Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method for detecting characters in complex natural scene image

A natural scene image and text detection technology, applied in the field of computer vision and pattern recognition, can solve problems such as redundancy and false detection of detection frames

Pending Publication Date: 2021-02-26
HUNAN NORMAL UNIVERSITY
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In turn, in the process of text detection, there are certain problems of false detection and redundant detection frames.

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
  • Method for detecting characters in complex natural scene image
  • Method for detecting characters in complex natural scene image
  • Method for detecting characters in complex natural scene image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0071] A text detection method in complex natural scene images, comprising the following steps:

[0072] Step 1: Scene preprocessing of image data, prepare datasets, the datasets used include: ICDAR2013, ICDAR2015, ICDAR2017, CTW-1500, MSRA-TD500, VGGSynthText-part;

[0073] Step 2: Building a network framework: The present invention uses a batch-normalized VGG-16-based fully convolutional network architecture as the backbone network (the present invention can also use other convolutional networks such as Resnet as the backbone network). Our model has skip connections in the decoding part, which is similar to U-net in that it aggregates low-level features. For VGG-16, the convolutional feature maps with 4 upsampling ratios are used as the final convolutional map. The final output has two branches: the character detection branch and the text line detection branch. The image is first sent to a fully convolutional neural network for feature extraction, and then the text discrim...

Embodiment 2

[0117] Such as Figure 7 Shown is a functional block diagram of a text detection device 100 in a complex natural scene image of the present invention.

[0118] The device 100 for character detection in a complex natural scene image of the present invention can be installed in an electronic device. According to the realized functions, the device 100 for character detection in complex natural scene images may include an image acquisition module 101 , a feature extraction and discrimination module 102 , a character line detection module 103 , a character detection module 104 , and a text correction module 105 . The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of the electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0119] In this embodiment, the functions of each module / unit are as follows:

[0120...

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 discloses a method for detecting characters in a complex natural scene image, belongs to the field of computer vision and pattern recognition, and relates to the technical field of neural networks and computer vision, in particular to a method for realizing character detection in a complex scene based on deep learning. The character detection method based on character annotation andthe character detection method based on word annotation are fused, the combined characteristics between characters are learned, the false detection rate of the characters can be reduced, the redundancy of a detection box is reduced, and the method has the capacity of flexibly coping with the characters of any shape. The invention discloses a character detection method in a complex scene. The method comprises the following steps: preprocessing image data, constructing a network framework, pre-training a model, and training the network framework; and generating the character real label, inputting a character image to be detected in a natural scene, carrying out feature extraction, carrying out image judgment and using a character correction module.

Description

technical field [0001] The invention relates to the fields of computer vision and pattern recognition, in particular to a method for character detection in complex natural scene images. Background technique [0002] In recent years, with the rapid development of the Internet, communication technology, and the rise of social networks, multimedia data such as video, audio, and images have grown rapidly. This information makes communication between people very easy. Compared with video data, images require smaller storage capacity and simpler acquisition equipment. Compared with audio and text data, images contain richer and more intuitive information, so images are more in line with daily scene needs. In the face of massive image data, how to efficiently and automatically obtain useful text information from images has become a research hotspot in the field of computer vision in recent years. However, due to the diversity of natural scenes and the variability of shooting angle...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/32G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V30/1478G06V20/63G06V30/153G06V10/267G06V30/10G06N3/045G06F18/214
Inventor 王润民李秀梅张翔宇徐尉翔钱盛友
Owner HUNAN NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products