TWI808017B - Portably auxiliary system for visual impairment - Google Patents

Abstract

A portably auxiliary system for visual impairment includes a main body, a first image sensor unit, a control unit, a processing unit and a output unit. The main body is used for wearing on an user. The first image sensor unit is disposed on the main body and is used for detecting an environment image. The control unit is disposed on the main body, electrically connected to the first image sensor unit, and received the environment image. The processing unit is disposed in the control unit or a portable electronic device and in connection with the control unit, the processing unit processes the environment image to obtain an object information of an object which is existed in the surrounding environment. The output unit is disposed on the main body, electrically connected to the control unit and is used for receiving the object information, the output unit converts the object information to a cognition guidance, and the output unit transmits the cognition guidance to the user by the main body.

Description

Portable Assistive System for the Visually Impaired

The invention relates to a visually impaired assisting system, in particular to a portable visually impaired assisting system for assisting the visually impaired to obtain object information of objects in the surrounding environment.

Visually impaired people refer to people with visual impairments. The visually impaired people can be classified into total blindness, severe amblyopia (severe low vision) and slight amblyopia (slight low vision) according to the international visual impairment scale area. Physical mobility is one of the biggest challenges faced by the visually impaired, who can only rely on their keen senses of hearing, touch, and smell to judge the state of their surroundings. Common aids for the visually impaired include guide dogs and white canes. Due to the development of technology, portable global positioning system (GPS) devices that can be used by the visually impaired have been developed, which are connected to laptops or other devices dedicated to the visually impaired, so as to guide the visually impaired to their destinations.

However, whether it is a guide dog, a white cane, or a portable GPS device, there is no way to correctly provide environmental information, and each has its own problems. For example, guide dogs are expensive and cumbersome to maintain. For example, the white cane can only provide information about obstacles whose length can be touched, and the information of obstacles beyond the length of the white cane cannot be known because it cannot be touched. For example, while GPS technology can provide heading, GPS technology does not take into account any obstacles along the way.

In addition, currently on the market, distance sensors are also installed in earphones, glasses brackets, collars, necklaces or clothes, and the distance sensors are used to detect the distance of obstacles. However, such The device has the following disadvantages: first, it can only sense the distance of obstacles and cannot provide other information; second, it is bulky and bulky, making it uncomfortable to wear; third, it is easy to be found to be a special object for the visually impaired and be discriminated against.

The main purpose of the present invention is to provide a portable assisting system for the visually impaired, which can assist the visually impaired to know the object information of the objects in the surrounding environment.

In order to achieve the aforementioned objective, the present invention provides a portable visually impaired assistance system, which includes a main body, a first image sensing unit, a control unit, a processing unit and an output unit. The body is used to be worn on the body of the user. The first image sensing unit is disposed on the body and used for sensing an environment image. The control unit is arranged on the main body, is electrically connected to the first image sensing unit, and is used for receiving the environment image. The processing unit is disposed on the control unit, or disposed on a portable electronic device and connected to the control unit, the processing unit processes the environment image to obtain object information of an object existing in the environment item. The output unit is arranged on the body, is electrically connected to the control unit, and is used for receiving the object information. The output unit converts the object information into a cognitive guide, and transmits the cognitive guide to the user through the main body.

In some embodiments, the portable assistive system for the visually impaired further includes a second image sensing unit disposed on the main body, electrically connected to the control unit, and used for sensing a gesture of a user to obtain an image of a gesture.

In some embodiments, the processing unit further includes an image database and a comparison program; the object information includes an orientation information, a distance information and a shape information; the object is compared with the image database by the processing unit to obtain the shape information, and the output unit outputs the corresponding cognitive guidance.

In some embodiments, the comparison program further includes a deep learning algorithm, the shape information is calculated by the deep learning algorithm, and the corresponding cognitive guidance is output through the output unit.

In some embodiments, the portable assistive system for the visually impaired further includes an operation interface, the operation interface is disposed on the main body and electrically connected to the processing unit, or disposed on the portable electronic device, and the user obtains a target object information, a travel information and an obstacle information of the object through the operation interface setting.

In some embodiments, the portable visually impaired assistance system further includes an input unit, the input unit is disposed on the main body and electrically connected to the control unit, or is disposed on a portable electronic device and connected to the control unit, for the user to input a search command; the processing unit searches the environment image for the object matching the search command through the first image sensing unit and the comparison program, and converts the search command into a command guide, the output unit receives the command guide and transmits the command guide to the user through the body.

