Intelligent article searching method for visually impaired people

A technology of intelligent search and people, which is applied in the field of daily search for items by the visually impaired. It can solve the problems of long and small target detection needs of the required items, tools that cannot fully meet the basic needs of the visually impaired, and lack of surrounding items, etc., to achieve improved detection. The effect of recognition accuracy, solving the problem of gradient disappearance and overfitting, and enriching semantic feature information

Pending Publication Date: 2021-05-07
HEFEI UNIV OF TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] At present, the research on services for the visually impaired at home and abroad has not received enough attention and attention, some research is still in its infancy, and the services designed for the visually impaired are only simple tools such as guide dogs, blind sticks, and blind paths. And these tools cannot fully meet the basic needs in the life of the visually impaired
In addition, domestic and foreign researchers' research on intelligent services for the visually impaired mainly focuses on guiding the blind using geographic information, navigation based on special sensors, guiding blind based on special visual models perceived by the visually impaired, and vision-based wearable devices. In fields such as guiding the blind and avoiding obstacles, using smart devices to enhance the perception of the visually impaired, etc., most of the products are mainly for assisting travel, and only have the recognition function for some simple specific markers such as traffic lights, roadblocks, etc., and the visually impaired are Passively receive information, lack of technical products to help visually impaired people actively obtain information about surrounding objects
However, the research on item identification and search technology for the life of the visually impaired largely relies on dedicated high-end equipment, and some of them are in the conception and experimental stage, and have not yet entered the practical stage.
In the real life of the visually impaired, there are often a large number of detection needs for small targets due to factors such as the distance between the required items, so that it still cannot meet the actual needs in a large number of application scenarios.

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
  • Intelligent article searching method for visually impaired people
  • Intelligent article searching method for visually impaired people
  • Intelligent article searching method for visually impaired people

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] In this embodiment, an object intelligent search method for visually impaired persons. Such as figure 1 Shown is the process of searching for the items needed by the visually impaired. Input the voice of the visually impaired into the speech recognition model to obtain the text information of the items required by the visually impaired, convert the text image into the target image, and then use the target detection to determine the viewing angle. After the target object is detected, the information such as the target location category is converted from text to voice to inform the visually impaired to help the visually impaired obtain the surrounding target objects. Specific steps are as follows:

[0033] Step 1. Obtain the voice information of the visually impaired and map the voice and text through the voice recognition model of deep learning to obtain the matching keywords; then perform the relationship mapping between the text and the image on the keywords to obtain...

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 an intelligent article searching method for visually impaired people. The method comprises the following steps: 1, extracting keywords of article appeals required by the visually impaired people through a voice recognition model based on deep learning; 2, constructing a target detection model based on deep learning; 3, collecting a life article data set for visually impaired person target detection and training a target detection model; and 4, after searching the target, outputting the position information of the article appealed by the visually impaired person through voice. According to the invention, the voice appeal of the visually impaired person can be identified through artificial intelligence, and the article can be identified and positioned intelligently, so that the visually impaired person can be helped to search the required article.

Description

technical field [0001] The invention relates to a daily search method for visually impaired persons, in particular to an object intelligent search method for visually impaired persons. Background technique [0002] The visually impaired are a disadvantaged group that accounts for a large proportion of society. They face many difficulties in life, such as searching for items. The visually impaired cannot distinguish the category of items and obtain the location information of items, making it difficult to obtain items alone. . In 2017, about 253 million people worldwide suffered from visual impairment, of which 36 million were blind and 217 million had moderate to severe visual impairment. According to the total population of my country in the sixth national census and the results of the second national sampling survey of disabled people, the number of people with single visual disabilities in my country is 12.63 million. If multiple disabled people are included, the prevale...

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): G06K9/32G06K9/34G06N3/04G10L15/26
CPCG10L15/26G06V10/267G06V10/25G06N3/045
Inventor 赵仲秋班名洋田卫东胡健
Owner HEFEI UNIV OF TECH
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