CNN-SVM handwritten signature recognition method based on VGG16
A technology of handwritten signature and recognition method, which is applied in the field of text image recognition, and can solve problems such as difficulty in handwritten signature recognition, large number, difficulty in Chinese character recognition, etc.
Pending Publication Date: 2019-11-01
HUAIYIN INSTITUTE OF TECHNOLOGY
View PDF9 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
It is difficult to recognize handwritten signatures, mainly due to the complex structure of the signature, the variety of fonts, the large number, the material of the writing paper, the color of the pen used for writing, the force control when writing Chinese characters, and even the angle and pixel when shooting the image. , The difference in light will bring certain difficulties to Chinese character recognition
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0071] Below in conjunction with specific embodiment, further illustrate the present invention, should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention, after having read the present invention, those skilled in the art will understand various equivalent forms of the present invention All modifications fall within the scope defined by the appended claims of the present application.
[0072] like Figure 1~6 Shown, a kind of CNN-SVM handwritten signature recognition method based on VGG16 of the present invention, comprises the steps:
[0073]Step 1: Define the original handwritten signature image dataset as Image0, and obtain the handwritten signature image dataset Image1 after labeling. The specific method is:
[0074] Step 1.1: Define the original handwritten signature image dataset Image0, Image0={C 1 ,C 2 ,...,C m}, where C m It is the mth group of handwritten signatur...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a CNN-SVM handwritten signature recognition method based on VGG16. The method comprises the steps of 1, subjecting a handwritten signature image data set to tagging processing;2, preprocessing the data set through image graying, binaryzation and size normalization in sequence; 3, training a neural network model VGG16 by adopting a public data set ImageNet of a Kaggle company to obtain a weight set; 4, migrating the weight set to a CNN and training to obtain an initial feature matrix; and 5, performing PCA dimension reduction on the initial feature matrix, and inputtingthe initial feature matrix into an SVM for training to obtain a handwritten signature image recognition result. According to the method, the CNN-SVM is improved based on the VGG16. The handwritten signature recognition effect is effectively improved, and the use value of drawing signatures is increased.
Description
technical field [0001] The invention belongs to the technical field of text image recognition, in particular to a CNN-SVM handwritten signature recognition method based on VGG16. Background technique [0002] In the work of handwritten character recognition, there are certain differences in the handwritten glyphs of different writers. To solve this problem, existing papers mainly start with the recognition of a single handwritten Chinese character. However, there are few studies on offline handwritten signature recognition. At the same time, handwritten signatures have characteristics such as continuous strokes and simple strokes, which bring greater difficulty to the recognition work. [0003] The existing research foundations of Feng Wanli, Zhu Quanyin and others include: Wanli Feng. Research of theme statement extraction for chinese literature based on lexicalchain. International Journal of Multimedia and Ubiquitous Engineering, Vol.11, No.6(2016), pp.379- 388; Wanli Fen...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V30/333G06V30/36G06V30/10G06N3/045G06F18/2135G06F18/2411
Inventor 冯万利顾晨洁朱全银董甜甜张柯文
Owner HUAIYIN INSTITUTE OF TECHNOLOGY