Book point reading method and system based on deep learning

A deep learning and book point technology, applied in the field of book point reading, can solve the problems of cumbersome operation, fixed image and text recognition and poor accuracy, so as to improve accuracy, simplify point reading detection and recognition methods, and avoid point reading errors. Effect
CN111353501APending Publication Date: 2020-06-30暗物智能科技(广州)有限公司
13 Cites 8 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
暗物智能科技(广州)有限公司
Publication Date
2020-06-30

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a book point reading method and system based on deep learning, and the method comprises the steps: enabling an obtained finger point reading image photographed by an image collection device to be inputted into a finger detection model and a text detection model at the same time, and respectively detecting the fingertip position and all text regions in the image; conductingaffine transformation and cutting on the character area according to the detected fingertip position and the detected character area; inputting the cut character area into a character recognition model to recognize character information; carrying out sentence segmentation or segmentation processing on the text information by utilizing punctuation or segmentation character information; and outputting individual characters, words, sentences or text segments by voice according to a preset click-to-read demand. Finger point reading images are shot through the image acquisition device, so that theexpansibility of point reading content is improved; the finger detection model, the text detection model and the text recognition model are trained, point reading of single Chinese characters, words,sentences and text segments is achieved, a traditional point reading detection and recognition method is simplified, and the accuracy of character detection and recognition is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical field

[0001] The invention relates to the technical field of book point reading, in particular to a book point reading method and system based on deep learning. Background technique

[0002] In recent years, with the rapid development of computer vision and deep learning, technologies such as taking photos for literacy, taking photos to search for questions, and reading supplementary studies have been widely used in intelligent education. Among them, the point-reading machine is a popular learning aid tool. It uses finger detection and text detection functions to perceive the location of the content pointed to by the user’s finger, and the text information content of the location area can be identified based on the location. So as to complete the process of human-computer interaction. However, the dot reading machine in the prior art has the following disadvantages: first, it is necessary to use gestures to draw a rectangle on the text area to determine the dot reading...

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