Intelligent blind guiding instrument

Abstract

The application relates to the technical field of blind guiding, and particularly discloses an intelligent blind guiding instrument, which comprises a host computer internally provided with an MCU processor and a power module; a head-mounted device composed of a ring frame and a camera, a microphone and a loudspeaker mounted on the ring frame; and a blind guiding stick in Bluetooth connection with the MCU processor, wherein the MCU processor is in communication connection with a background server or a cloud server and loads an image database; the MCU processor is used for receiving image information sent by the camera and identifying and judging the image category, so as to control the blind guiding instrument to enter a voice navigation mode or a thing recognition mode; the MCU processor controls the blind guiding instrument to enter a voice control mode when detecting that the microphone sends predetermined voice content, so as to start or stop the blind guiding instrument or play predetermined audio. The above-mentioned intelligent blind guiding instrument combines the blind guiding stick with the head-mounted device, collects the images around the blind guiding stick to position a position to be guided, improves the navigation accuracy, has multiple functions, realizes one machine with multiple functions, and reduces the blind guiding cost of users.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of guiding the blind, in particular to an intelligent blind guiding instrument with precise guiding and multiple functions. BACKGROUND

[0002] In the daily life of the blind, the existence of visual impairment makes it difficult for them to see things, and brings many inconveniences to daily life. In particular, the increase in the number of vehicles and the continuous updating of roads pose a potential safety threat to the travel of the blind, which greatly weakens the willingness of the blind to travel and to a certain extent affects the quality of life of the blind.

[0003] At present, the blind mainly use a blind guiding stick equipped with a radar detector to identify the guiding route. Since the area around the blind is large when the blind travels, the radar detector is difficult to distinguish the forward direction of the blind, so it is difficult to effectively prompt the road conditions in front of the blind. The reliability of the radar blind guiding stick is poor. Secondly, the traditional blind guiding stick is only used for guiding route identification, and has few additional functions. The blind needs to use the blind guiding stick in combination with other blind guiding devices, thereby increasing the blind guiding cost of the blind. SUMMARY

[0004] Therefore, it is necessary to provide an intelligent blind guiding instrument with precise guiding and multiple functions in view of the technical problems of poor reliability and single function.

[0005] An intelligent blind guiding instrument, comprising:

[0006] A host configured to be arranged on clothes or accessories, wherein the host is internally provided with an MCU processor for data operation and signal response processing, and a power module electrically connected with the MCU processor and configured to supply power for the blind guiding instrument; the MCU processor is in communication connection with a background server or a cloud server, and loads an image database sent by the background server or the cloud server, wherein the image database comprises different object image data, different scene image data, an electronic map and dangerous goods image data;

[0007] A head-mounted device comprising a ring frame configured to be worn on the head or neck of a human body, a camera configured to capture images, a microphone configured to capture voice information, and a loudspeaker configured to output voice prompts, which are installed on the ring frame and electrically connected with the MCU processor through connecting lines, respectively; and

[0008] A blind guiding stick, wherein the blind guiding stick is in Bluetooth connection with the MCU processor and is configured to identify and position the camera;

[0009] The MCU processor is used for receiving image information sent by the camera, and identifying and judging the image information category to control the blind guiding instrument to enter a voice navigation mode or an object recognition mode; when the camera collects image information of a preset position on the blind guiding stick, the MCU processor controls the speaker to be turned on to enter the voice navigation mode, and the MCU processor compares image information around the blind guiding stick collected by the camera in real time with electronic map and dangerous object image data, and outputs guiding prompt information and dangerous prompt information; when the camera collects a preset gesture action, the MCU processor controls the speaker to be turned on to enter the object recognition mode; and the MCU processor also controls the blind guiding instrument to enter a voice control mode when detecting that the microphone sends a predetermined voice content, so as to start or stop the blind guiding instrument or play predetermined audio content.

[0010] In one of the embodiments, a display screen electrically connected to the MCU processor is arranged on the shell of the host, and used for displaying personal information and device information of the host.

[0011] In one of the embodiments, the frame is in a structure of glasses, a headband, a headband or a hat.