In some embodiments, the first image sensing unit is used for sensing a user's reaction behavior, the control unit is used for receiving the reaction behavior, and the processing unit compares the reaction behavior with the image database through the comparison program to obtain the shape information.

In some embodiments, the shape information further includes an abstract class of the object.

In some embodiments, the first image sensing unit senses the environment image by optical radar technology, structured light technology, indirect time-of-flight technology, direct time-of-flight technology or image recognition technology.

In some embodiments, the cognitive guidance is microcurrent, sound, voice or vibration, and the output unit adjusts the amplitude or frequency of the microcurrent, sound, voice or vibration and the volume of the sound or voice according to the object information.

The effect of the present invention is that the portable visually impaired assistance system of the present invention can assist the visually impaired to know the object information of the objects in the surrounding environment, and is easy to wear on the visually impaired, so that the visually impaired can understand the situation of the surrounding environment, and engage in general daily activities, and even exercise.

1: Portable Electronic Devices

10: Ontology

11: head ring

111: The first belt body

112: the second belt body

12: hat

121: perforation

20: The first image sensing unit

30: Control unit

40: Processing unit

41: Image database

42: Comparison program

50: output unit

60: Operation interface

70: The second image sensing unit

80: input unit

FIG. 1 is a schematic diagram of the first embodiment of the portable visually impaired assistance system of the present invention.

Fig. 2A is an exploded view of the body of the present invention with a headband and hooded structure.

Fig. 2B is a schematic diagram of the body of the present invention having a headband-hooded structure and being worn on the head.

FIG. 3 is a schematic diagram of the portable visually impaired assistance system of the present invention sensing environment images, inputting search commands and providing corresponding functions.

FIG. 4 is a schematic diagram of the portable visually impaired assistance system of the present invention sensing a user's gesture and providing corresponding functions.

FIG. 5 is a schematic diagram of a second embodiment of the portable visually impaired assistance system of the present invention.

The implementation of the present invention will be described in more detail below with reference to the drawings and reference symbols, so that those skilled in the art can implement it after studying this specification.

As shown in FIG. 1 , FIG. 2A and FIG. 2B , the present invention provides a portable assistive system for the visually impaired, including a main body 10 , a first image sensing unit 20 , a control unit 30 , a processing unit 40 , an output unit 50 , an operation interface 60 , a second image sensing unit 70 and an input unit 80 . Book The body 10 is used to be worn on the body of a user. The first image sensing unit 20 is disposed on the main body 10 . The control unit 30 is disposed on the body 10 and electrically connected to the first image sensing unit 20 . The processing unit 40 is disposed on the control unit 30 . The output unit 50 is disposed on the body 10 and electrically connected to the control unit 30 . The operation interface 60 , the second image sensing unit 70 and the input unit 80 are all disposed on the main body 10 and electrically connected to the control unit 30 .

FIG. 3 is a schematic diagram of the portable visually impaired assistance system of the present invention sensing environment images, inputting search commands and providing corresponding functions. As shown in FIG. 3 , the first image sensing unit 20 is used to sense an environment image; the control unit 30 is used to receive the environment image; the processing unit 40 processes the environment image to obtain object information of an object existing in the environment image; the output unit 50 is used to receive the object information, convert the object information into a cognition guide, and convey the cognition guide to the user through the main body 10. In this way, the user can know the object information of the objects in the surrounding environment through the cognitive guidance.

In some embodiments, the first image sensing unit 20 senses the environment image by optical radar technology (LiDAR), structured light technology, indirect time-of-flight technology (iToF), direct time-of-flight technology (dToF) or image recognition technology.

In some embodiments, the output unit 50 is an electrode, a speaker, a vibrator or a combination thereof. The cognitive guidance generated by the electrode is microcurrent, the cognitive guidance generated by the speaker is sound or voice, and the cognitive guidance generated by the vibrator is vibration.

As shown in FIGS. 1 and 3 , in a preferred embodiment, the processing unit 40 further includes an image database 41 and a comparison program 42. The comparison program 42 includes a deep learning algorithm. The object information includes an orientation information, a distance information and a shape information. The object obtains shape information by comparing the image database 41 through the comparison program 42. The shape information is calculated by the deep learning algorithm, and the corresponding cognitive guidance is output through the output unit 50. Therefore, the user can know the orientation information, distance information and shape information of the object through the cognitive guidance.

