Intelligent blind guide method and apparatus
View Patent ↗The embodiment of the present disclosure discloses an intelligent blind guide method and apparatus, and relates to the field of artificial intelligence, so as to solve the problem that an intelligent blind guide system requires the whole-process intervention of an artificial customer service staff, which results in high work intensity. The intelligent blind guide method includes: obtaining a confidence of an intelligent blind guide according to sensor information, wherein the confidence indicates the reliability of blind guide information generated according to the sensor information without artificial decision; generating the blind guide information according to the sensor information if the confidence is greater than or equal to a preset threshold, and importing artificial decision to generate the blind guide information if the confidence is less than the preset threshold. The embodiment of the present disclosure is applied to a blind guide helmet.
1. An intelligent blind guide method, comprising:
obtaining, through a sensor, sensed information about an environment surrounding an intelligent blind guide;
obtaining a confidence value of the intelligent blind guide by evaluating the sensed information, wherein the confidence value of the intelligent blind guide indicates the reliability of blind guide information generated by processing the sensed information; and
generating blind guide information by processing the sensed information using an artificial intelligence algorithm in a case that the confidence value of the intelligent blind guide is greater than or equal to a threshold, and triggering artificial blind guide in a case that the confidence value of the intelligent blind guide is less than the threshold;
wherein the obtaining the sensed information includes obtaining sensed information on at least two different categories of information about the environment;
wherein the obtaining the confidence value of the intelligence blind guide includes analyzing the sensed information on the at least two different categories of information about the environment; and
wherein the analyzing the sensed information on the at least two different categories of information about the environment includes:
obtaining confidence sub-values on one or more of: positioning accuracy confidence, obstacle avoidance success confidence and object identification confidence; and
fusing at least two of the obtained confidence sub-values to obtain the confidence value of the intelligent blind guide.
2. The method according to claim 1 , wherein the sensed information comprises ultrasonic sensed information and visual sensed information.
3. The method according to claim 1 , wherein the generating the blind guide information by processing the sensed information using the artificial intelligence algorithm includes:
generating one or more of the following information according to the sensed information:
current position coordinates, current travelable direction and object identification tag.
4. The method of claim 1 , wherein the at least two different categories of information include positioning information, obstacle information or object identification information.
5. The method according to claim 1 , wherein the obtaining the positioning accuracy confidence includes obtaining the positioning accuracy confidence according to one or more of the following parameters: texture quality, number of tracking or quality of motion.
6. The method according to claim 1 , wherein the obtaining the obstacle avoidance success confidence includes:
analyzing a size ratio of a passable area in a scene view based on a depth reconstruction result; and
determining the obstacle avoidance success confidence according to the obtained size ratio.
7. The method according to claim 1 , wherein the obtaining the object identification confidence includes:
extracting features in an image;
matching the extracted features with a pre-trained feature library; and
determining the object identification confidence according to a result of the matching.
8. A non-transitory computer storage medium, storing computer executable instructions used with an intelligent blind guide apparatus, the computer executable instructions, when executed by a processor, configures the processor to implement actions comprising:
causing a sensor to obtain sensor information about an environment surrounding an intelligent blind guide;
obtaining a confidence value of the intelligent blind guide by evaluating the sensor information; and
generating blind guide information by processing the sensor information using an artificial intelligence algorithm in a case that the confidence value of the intelligent blind guide is greater than or equal to a threshold, and triggering artificial blind guide in a case that the confidence value of the intelligent blind guide is less than the threshold;
wherein the obtaining the sensor information includes obtaining sensor information on at least two different categories of information about the environment;
wherein the obtaining the confidence value of the intelligence blind guide includes analyzing the sensor information on the at least two different categories of information about the environment; and
wherein the analyzing the sensor information on the at least two different categories of information about the environment includes:
obtaining confidence sub-values on one or more of: positioning accuracy confidence, obstacle avoidance success confidence and object identification confidence; and
fusing at least two of the obtained confidence sub-values to obtain the confidence value of the intelligent blind guide.
9. The computer storage medium according to claim 8 , wherein the sensor information comprises ultrasonic sensor information and visual sensor information.
10. The computer storage medium according to claim 8 , wherein the generating the blind guide information by processing the sensor information using the artificial intelligence algorithm includes:
generating one or more of the following information according to the sensor information: current position coordinates, current travelable direction and object identification tag.
11. The computer storage medium according to claim 8 , wherein the obtaining the confidence sub-value on the positioning accuracy confidence includes obtaining the positioning accuracy confidence according to one or more of the following parameters: texture quality, number of tracking or quality of motion.
12. The computer storage medium according to claim 8 , wherein the obtaining the confidence sub-value on the obstacle avoidance success confidence includes:
analyzing a size ratio of a passable area in a scene view based on a depth reconstruction result; and
determining the obstacle avoidance success confidence according to the obtained size ratio.
13. The computer storage medium according to claim 8 , wherein the obtaining the confidence sub-value on the object identification confidence includes:
extracting features in an image;
matching the extracted features with a pre-trained feature library; and
determining the object identification confidence according to a result of the matching.
14. A system, comprising:
a processor;
a communication interface configured to perform data transmission between the server and an external device; and
a memory storing computer executable instructions, which when executed by the processor, configures the processor to conduct actions including:
causing a sensor of the external device to obtain sensor information about an environment surrounding the external device;
receiving data on the sensor information from the external device;
evaluating the sensor information to obtain an evaluation value; and
generating a feedback information about the environment by processing the sensor information using an artificial intelligence algorithm in a case that the evaluation value is greater than or equal to a threshold, and triggering human intervention in a case that the evaluation value is less than the threshold;
wherein the obtaining the sensor information includes obtaining sensor information on at least two different categories of information about the environment;
wherein the evaluating the sensor information includes analyzing the sensor information on the at least two different categories of information about the environment; and
wherein the analyzing the sensor information on the at least two different categories of information about the environment includes:
obtaining confidence sub-values on one or more of: positioning accuracy confidence, obstacle avoidance success confidence and object identification confidence; and
fusing at least two of the obtained confidence sub-values to obtain the confidence value of the intelligent blind guide.
15. The system of claim 14 , wherein the at least two different categories of information include positioning information, obstacle information or object identification information.
16. The system according to claim 14 , wherein the obtaining the confidence sub-value on the positioning accuracy confidence includes obtaining the positioning accuracy confidence according to one or more of the following parameters: texture quality, number of tracking or quality of motion.
17. The system according to claim 14 , wherein the obtaining the confidence sub-value on the obstacle avoidance success confidence includes:
analyzing a size ratio of a passable area in a scene view based on a depth reconstruction result; and
determining the obstacle avoidance success confidence according to the obtained size ratio.
18. The system according to claim 14 , wherein the obtaining the confidence sub-value on the object identification confidence includes:
extracting features in an image;
matching the extracted features with a pre-trained feature library; and
determining the object identification confidence according to a result of the matching.