[0012] In one of the embodiments, a radar detection head electrically connected to the MCU processor is further arranged on the frame, and when the blind guiding instrument is in the voice navigation mode, the MCU processor fits image information sent by the camera and radar signal data sent by the radar detection head to correct road condition information.

[0013] In one of the embodiments, when detecting a moving object, a preset dangerous image, a preset terminal position or a preset distance, the speaker outputs a predetermined voice signal correspondingly when the blind guiding instrument is in the voice navigation mode.

[0014] In one of the embodiments, the object recognition mode includes a local scene recognition sub-mode switched when detecting a first gesture image, an online scene recognition sub-mode switched when detecting a second gesture image, and an object recognition sub-mode switched when detecting a third gesture image.

[0015] In one of the embodiments, when receiving a place marking voice instruction, the microphone sends a request to the MCU processor, and the MCU processor marks a current position in an electronic map.

[0016] In one of the embodiments, the MCU processor further plans a navigation path in the electronic map according to a blind guiding route instruction sent by the microphone.

[0017] In one of the embodiments, a plurality of hand controls for controlling starting and stopping of the blind guiding instrument, mode switching and volume and speed adjustment are further arranged on the frame.

[0018] In one of the embodiments, the MCU processor further interacts with the third-party APP for authentication, and switches to the remote control mode when receiving the predetermined gesture, voice or key instruction, and when the blind guiding instrument switches to the remote control mode, the user of the third-party APP receives the image information sent by the camera and calls the speaker to send voice prompt information.

[0019] The intelligent blind guiding instrument of the present application separates the host and the head-mounted device, reduces the head load of the user during use, sets the blind guiding stick, and turns on the navigation function of the blind guiding instrument when the camera detects the image of the predetermined part of the blind guiding stick, and the MCU processor processes the image around the blind guiding stick collected by the camera in real time to output the guiding prompt information, so that the user can accurately obtain the road conditions and navigation information of the direction in front of him, and the reliability of the navigation function of the blind guiding instrument is improved. By loading the image database, the object recognition mode can be entered on the basis of recognizing the gesture, or the voice control mode can be entered to play audio when receiving the voice instruction, so as to provide multiple ways for the user to recognize objects and obtain information, realize one machine with multiple functions, expand the application scenarios of the blind guiding instrument, and reduce the blind guiding cost of the user, thereby providing convenience for the life of the blind people. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 Fig. 1 is a structural schematic diagram of the intelligent blind guiding instrument in one embodiment of the present application;

[0021] Figure 2 Fig. 2 is a module structure schematic diagram of the intelligent blind guiding instrument in the embodiment shown in Fig. 1. Figure 1 Specifically, the intelligent blind guiding instrument of the present application comprises a host and a head-mounted device. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below. In the following description, a large number of specific details are set forth in order to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the scope of the present application, so the present application is not limited to the specific embodiments disclosed below.