The orientation information of the object can define the orientation of the object in the environment image relative to the first image sensing unit 20 . For example, if there is a car parked on the left side of the user, the first image sensing unit 20 can sense that there is a car parked on the left side of the user. However, the description of the orientation information of the above-mentioned objects is only an example, and is not limited thereto.

The distance information of the object can define the distance of the object in the environment image relative to the first image sensing unit 20 . For example, the user can set the sensing range of the first image sensing unit 20 to objects within 10 meters through the operation interface 60. If a dog lover leads a puppy into the sensing range of the first image sensing unit 20, the first image sensor 20 can sense a person and a dog within 10 meters. However, the description of the distance information of the above-mentioned objects is only an example, and is not limited thereto.

In some embodiments, the shape information of the object further includes an abstract category or a concrete form of the object. Therefore, the user can further know the concrete form or abstract category of the object through the cognitive guidance.

Object specifics are individual nouns that define objects in the environment image. For example, individual nouns of objects include vehicles such as buses, taxis, locomotives, and bicycles. For example, individual nouns of objects include crosswalks, sidewalks, pedestrian-only signs, traffic signs, arterial road ahead, attention to falling rocks, and left-turn traffic control facilities. For example, individual nouns for objects include animals such as people, cats, dogs, and birds. For example, individual nouns for objects include indoor objects such as tables, chairs, cups, computers, keyboards, mice, televisions, and washing machines. For example, individual nouns of objects include moving vehicles such as elevators and escalators. However, the description of the specific form of the above-mentioned objects is only an example, and is not limited thereto. The specific form of the object can make the user clearly understand what objects are in the surrounding environment.

The abstract category of objects refers to collective nouns that can define objects in an environment image, such as vehicles, traffic control facilities, animals, indoor objects, and mobile vehicles. However, the description of the above abstract categories of objects is only an example, not a limitation. The user can set the object information provided by the processing unit 40 to the output unit 50 through the operation interface 60 as the abstract category of the object. The abstract category of the object can allow the user to understand which category the objects in the surrounding environment belong to, and can avoid confusion caused by too many objects causing the user to receive too much information at the same time.

In some embodiments, the processing unit 40 further obtains a target information, a traveling information, and an obstacle information of the object after being set by the user through the operation interface 60 . The target object information is the place to be reached or the item to be taken. The traveling information is the real-time direction that the user wants to go to the target. Obstacle information is an object that reminds the user to avoid to avoid collision. For example, the user can set the object through the operation interface 60 to be a water cup placed in the meeting room, and the target object information obtained by the processing unit 40 includes the meeting room (the place to be reached) and the water cup (the item to be taken); the travel information obtained by the processing unit 40 is "go straight ahead 10 meters and then turn right to enter the meeting room"; if there is a pillar on the road, the obstacle information obtained by the processing unit 40 is "there is a pillar 3 meters ahead, please pay attention", the user can easily flash past the pillar and continue moving forward; etc. After the user enters the meeting room, the processing unit 40 obtains the traveling information as "walk straight forward 3 meters"; after the user walks straight forward 3 meters, the processing unit 40 obtains the traveling information as "reach the destination, the water glass is on your left table". Regarding the target information and traveling information of the object, the output unit 50 transmits voice or sound as cognitive guidance to the user through the main body 10; regarding the obstacle information of the object, the output unit 50 transmits vibration or micro current or as a cognitive guidance to the user through the main body 10. However, the descriptions of the target information, travel information, and obstacle information of the above-mentioned objects are only examples, and are not limited thereto.

As shown in Figure 2A, in a preferred embodiment, the body 10 is a headband hood structure and includes a headband 11 and a hat 12, the headband 11 includes a first belt 111 and a second belt 112, the first belt 111 is circular, the second belt 112 is arched and its two ends are respectively connected to the first belt 111, four first image sensing units 20 are arranged on the front side, rear side, left side and right side of the first belt 111, a first image sensor The unit 20 is disposed on the top of the second belt 112 , and the front side, the rear side, the left side, the right side and the top of the hat 12 respectively define a through hole 121 . As shown in FIG. 2B , the headband 11 is worn on the top of the user's head, and the hat 12 is worn on the top of the user's head and is located outside the headband 11 . The positions of the first image sensing units 20 correspond to the through holes 121 of the hat 12 . Therefore, the first image sensing units 20 can penetrate through the through holes 121 and sense environmental images with different orientation information. For example, FIG. 2B represents the field of view of the first image sensing unit 20 (that is, a certain orientation that it can sense) with an imaginary line. The first image sensing unit 20 on the front side can sense objects in the front and right front, and the first image sensing unit 20 on the left side can sense objects in the left and rear left, and both can jointly sense objects in the front left.

