Voice prompting method of safety warning
AssigneeCompal Electronics, Inc.
InventorsShih-Hao Yang, Hui-Chen Chen, Chieh-Yuan Huang
Life of this patent
Estimating the term…
Abstract
A voice prompting method includes steps of: providing a wearable electronic device, obtaining an image and performing an image processing on the image, detecting and identifying a plurality of objects in the image, classifying and judging danger levels of the objects to generate a judgment result, generating a safety warning message corresponding to each of the objects presently having the danger level higher than a specific danger level according to the judgment result, and transforming the safety warning message into a voice form and prompting with the voice form of the safety warning message. Therefore, it can help a user avoid danger and effectively enhance the safety of walking of the user.
Claims8
A voice prompting method of safety warning, comprising steps of:
- (a) providing a wearable electronic device;
- (b) obtaining an image and performing an image processing on the image;
- (c) detecting and identifying a plurality of objects in the image;
- (d) classifying and judging danger levels of the objects to generate a judgment result;
- (e) generating a safety warning message corresponding to each of the objects presently having the danger level higher than a specific danger level according to the judgment result;
- and
- (f) transforming the safety warning message into a voice form and prompting with the voice form of the safety warning message,
- wherein the voice prompting method of safety warning further comprises a step, between the step (c) and the step (d), of:
- (g) obtaining a speed of a user of the wearable electronic device and speeds of the objects according to the image,
- wherein the step (d) comprises steps of:
- (d1) classifying the danger levels of the objects to the user as a first classification;
- and
- (d2) classifying the speeds of the objects as a second classification and judging the danger levels of the objects according to the first classification and the second classification to generate the judgment result.
The voice prompting method of safety warning according to claim wherein the step (e) comprises steps of:
- (e1) using the judgment result, the speed of the user and the speeds of the objects as input of a recurrent neural network of an image captioning unit;
- and
- (e2) allowing the image captioning unit to output a paragraph of text matched with the image.
Dependencies are inferred from the claim text.