Deep learning digital handwriting identification method based on ARM (Advanced RISC Machines) platform

A technology of deep learning and recognition methods, applied in character and pattern recognition, instruments, computer components, etc., can solve the problems of non-real-time performance, complex structure, and high cost of desktop operating systems, and achieve fast modeling speed and simple development , low cost effect

Pending Publication Date: 2018-11-13
NANJING UNIV OF POSTS & TELECOMM
View PDF0 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

PC-based image processing systems often easily show many shortcomings such as complex structure, high cost, large volume, and huge power c

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep learning digital handwriting identification method based on ARM (Advanced RISC Machines) platform
  • Deep learning digital handwriting identification method based on ARM (Advanced RISC Machines) platform
  • Deep learning digital handwriting identification method based on ARM (Advanced RISC Machines) platform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] Specific embodiments of the present invention are described in detail below, but it should be understood that the protection scope of the present invention is not limited by the specific embodiments.

[0050] The present invention provides a handwritten digit recognition method based on the ARM platform and deep learning, and the specific process is as follows figure 1 shown.

[0051] Step (1): Construct a softmax deep learning neural network model, such as figure 2 shown.

[0052] Weight initialization: initialize the weight, the weight is initialized with truncated_normal, and the standard deviation of stddev is defined as 0.1. In order to avoid the problem of zero gradient, a constant 0.1 is used for bias initialization. truncated_normal: Truncated normal distribution random numbers, mean mean, standard deviation stddev, only keep random numbers in the range of [mean-2×stddev, mean+2×stddev].

[0053] Convolution and pooling: TensorFlow has great flexibility in ...

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

No PUM Login to view more

Abstract

The invention relates to a deep learning digital handwriting identification method based on an ARM (Advanced RISC Machines) platform, and belongs to the field of image recognition. The method comprises the following specific steps that: constructing a softmax deep learning neural network model; utilizing an MNIST training set to train the model; transplanting the deep learning model to the ARM platform; preprocessing a handwriting digital photo shot by a camera; and identifying and outputting a result. By use of the method, TensorFlow is used for carrying out deep learning, so that the methodhas the advantages of being simple in development, high in modeling speed and high in identification accuracy, and in addition, the ARM embedded platform has the advantages of low cost, good stability, high instantaneity and the like.

Description

technical field [0001] The invention relates to a digital handwriting recognition method, in particular to a digital handwriting recognition method based on an ARM platform and using a convolutional neural network and a softmax regression calculation method, belonging to the technical field of image recognition processing. Background technique [0002] The main function of image recognition technology is to distinguish the objects in the image according to the observed image, so as to make corresponding meaningful judgments. The specific realization is to apply modern information processing technology and computer technology to human Cognitive processes are simulated. Usually, an image recognition system consists of three parts: image segmentation, image feature extraction, and classifier recognition. Among them, the main function of image segmentation is to divide the image into multiple regions; The corresponding features are extracted from the image of each area; the rec...

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
no application Login to view more
IPC IPC(8): G06K9/68
CPCG06V30/2455
Inventor 杨敏黄佳凯荆晓远程雷
Owner NANJING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap