CN109381329A - A kind of intelligent blind-guiding helmet and its operation method - Google Patents

Abstract

The invention discloses an intelligent blind guidance helmet and an operation method thereof, comprising: a helmet; a binocular camera, a processing module, a communication module, a positioning module and a voice module are installed on the helmet. The binocular camera is used to transmit the collected image information to the processing module for analysis, the processing module analyzes the image information, and issues instructions to the voice module, and the voice module feeds back to the user according to the analysis result. The invention has the advantages that the portability of the blind person is improved, and the practicability of the device is enhanced; the manufacture is simple, the manufacturing cost of the device is reduced, the threshold for use is lowered, and the popularity of the device is improved.

Description

An intelligent blind guide helmet and its operation method

technical field

The invention relates to the technical field of blind guidance, in particular to an intelligent blind guidance helmet and a method for operating the same for providing travel convenience for visually impaired people.

Background technique

At present, the number of blind people in the world continues to increase, and the number of blind people in developing countries accounts for the vast majority of the world's total. As a developing country, my country has a huge number of blind people on the basis of a huge population base, and there are not many blind guide devices on the market at present. At present, blind guidance technologies mainly include blind guiding robots, GPS blind guiding technology, and sensor-based whole-body wearable blind guiding technology.

The prior art relevant to the present invention is as follows:

Prior art 1: Chinese invention patent, patent name: An intelligent blind guidance robot, application number: CN201810844525.1.

Disadvantages of prior art one:

1. Unable to travel freely, and accidents are prone to occur when faced with various road conditions; 2. High cost; 3. Low travel portability.

Prior art 2: Chinese utility model patent, patent name: GPS guide for the blind, application number: CN201020231887.2.

Disadvantages of the existing technology two:

1. It is greatly affected by the weather; 2. The indoor blindness guidance effect is poor; 3. The precision is low.

SUMMARY OF THE INVENTION

Aiming at the defects of the prior art, the present invention provides an intelligent blind guide helmet and an operation method thereof, which can effectively solve the problems existing in the prior art.

In order to realize the above purpose of the invention, the technical scheme adopted by the present invention is as follows:

An intelligent blind guidance helmet includes: a helmet 1 ; a binocular camera 2 , a processing module 3 , a communication module 4 , a storage module 5 , a positioning module 6 , a voice module 7 and a battery 8 are installed on the helmet 1 .

The binocular camera 2 is fixed on the front of the helmet 1; the communication module 4 and the positioning module 6 are fixed on the top of the helmet 1; the two voice modules 7 are located on both sides of the helmet 1 corresponding to the ears; left and right.

The binocular camera 2 adopts a horizontal fixed-baseline camera, the data interface is USB3.0, the interface protocol is UVC protocol, the timing of the two cameras is absolutely synchronized, and the collected image information is transmitted to the processing module 3 for analysis. According to the processing results, the voice module 7 feedback to users.

The processing module 3 identifies obstacles and their distances through the Faster RCNN convolutional neural network algorithm and the binocular stereo matching algorithm, and issues an instruction to the voice module 7 .

The positioning module 6 transmits the positioning information to the processing module 3 .

The storage module 5 is connected to the processing module 3, and is used for storing the image information and position information acquired by the device.

The communication module 4 transmits the real-time location information to the cloud, and family members can view the device location in real time; at the same time, this module transmits the image information in the storage module 6 to the cloud for continuing to train the FasterRCNN convolutional neural network.

The battery 8 provides power for the binocular camera 2 , the processing module 3 , the communication module 4 , the storage module 5 , the positioning module 6 and the voice module 7 .

An operation method of an intelligent blind guide helmet. After the user wears the device, the binocular camera 2 will collect the surrounding environment image information and deliver it to the processing module 3. The surrounding road conditions are identified by the FasterRCNN algorithm and the binocular stereo matching algorithm, and the voice Module 7 provides voice prompts to the user. The communication module 4 transmits the real-time location information of the positioning module 6 to the cloud, and the family members can view the device location in real time.

Compared with the prior art, the present invention has the advantages that: by using two cameras to simulate the way normal human eyes collect information, and by processing the image information collected by the two cameras to identify the surrounding environment, it can greatly It simulates the daily behavior of normal people. Deep learning algorithms temporarily have absolute advantages in the field of image recognition. Compared with traditional image recognition algorithms, the accuracy of image recognition is greatly improved. Using fewer hardware modules to achieve more functions, through the combination of various algorithms, the composition of the hardware modules is further simplified, the weight of the device is reduced, the portability of the blind person is improved, and the practicability of the device is enhanced. On the other hand, by simplifying the composition of the equipment, the manufacturing cost of the equipment is reduced, thereby lowering the threshold for use and increasing the possibility of popularization of the equipment.