[0023] Please refer to Figure 1 and Figure 2The application provides a precise guide and multifunctional intelligent blind guide instrument 10, which is used for providing road navigation, object recognition learning and audio playing for visually impaired people (blind people), and compared with traditional blind guide glasses or other blind guide devices, the intelligent blind guide instrument 10 has more functions, and a user can meet various needs by using only one device, so that the problems of increased blind guide cost and inconvenience caused by carrying multiple blind guide devices are avoided, and more convenience is provided for the blind people. The intelligent blind guide instrument 10 comprises a host 100, a head-mounted device 200 which is electrically connected with the host 100 through a connecting line, and a blind guide stick 300 which is separately arranged with the host 100 and the head-mounted device 200, wherein the host 100 is arranged on clothes or accessories when used, for example, the host 100 is stored in a clothes pocket of a user, or a hook is arranged on the shell of the host 100 so as to hang the host 100 on a belt or other accessories of the user, so that the host 100 is worn. The host 100 is internally provided with an MCU processor 110 for data operation and signal response processing, and a power module 120 which is electrically connected with the MCU processor 110 and is used for power supply of the blind guide instrument; the MCU processor 110 is in communication connection with a background server or a cloud server, and loads an image database sent by the background server or the cloud server, wherein the image database comprises different object image data, different scene image data, an electronic map and dangerous goods image data. The power module 120 comprises a power storage element and a charging element, preferably, the charging element is a contact charger which is arranged through the shell of the host 100, and when the contact charger is connected with an external power supply, the power of the external power supply is transmitted to the power storage element, so that the power storage element provides sufficient working power for the blind guide instrument. In the embodiment, the capacity of the power storage element is as high as 10000 mAh, so that the blind guide instrument can be on standby for at least two days under the condition of continuous working, the standby time of the blind guide instrument is prolonged, the inconvenience caused by power failure of the blind guide instrument to a user is avoided, and the reliability of the blind guide instrument is improved. It should be noted that, in order to prolong the service life of the blind guide instrument and further improve the reliability of the working, a waterproof sleeve is arranged on the shell of the host 100, the waterproof sleeve can be made of rubber material, and the thickness of the waterproof sleeve is greater than 3 mm, so that the waterproof and dustproof functions are provided for the host 100, dust impurities or water is prevented from entering the inside of the host 100, and the damping function is provided for the host 100, so that the connection between components in the host 100 is not loose under the condition of external force impact, and the shock resistance of the host 100 and the reliability of the equipment use are improved.

[0024] Further, in an embodiment, the shell of the host 100 is provided with a display screen 130 electrically connected with the MCU processor 110, for displaying the personal information of the host and the equipment information. Preferably, the personal information of the host and the equipment information are stored in a two-dimensional code, the personal information including the name, age, emergency contact phone, address, etc. of the host, and the equipment information including the factory model and parameter information of the guide blind instrument, the transaction information of the guide blind instrument, etc., so that when any emergency occurs or the guide blind instrument is lost during the use of the guide blind instrument by the host or other users, other personnel can obtain the personal information and the equipment information of the host by scanning and identifying the two-dimensional code, so as to provide assistance for the user or find the owner, in addition, the family members or other related technical personnel of the user can also obtain the equipment parameters by scanning the two-dimensional code, so as to evaluate the performance of the equipment. The host 100 is also provided with a plurality of buttons or knobs for controlling the start and stop and mode switching of the guide blind instrument, so that the user can manually control the guide blind instrument by operating the buttons or knobs.

[0025] The head-mounted device 200 includes a ring frame 210 for wearing on the head or neck of a human body, a camera 220 for image acquisition, a microphone 230 for voice information acquisition, and a loudspeaker 240 for outputting voice prompts, which are installed on the ring frame 210 and are electrically connected with the MCU processor 110 through connecting wires respectively. The ring frame 210 is mainly used as a mounting carrier for the camera 220, the microphone 230, and the loudspeaker 240, by wearing the ring frame 210 on the head or neck of the human body, the horizontal height of the camera 220 is lifted, which is conducive to the camera 220 to collect clear and comprehensive images, expands the area range of the images, and further improves the reliability of the images collected by the camera 220. In an embodiment, the ring frame 210 is in the form of glasses, a headband, a headband, or a hat, that is, the ring frame 210 can be designed as a decoration such as glasses, a headband, a headband, and a hat, so as to weaken the sense of strangeness caused by wearing the ring frame 210.

[0026] Further, the ring frame 210 is also provided with a radar probe 250 electrically connected with the MCU processor 110. When the blind guiding instrument is in the voice navigation mode, the MCU processor 110 fits the image information sent by the camera 220 and the radar signal data sent by the radar probe 250 to correct the road condition information. Specifically, in the process of using the blind guiding instrument, the image information collected by the camera 220 is mostly planar image information, and it is difficult to identify the thickness, depth, distance from the user and other information of the collected image object. By using the camera 220 and the radar probe 250 together, the MCU processor 110 receives the planar image sent by the camera 220, and marks each point in the planar image according to the data fed back by the radar probe 250. It can also be understood that the MCU processor 110 constructs an environment three-dimensional model according to the radar data after receiving the data fed back by the radar probe 250, and then imports the planar image photographed by the camera 220 into the environment three-dimensional model, so as to mark each point in the planar image, more accurately identify the environment image in the photographing range of the camera 220, and improve the reliability of the blind guiding instrument.