In some embodiments, the material of the headband 11 is elastic cloth and the lengths of the first strap 111 and the second strap 112 can be adjusted, so that the headband 11 can be worn on heads of any size.

In some embodiments, the first belt body 111 and the cap 12 can be fixed or detached to each other by Velcro, buttons, clasps or other common fixing structures.

In some embodiments, the hat 12 may not have the perforations 121 , and the user may put on the hat 12 first, and then put the headband 11 on the hat 12 .

In some embodiments, the body 10 is a finger ring structure (not shown in the figure) and includes a ring body and a mounting base, the ring body is worn on the user's finger, the mounting base is disposed on the ring body, and a first image sensing unit 20 is disposed on the front side of the mounting base. In other words, the first image sensing unit 20 faces toward the fingertip, so it can sense the environment image in the direction the finger is pointing. In some embodiments, the mount can Rotate horizontally or vertically relative to the ring body, thereby expanding the sensing range of the first image sensing unit 20, which is helpful for collecting environmental images in all directions.

In some embodiments, the body 10 is an earhook structure (not shown in the figure) and includes an earhook and an earphone, the earhook is hung on the user's ear, the earphone is set on the earhook, and a first image sensing unit 20 is set on the earphone. The earphone 16 may be a bone conduction earphone or a general earphone with a speaker.

In some embodiments, the body 10 is a neck hanging structure (not shown in the figure) and includes a neck hanging and a casing, the neck hanging is hung on the user's neck, the casing is arranged on the neck hanging, and a first image sensing unit 20 is arranged on the casing.

However, the specific form of the main body 10 is not limited to the above four, and any accessory that can be worn on the body of the user can become the specific form of the main body 10, such as a white cane, gloves, watches, bracelets, necklaces, glasses, earrings or hats.

It should be noted that since the first image sensing unit 20 includes a substrate (not shown in the figure) and a plurality of unit pixels (not shown in the figure), the unit pixels are arranged on the substrate, so the volume of the first image sensing unit 20 is relatively small, and the very small first image sensing unit 20 is suitable for being installed on the above-mentioned specific forms of the body 10, and will not cause a burden to the user, and will not be discriminated against because it is a special object for the visually impaired.

In some embodiments, the output unit 50 adjusts the amplitude or frequency of the microcurrent, sound, voice or vibration, and the volume of the sound or voice according to the object information. For example, the closer the user is to the object, the greater the amplitude or frequency of the microcurrent, sound, voice or vibration, and the greater the volume of the sound or voice; conversely, the farther the user is from the object, the smaller the amplitude or frequency of the microcurrent, sound, voice or vibration, and the smaller the volume of the sound or voice. The abstract category of objects is more suitable for informing users with cognitive guidance such as microcurrent or vibration, and the concrete form of objects is more suitable Cognitive guidance such as sound or voice is used to notify the user, so as to prevent the output unit 50 from simultaneously issuing multiple complicated cognitive guidance to disturb the user.

In some embodiments, the number of output units 50 may be plural and disposed at different positions of the body 10 . When there is only one object around the user, the processing unit 40 can transmit the orientation information through the control unit 30 to the output unit 50 at a specific position of the main body 10 according to the orientation of the object, and the user can know the actual orientation of the object from the cognitive guidance. When there are multiple objects around the user, the present invention provides the following two implementations: first, the processing unit 40 can determine which object is closest to the user according to the distance information of the objects, and the processing unit 40 further transmits the object information of the closest object to the output unit 50 at a specific position of the main body 10 through the control unit 30, and the user can know the object information of the closest object from the cognitive guidance; The unit 30 transmits to a plurality of output units 50 at a specific position of the main body 10. The output units 50 output micro-currents, sounds, voices or vibrations with different amplitudes or frequencies and different volumes of the sounds or voices. The user can learn the object information of these objects from the micro-currents, sounds, voices or vibrations with amplitudes or frequencies and the volume of the sounds or voices.

Generally speaking, the user is usually a visually impaired person, and can wear a plurality of portable visually impaired assistance systems of the present invention on his body according to his needs.

For example, when the user wears the ear-hanging structure and the neck-hanging structure at the same time, the first image sensing unit 20 in the ear-hanging structure is responsible for sensing short-distance horizontal and vertical environmental images, and the first image sensing unit 20 in the neck-hanging structure is responsible for sensing long-distance horizontal environmental images.