Instruction drawings

1 is a schematic front view of an intelligent blind guide helmet according to an embodiment of the present invention;

2 is a left side view of an intelligent blind guide helmet according to an embodiment of the present invention;

3 is a right side view of an intelligent blind guide helmet according to an embodiment of the present invention;

4 is a schematic structural diagram of a circuit of an intelligent blind guidance helmet according to an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail below with reference to the accompanying drawings and examples.

As shown in Figures 1 to 4, an intelligent blind guidance helmet, comprising: a helmet 1;

A binocular camera 2 , a processing module 3 , a communication module 4 , a storage module 5 , a positioning module 6 , a voice module 7 and a battery 8 are installed on the helmet 1 .

The binocular camera 2 is fixed on the front of the helmet 1; the communication module 4 and the positioning module 6 are fixed on the top of the helmet 1; the two voice modules 7 are located on both sides of the helmet 1 corresponding to the ears; left and right.

The binocular camera 2 adopts a horizontal fixed-baseline camera, the data interface is USB3.0, the interface protocol is UVC protocol, the timing of the two cameras is absolutely synchronized, and the collected image information is transmitted to the processing module 3 for analysis. According to the processing results, the voice module 7 feedback to users.

The processing module 3 adopts Raspberry PI 3rd generation B+ type as the main body. The obstacles and their distances are identified through the Faster RCNN convolutional neural network algorithm and the binocular stereo matching algorithm, and an instruction is issued to the voice module 7, which is BY-F610.

The positioning module 6 uses the S1216 positioning module to achieve positioning.

The storage module 5 is a memory, and the model is K9F1G08U0E, which is used to store the image information obtained by the device.

The communication module 4 uses the USR-LTE-7s44gDTU module to transmit the real-time location information to the cloud, so that family members can view the device location in real time; at the same time, this module transmits the image information in the storage module 6 to the cloud, which is used to continue training the FasterRCNN convolutional neural network network.

The battery 8 provides power for the binocular camera 2 , the processing module 3 , the communication module 4 , the storage module 5 , the positioning module 6 and the voice module 7 .

After the user wears the device, the binocular camera 2 will collect the surrounding environmental image information and deliver it to the processing module 3, identify the surrounding road conditions through the FasterRCNN algorithm and the binocular stereo matching algorithm, and the voice module 7 will give the user a voice prompt. The communication module 4 transmits the real-time location information of the positioning module 6 to the cloud, and the family members can view the device location in real time.

Those of ordinary skill in the art will appreciate that the embodiments described herein are intended to help readers understand the implementation method of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations without departing from the essence of the present invention according to the technical teaching disclosed in the present invention, and these modifications and combinations still fall within the protection scope of the present invention.

Claims (2)

1. a kind of intelligent blind-guiding helmet characterized by comprising the helmet (1)；Binocular camera (2), place are installed on the helmet (1) Manage module (3), communication module (4), memory module (5), locating module (6), voice module (7) and battery (8)；

Binocular camera (2) is fixed on the helmet (1) front；Communication module (4) and locating module (6) are fixed at the top of the helmet (1)； Two side positions that two voice modules (7) are located at the helmet (1) correspond to ear；Memory module (5) is located at processing module (3) The left and right sides of the helmet (1)；

Binocular camera (2) determines baseline camera, data-interface USB3.0 using level, and interface protocol is UVC agreement, and two take the photograph As head timing absolute synchronization, acquired image information is transmitted to processing module (3) and is analyzed, according to processing result by language Sound module (7) feeds back to user；

Processing module (3) identifies barrier by Faster RCNN convolutional neural networks algorithm and Binocular Stereo Matching Algorithm And its distance, and voice module (7) are issued and are instructed, user is prompted by voice module (7)；

Locating module (6), which is realized, to be positioned and location information is transmitted to processing module (3)；

Memory module (5) connects processing module (3), for storing the image information and location information of equipment acquisition；

Real-time location information is transferred to cloud by communication module (4), and household can real time inspection device location；This module will simultaneously Image information in memory module (6) is transmitted to cloud, trains FasterRCNN convolutional neural networks for continuing；

Battery (8) be binocular camera (2), processing module (3), communication module (4), memory module (5), locating module (6) and Voice module (7) provides electric energy.

2. a kind of operation method of intelligent blind-guiding helmet according to claim 1, it is characterised in that: user's wearable device Afterwards, binocular camera (2) will collect the ambient image information of surrounding, consigns to processing module (3), is calculated by FasterRCNN Method and Binocular Stereo Matching Algorithm identify surrounding road condition, carry out voice prompting to user by voice module (7)；Communication module (4) by locating module (6), location information is transferred to cloud in real time, and household can real time inspection device location.