[0027] The microphone 230 is used to input voice instructions, such as marking a place by voice, issuing a navigation route request, and voice controlling the work of the blind guiding instrument. The loudspeaker 240 is used to receive the signal sent by the MCU processor 110 and play it in the form of sound to the user and the people around the user to prompt the relevant people. The loudspeaker 240 can be a conventional sound conduction loudspeaker 240, or a bone conduction loudspeaker 240, which can be designed as an in-ear structure or an external non-in-ear structure, and the specific design is based on the user's use habit. In an embodiment, the ring frame 210 is also provided with a plurality of hand controls 211 for controlling the start and stop of the blind guiding instrument, mode switching, and volume and speed adjustment. Each hand control 211 can be regarded as a physical button. In this way, the user can not only control the work of the blind guiding instrument through the buttons or knobs on the main machine 100, but also control the start and stop of the blind guiding instrument and adjust the working parameters of the blind guiding instrument through the hand controls 211 on the ring frame 210, thereby providing multiple control operation methods for the blind guiding instrument.

[0028] The blind guiding stick 300 is matched with the host 100, and is used for positioning the blind guiding point and determining the blind guiding direction, and triggering the opening of the blind guiding function of the blind guiding instrument. The blind guiding stick 300 is connected with the MCU processor 110 through Bluetooth, and is used for the positioning of the camera 220. In the working process of the blind guiding instrument, the MCU processor 110 is used for receiving the image information sent by the camera 220, and identifying and judging the image information category, so as to control the blind guiding instrument to enter the voice navigation mode or the object recognition mode. Specifically, when the camera 220 collects the image information of the preset position of the blind guiding stick 300, for example, the camera 220 collects the image of the handle of the blind guiding stick 300 or collects the image of the end part of the blind guiding stick 300 contacting the ground, the MCU processor 110 judges that the blind guiding stick 300 enters the shooting area of the camera 220 and controls the speaker 240 to open to enter the voice navigation mode. In this case, the camera collects the image around the blind guiding stick 300, that is, collects the image of the forward direction of the user, and then the MCU processor 110 compares the image information around the blind guiding stick 300 collected by the camera 220 in real time with the electronic map and the dangerous object image data, and outputs the guiding prompt information and the dangerous prompt information. It should be noted that in this case, the user does not need to input the destination instruction to the MCU processor 110, that is, the MCU processor 110 does not need to request and plan the active path from the local server, and the MCU processor 110 can obtain the image data around it under the cooperation of the camera and the radar probe 250, and output the voice in real time to prompt the user the things around him, and make warning and reminding. The reminding content can include the straightness, length, whether there is a pit, facilities around the road section, whether there is a moving object and the speed of the moving object, etc. of the road section in front of the user, so as to ensure the safety of the user when walking.

[0029] It should be noted that in an embodiment, the blind guiding stick 300 is provided with a signal transceiver module 310 connected with the MCU processor 110 through Bluetooth, and an operation key 320 electrically connected with the signal transceiver module 310. In this way, the user can control the operation key 320 on the blind guiding stick 300 to send a control signal to the MCU processor 110, so as to start or stop the blind guiding instrument, switch the working mode of the blind guiding instrument or adjust the working parameters of the blind guiding instrument. Further, the inside of the blind guiding stick 300 is also provided with a vibration module 330 and a battery 340 for supplying power to the vibration module 330. The vibration module 330 can be a vibration motor, which is electrically connected with the signal transceiver module 310. When the MCU processor 110 judges that the image collected by the camera 220 includes a dangerous object, the MCU processor 110 controls the vibration motor to work through the signal transceiver module 310, so as to prompt the user of the danger.

