extracting basic stroke features of the test handwriting picture set by using the basic stroke feature map model to obtain a basic stroke feature map set;
the radical identification module is used for training a pre-constructed second convolutional neural network model by utilizing the handwriting original image set to obtain a radical feature map model; performing radical feature extraction on the test handwriting picture set by using the radical feature picture model to obtain a radical feature picture set;
the handwriting recognition module is used for training a pre-constructed third convolutional neural network model by utilizing the test handwriting picture set, the basic stroke feature map set and the radical feature map set to obtain a handwriting recognition model; and when the picture to be detected is received, identifying the picture to be detected by utilizing the handwriting identification model to obtain an identification result.
8. The stroke recognition apparatus of claim 7 wherein the base stroke recognition module obtains the base stroke feature map model by:
taking the handwriting original picture set as a first training set;
carrying out basic stroke marking on the handwriting original picture set to obtain a first label set;
and training the first convolutional neural network model by using the first training set and the first label set to obtain the basic stroke feature diagram model.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method of handwriting recognition according to any one of claims 1 to 6.
10. A computer-readable storage medium, storing a computer program, characterized in that the computer program, when executed by a processor, implements a method of handwriting recognition according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010752375.9A CN111814743A (en) | 2020-07-30 | 2020-07-30 | Handwriting recognition method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010752375.9A CN111814743A (en) | 2020-07-30 | 2020-07-30 | Handwriting recognition method and device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111814743A true CN111814743A (en) | 2020-10-23 |
Family
ID=72862860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010752375.9A Pending CN111814743A (en) | 2020-07-30 | 2020-07-30 | Handwriting recognition method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111814743A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112198966A (en) * | 2020-12-08 | 2021-01-08 | 中南大学 | Stroke identification method and system based on FMCW radar system |
| CN112699780A (en) * | 2020-12-29 | 2021-04-23 | 上海臣星软件技术有限公司 | Object identification method, device, equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0896081A (en) * | 1994-09-29 | 1996-04-12 | Toshiba Corp | Character recognizing device and character recognizing method |
| CN1641686A (en) * | 2004-01-17 | 2005-07-20 | 中国科学院计算技术研究所 | Confused stroke order library establishing method and on-line hand-writing Chinese character identifying and evaluating system |
| JP2005208729A (en) * | 2004-01-20 | 2005-08-04 | Toshiba Corp | Writer identification method and writer identification apparatus |
| CN105809100A (en) * | 2014-12-31 | 2016-07-27 | 中国刑事警察学院 | Handwriting and signature quantization inspection and identification characteristic product method |
| CN109446905A (en) * | 2018-09-26 | 2019-03-08 | 深圳壹账通智能科技有限公司 | Sign electronically checking method, device, computer equipment and storage medium |
| CN110555403A (en) * | 2019-08-29 | 2019-12-10 | 韶关市启之信息技术有限公司 | handwritten character evaluation method and system |
| CN111046774A (en) * | 2019-12-06 | 2020-04-21 | 国网湖北省电力有限公司电力科学研究院 | Chinese signature handwriting identification method based on convolutional neural network |
| CN111079503A (en) * | 2019-08-02 | 2020-04-28 | 广东小天才科技有限公司 | Character recognition method and electronic equipment |
| CN111259726A (en) * | 2019-12-04 | 2020-06-09 | 湖北工业大学 | Method for extracting and authenticating stroke dynamic characteristics in online handwriting authentication |
-
2020
- 2020-07-30 CN CN202010752375.9A patent/CN111814743A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0896081A (en) * | 1994-09-29 | 1996-04-12 | Toshiba Corp | Character recognizing device and character recognizing method |
| CN1641686A (en) * | 2004-01-17 | 2005-07-20 | 中国科学院计算技术研究所 | Confused stroke order library establishing method and on-line hand-writing Chinese character identifying and evaluating system |
