Intelligent voice blind guiding system based on artificial intelligence (AI) open platform and application method
An open platform and intelligent voice technology, applied in radio wave measurement systems, satellite radio beacon positioning systems, and appliances to help people move around, can solve problems such as insufficient prompts and blind people's inability to know accurately, and achieve a strong satellite search ability Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0078] Embodiment 1, Universal Object Recognition:
[0079] Figure 1~3 The shown a kind of intelligent speech guiding system based on AI open platform includes interactive box 1 and control and voice output box 2; the upper end of interactive box 1 is equipped with camera 3, and the side is provided with a plurality of interactive box USB interfaces 4, wherein A microphone assembly is connected to an interactive box USB interface through a USB plug 5;
[0080] The interior of the interactive box 1 is provided with a raspberry pie main control board; the raspberry pie main control board is equipped with onboard bluetooth and onboard WIFI, and the raspberry pie main control board is electrically connected with the camera and the USB interface of the interactive box; The camera is connected to Raspberry CSICamera interface.
[0081] The box body of the interactive box is provided with a power indicator light 14 . According to the actual application, a battery is installed ins...
Embodiment 2
[0095] Embodiment 2. Customized object recognition:
[0096] The difference between this embodiment and Embodiment 1 is that the usage method of this embodiment can customize and customize recognition targets, such as various information signs such as bus stops and signs, and return results when specified objects appear in the image. The function is realized by using Baidu EasyDL customized training and service platform. Users customize the training model according to their own needs. The recognition object is determined by the user. After the model training is completed, apply for release. After the release is successful, it can be called. The program passes the image to be recognized and the model Match and compare to get the recognition result. The calling steps are similar to the general object recognition steps, the difference is that the address of the calling interface is different. If the specified object is recognized in the image, the recognition accuracy and the po...
Embodiment 3
[0101] Embodiment three, identification of traffic lights:
[0102] This embodiment first judges whether there is a traffic light in the image through customized image recognition. If the recognition judgment result is that a traffic light is found in front, then the traffic light is cut out according to the position coordinates of the traffic light in the image in the data return result, saved separately, and then judged by Opencv Whether it is a red light or a green light, it guides the blind to pass. Such as Figure 10 As shown, the difference from Embodiment 1 is as follows:
[0103] a. Recognition command; the microphone receives the voice command, and preprocesses the voice command through the main control board of the Raspberry Pi. Text, and match with the command model, if the match is successful, the command is determined to be valid;
[0104] b. Judging that the instruction type is a traffic light identification instruction;
[0105] c. Analysis and processing in...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com