[0030] In an embodiment, the blind guiding device outputs predetermined voice signals through the speaker 240 in the voice navigation mode when detecting a moving object, a preset dangerous image, a preset end position or a preset route. For example, the speaker 240 simulates a dog barking or other prompt sound to remind the user that the blind guiding function is turned on when the camera 220 captures an image of the blind guiding stick 300 or the button is started or turned off. When the MCU processor 110 receives the image captured by the camera 220 and determines that the user is in a safe state, the speaker 240 emits an interval and low-pitched dog panting sound. It should be noted that before the first use of the blind guiding device or the voice navigation mode is turned on, the host 100 needs to input instructions and define the parameters of the MCU processor 110. In this embodiment, the "safe area" is defined as an area 2 meters in front of the blind guiding stick 300, 1.2 meters wide on the left and right, and 30 centimeters high from the bottom of the blind guiding stick 300. When the safe area is identified as a sidewalk, the MCU processor 110 controls the speaker 240 to emit a dog panting sound, and the road surface state is prompted in the form of voice broadcast. When the MCU processor 110 detects that there is a moving object in the safe area, such as a moving person, a vehicle or an animal, the speaker 240 immediately emits a dog barking sound. The MCU processor 110 controls the frequency of the sound output by the speaker 240 according to the distance between the moving object and the user as fed back by the radar probe 250. Specifically, when the moving object approaches the user, the dog barking sound emitted by the speaker 240 becomes more urgent, and when the moving object moves away from the user, the dog barking sound emitted by the speaker 240 becomes slower. When the MCU processor 110 detects that there is a dangerous object in the safe area, such as a step, a staircase, an elevator, a column, a railing, a pit, water, fire, a wall, a vehicle, a linear object, a cliff, glass, a zebra crossing, a vertical elevator and other obstacles, the MCU processor 110 controls the speaker 240 to output the name information of the dangerous object when identifying and judging the above dangerous objects, so as to prompt the user of the dangerous situation.

[0031] In an embodiment, the MCU processor 110 further has a voice recognition module. When the blind guiding instrument enters the voice navigation mode, the MCU processor 110 further receives the sound signal transmitted by the microphone 230, and sets the navigation parameters of the blind guiding instrument according to the sound signal. For example, the user can issue a voice instruction of "going to xx place". After the microphone 230 collects the voice instruction, the voice instruction is sent to the MCU processor 110. The voice recognition module recognizes the voice signal and sends the recognition result to the MCU processor 110. The MCU processor 110 plans the optimal navigation path in the electronic map according to the recognition result. When the blind guiding instrument enters the voice navigation mode, the MCU processor 110 substitutes the planar image transmitted by the camera 220 and the simulation result of the radar data transmitted by the radar probe 250 into the navigation path for comparison, and broadcasts the change points in the navigation path, such as the turning, reversing, position condition, orientation and the like. At the same time, the MCU processor 110 also controls the loudspeaker 240 to broadcast the parameters such as the distance from the end point and the current walking distance after walking a predetermined distance, so that the user can know the navigation condition.

[0032] When the camera 220 captures the preset gesture action, the MCU processor 110 controls the speaker 240 to open to enter the thing identification mode. Further, the thing identification mode includes a local scene identification sub-mode switched when the first gesture image is detected, an online scene identification sub-mode switched when the second gesture image is detected, and an article identification sub-mode switched when the third gesture image is detected. For example, when the camera 220 captures the user's five fingers into a fist, it is determined as the first gesture image, in which case the blind guide instrument switches to the local scene identification sub-mode, and when the MCU processor 110 receives the image transmitted by the camera 220, the image is substituted into the image database to match and identify different scene image data, when the captured image matches any scene image data in the image database, the MCU processor 110 controls the speaker 240 to output the name of the matched scene, so that the user knows the current environment. When the camera 220 captures the user's five fingers spread out, it is determined as the second gesture image, in which case the blind guide instrument switches to the online scene identification sub-mode, the MCU processor 110 is connected to the third party platform through the cloud server, and is authorized by the third party platform, for example, an APP that needs ID login and information matching, so that the blind guide instrument is connected with the APP end, the user can provide volunteer service or select paid service provided by the marketing personnel of the APP end, the APP end user can obtain the permission of the speaker 240, the camera 220 and the microphone 230, so that the blind guide instrument user can point out, ask questions by voice, etc. In the case of microphone 230 collecting voice instructions or camera 220 collecting image instructions, the APP end user obtains the instructions, and calls the speaker 240 to output voice answers to the blind guide instrument user, thereby realizing online interaction between the APP end user and the blind guide instrument user. When the camera 220 captures the user's "scissors hand" gesture, it is determined as the third gesture image, in which case the blind guide instrument switches to the article identification sub-mode, and when the camera 220 captures the user's index finger pointing at the article, the article contacted with the index finger is identified, searched and matched in the image database by the MCU processor 110, until the matched article information is output, and the article name is played by the speaker 240, so that the user can touch the article to deepen the understanding of the article under the condition of knowing the article name. It should be noted that the above gestures are only used to illustrate the starting conditions of the first gesture image, the second gesture image and the third gesture image, and in actual use, the first gesture image, the second gesture image and the third gesture image can be customized before the blind guide instrument is used for the first time to meet the user's habits.