| JP2005208729A (en) * | 2004-01-20 | 2005-08-04 | Toshiba Corp | Writer identification method and writer identification apparatus |
| CN105809100A (en) * | 2014-12-31 | 2016-07-27 | 中国刑事警察学院 | Handwriting and signature quantization inspection and identification characteristic product method |
| CN109446905A (en) * | 2018-09-26 | 2019-03-08 | 深圳壹账通智能科技有限公司 | Sign electronically checking method, device, computer equipment and storage medium |
| CN111079503A (en) * | 2019-08-02 | 2020-04-28 | 广东小天才科技有限公司 | Character recognition method and electronic equipment |
| CN110555403A (en) * | 2019-08-29 | 2019-12-10 | 韶关市启之信息技术有限公司 | handwritten character evaluation method and system |
| CN111259726A (en) * | 2019-12-04 | 2020-06-09 | 湖北工业大学 | Method for extracting and authenticating stroke dynamic characteristics in online handwriting authentication |
| CN111046774A (en) * | 2019-12-06 | 2020-04-21 | 国网湖北省电力有限公司电力科学研究院 | Chinese signature handwriting identification method based on convolutional neural network |
Non-Patent Citations (3)
| Title |
|---|
| 文颖, 龚裕, 张宪荣: "基于基本笔画运笔特征的汉字笔迹鉴定的研究", 上海大学学报(自然科学版), no. 01, 28 February 2002 (2002-02-28) * |
| 李琰延;李大娟;李楠;邢问天;: "基于狼群优化概率神经网络的笔迹鉴定应用", 中国新通信, no. 10, 20 May 2015 (2015-05-20) * |
| 杜英杰;陈杰;李晴;: "笔迹检验鉴定量化研究理论与方法", 山东化工, no. 22, 23 November 2017 (2017-11-23) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112198966A (en) * | 2020-12-08 | 2021-01-08 | 中南大学 | Stroke identification method and system based on FMCW radar system |
| CN112198966B (en) * | 2020-12-08 | 2021-03-16 | 中南大学 | Stroke identification method and system based on FMCW radar system |
| CN112699780A (en) * | 2020-12-29 | 2021-04-23 | 上海臣星软件技术有限公司 | Object identification method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112528863A (en) | Identification method and device of table structure, electronic equipment and storage medium | |
| CN112699775A (en) | Certificate identification method, device and equipment based on deep learning and storage medium | |
| CN111932564B (en) | Picture identification method and device, electronic equipment and computer readable storage medium | |
| CN111932534B (en) | Medical image picture analysis method and device, electronic equipment and readable storage medium | |
| CN112396005A (en) | Biological characteristic image recognition method and device, electronic equipment and readable storage medium | |
| CN112052850A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN113283446A (en) | Method and device for identifying target object in image, electronic equipment and storage medium | |
| CN112581227A (en) | Product recommendation method and device, electronic equipment and storage medium | |
| CN112860905A (en) | Text information extraction method, device and equipment and readable storage medium | |
| CN112528616A (en) | Business form generation method and device, electronic equipment and computer storage medium | |
| CN111814743A (en) | Handwriting recognition method and device and computer readable storage medium | |
| CN112885423A (en) | Disease label detection method and device, electronic equipment and storage medium | |
| CN111933274A (en) | Disease classification diagnosis method and device, electronic equipment and storage medium | |
| CN114491047A (en) | Multi-label text classification method and device, electronic equipment and storage medium | |
| CN112132016A (en) | Bill information extraction method and device and electronic equipment | |
| CN114708461A (en) | Multi-modal learning model-based classification method, device, equipment and storage medium | |
| CN111930897B (en) | Patent retrieval method, device, electronic equipment and computer-readable storage medium | |
| CN112528984A (en) | Image information extraction method, device, electronic equipment and storage medium | |
| CN112329666A (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN114996386A (en) | Business role identification method, device, equipment and storage medium | |
| CN114120347A (en) | Form verification method and device, electronic equipment and storage medium | |
| CN113869385A (en) | Poster comparison method, device and equipment based on target detection and storage medium | |
| CN112561893A (en) | Picture matching method and device, electronic equipment and storage medium | |
| CN113255456A (en) | Non-active living body detection method, device, electronic equipment and storage medium | |
| CN113705686A (en) | Image classification method and device, electronic equipment and readable 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 |