[0033] The MCU processor 110 also controls the blind guide instrument to enter the voice control mode when detecting that the microphone 230 emits the predetermined voice content, to start or stop the blind guide instrument or play the predetermined audio content. In an embodiment, the microphone 230 sends a request to the MCU processor 110 when receiving the place marking voice instruction, and the MCU processor 110 marks the current position in the electronic map, so that when the user passes through the current place next time, the MCU processor 110 will read the marked position data and control the speaker 240 to voice broadcast, thereby improving the accuracy of the electronic map of the blind guide instrument. In an embodiment, the MCU processor 110 also plans the navigation path in the electronic map according to the blind guide route instruction sent by the microphone 230. Further, in an embodiment, the MCU processor also interacts with the third-party APP for authentication, and switches to the remote control mode when receiving the predetermined gesture, voice or key instruction, when the blind guide instrument switches to the remote control mode, the third-party APP user receives the image information sent by the camera 220 and calls the speaker 240 to send voice prompt information. For example, when the user inputs the "start the remote control mode" voice instruction to the blind guide instrument, the MCU processor 110 receives the voice instruction and provides the permission to the third-party APP, which is the third-party user end bound with the blind guide instrument device ID, and the APP includes the functions of family binding, remote voice navigation, non-disturbance link remote video, position positioning, activity route monitoring, marking editing, blind guide route setting, volunteer voice service, music and e-book playing, voice blogger, etc. In this way, the user's family members can remotely monitor the user's whereabouts through the third-party client, and can remotely talk to the user or remotely guide the user or perform object recognition learning, etc. When the user's activity reaches the predetermined position marked on the electronic map, the MCU processor 110 also sends the user's position information to the third-party APP, so that the user's family members can know the user's whereabouts, thereby improving the safety of the user's travel; the user can also activate the APP according to his own needs through voice, gesture or key form, establish a connection with the APP, and then send the book listening, voice navigation and other instructions to the APP end, and the APP provides the corresponding services. In addition, the MCU processor 110 also communicates with the cloud background, so that when the user encounters an emergency and has no way to ask for help, the user can also send a request instruction to the cloud background, and the staff on the cloud background end can provide remote telephone navigation or other related services, which will not be described here.

[0034] The intelligent blind guiding instrument 10 of the present application is arranged separately from the host 100 and the head-mounted device 200, reducing the head load of the user when using, the blind guiding stick 300 is arranged, the navigation function of the blind guiding instrument is started when the camera 220 detects the image of the predetermined position on the blind guiding stick 300, and the MCU processor 110 processes the image around the blind guiding stick 300 collected by the camera 220 in real time, so as to output the guiding prompt information, facilitating the user to accurately obtain the road condition and navigation information of the advancing direction, and improving the reliability of the navigation function of the blind guiding instrument; by loading the image database, the object recognition mode can be entered on the basis of recognizing the gesture, or the audio can be played in the voice control mode when the voice instruction is received, so as to provide multiple ways for the user to recognize objects and obtain information, realize one machine with multiple functions, expand the application scene of the blind guiding instrument, the user only needs one device to meet multiple functional requirements, reduce the blind guiding cost of the user, and provide convenience for the life of the blind people.

[0035] The technical features of the above-described embodiments can be combined arbitrarily, and to make the description concise, all possible combinations of the technical features in the above-described embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, it should be considered that they are within the scope of the present application.

[0036] The above-described embodiments only express several implementation manners of the present application, the description is more specific and detailed, but it should not be understood as the limitation of the patent scope of the present application. It should be pointed out that, for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which belong to the protection scope of the present application. Therefore, the patent protection scope of the present application should be subject to the appended claims.

Claims

1. An intelligent guide device for the visually impaired, characterized in that, include: The host is used to be installed on clothing or accessories. The host contains an MCU processor for data processing and signal response processing, and a power module electrically connected to the MCU processor for powering the guide device. The MCU processor is communicatively connected to a backend server or cloud server and loads an image database sent by the backend server or cloud server. The image database includes image data of different objects, image data of different scenes, electronic maps, and image data of dangerous items. The head-mounted device includes a frame for wearing on the head or neck of a human body, a camera for image acquisition, a microphone for voice information acquisition, and a speaker for outputting voice prompts, all mounted on the frame and electrically connected to the MCU processor via connecting wires. as well as The guide cane is connected to the MCU processor via Bluetooth. It is used for camera recognition and positioning, and for locating guide points and determining the direction of guidance. At the same time, it triggers the activation of the guide device's guidance function. The MCU processor receives image information sent by the camera and identifies and judges the image information category to control the guide device to enter voice navigation mode or object recognition mode. When the camera captures image information of a preset position on the guide cane, the MCU processor controls the speaker to turn on to enter voice navigation mode, captures images of the user's direction of travel, and compares the image information around the guide cane captured by the camera in real time with electronic maps and images of dangerous objects, and outputs directional prompts and danger warnings. When the camera captures a preset gesture, the MCU processor controls the speaker to turn on to enter object recognition mode. The MCU processor also controls the guide device to enter voice control mode when it detects that the microphone is emitting predetermined voice content, so as to start or stop the guide device or play predetermined audio content. The MCU processor also interacts and authenticates with a third-party APP, and switches to remote control mode when it receives a predetermined gesture, voice or button command. When the guide device switches to remote control mode, the third-party APP user receives the image information sent by the camera and calls the speaker to send voice prompt information.

2. The intelligent guide device for the visually impaired according to claim 1, characterized in that, The host's casing is equipped with a display screen that is electrically connected to the MCU processor, used to display the owner's personal information and device information.

3. The intelligent guide device for the visually impaired according to claim 2, characterized in that, The frame is shaped like glasses, a headband, a head strap, or a hat.

4. The intelligent guide device for the visually impaired according to claim 3, characterized in that, The frame is also equipped with a radar detector head that is electrically connected to the MCU processor. When the guide device is in voice navigation mode, the MCU processor fits the image information sent by the camera and the radar signal data sent by the radar detector head to correct the road condition information.

5. The intelligent guide device for the visually impaired according to claim 4, characterized in that, In voice navigation mode, when the device detects a moving object, a preset danger image, a preset destination, or a preset route, the speaker outputs a predetermined voice signal accordingly.

6. The intelligent guide device for the visually impaired according to claim 5, characterized in that, The object recognition modes include a local scene recognition sub-mode that switches when a first gesture image is detected; an online scene recognition sub-mode that switches when a second gesture image is detected; and an item recognition sub-mode that switches when a third gesture image is detected.

7. The intelligent guide device for the visually impaired according to claim 6, characterized in that, When the microphone receives a voice command indicating a location, it sends a request to the MCU processor, which then marks the current location on an electronic map.

8. The intelligent guide device for the visually impaired according to claim 7, characterized in that, The MCU processor also plans a navigation route on an electronic map based on the guide route instructions sent by the microphone.

9. The intelligent guide device for the visually impaired according to claim 8, characterized in that, The frame is also equipped with multiple hand controls for starting and stopping the guide device, switching modes, and adjusting volume and speech rate.

Images