Take badminton as an example. The first image sensing unit 20 in the ear-hook structure is responsible for sensing the image of the badminton court (environmental image). The processing unit 40 in the ear-hook structure judges the orientation information and distance information of the shuttlecock (object) in the image of the badminton court according to the image of the badminton court. The loudspeaker or vibrator (output unit 50) in the ear-hook structure transmits sound or vibration (cognitive guidance) to the user through the ear-hook structure. Loud, or the volume of the sound is getting louder and louder, knowing that the shuttlecock is getting closer to the user. The first image sensing unit 20 in the neck hanging structure is responsible for sensing the image (environmental image) of the badminton court. The processing unit 40 in the neck hanging structure judges the orientation information and distance information of the opponent (object) in the image of the badminton court according to the image of the badminton court. The speaker (output unit 50) in the neck hanging structure transmits the voice (cognitive guidance) to the user through the neck hanging structure.

Take advancing to the meeting room (target object information of the object) as an example. The first image sensing unit 20 in the ear-hook structure is responsible for sensing the image of the road (environmental image), and the processing unit 40 in the ear-hook structure judges the orientation information and distance information of multiple obstacles (obstacle information of objects) in the image of the road according to the image of the road. The azimuth information and distance information of different obstacles, as well as the azimuth information and distance information of the closest obstacle can be obtained by the amplitude or frequency of the vibration. The first image sensing unit 20 in the neck hanging structure is responsible for sensing the image (environmental image) of the doorway of the meeting room, and the processing unit 40 in the neck hanging structure judges the orientation information and distance information of the door or the door plate (object) of the meeting room in the image of the doorway of the meeting room according to the image of the doorway of the meeting room. The loudspeaker (output unit 50) transmits sound or voice (cognitive guidance) to the user through the neck hanging structure, and the user knows the direction information and distance information of the door or door plate of the meeting room according to the sound or voice.

For example, when the user wears the ring structure, the ear-hook structure and the neck-hook structure at the same time, the first image sensing unit 20 in the ring structure is responsible for sensing the environment image in the direction the finger is pointing, the first image sensing unit 20 in the ear-hook structure is responsible for sensing the long-distance environment image, and the first image sensing unit 20 in the neck-hook structure is responsible for sensing the short-range environment image.

Take crossing the road as an example. The first image sensing unit 20 in the ear-hook structure is responsible for sensing the image of the road (environmental image). The processing unit 40 in the ear-hook structure judges the passable state and remaining passable time (specific form of object information) of the pedestrian-specific signs (objects) in the image of the road based on the image of the road. The first image sensing unit 20 in the neck hanging structure is responsible for sensing the image of the road (environmental image). The processing unit 40 in the neck hanging structure judges the orientation information and distance information of the pedestrian crossing (object) in the image of the road according to the image of the road. crosswalk and guide users back to the crosswalk. The first image sensing unit 20 in the ring structure is responsible for sensing the image of the obstacle in the direction pointed by the finger (environment image). The processing unit 40 in the ring structure judges the orientation information and distance information of the obstacle (object) according to the image of the obstacle in the direction pointed by the finger. The electrode or vibrator (output unit 50) in the ring structure transmits the micro current or vibration (cognitive guidance) to the user through the ring structure, and the user learns the orientation information and distance information of the obstacle according to the micro current or vibration. For example, the direction pointed by the finger is "front", and the obstacle is "Pedestrian", the user can dodge in advance. For example, the direction pointed by the finger is "left" or "right", and the obstacle is "vehicle". The user can stop and wait for the vehicle to pass before moving forward.

FIG. 4 is a schematic diagram of the portable visually impaired assistance system of the present invention sensing a user's gesture and providing corresponding functions. As shown in FIG. 4 , the second image sensing unit 70 is used to sense a user's gesture to obtain an image of a gesture, and the input unit is a microphone and used to receive a voice. Users can give instructions through gestures or voice. For example, the user can notify the portable visually impaired assistance system of the present invention to point to an object through gestures or voice, and the output unit 50 will convey the cognitive guidance to the user through the main body 10, and the user can know the distance information or shape information of the object through the cognitive guidance. For example, the portable visually impaired assistance system of the present invention can ask questions through the output unit 50 , and the user can answer the questions of the portable visually impaired assistance system of the present invention through gestures or voice. For example, the user can set the operation mode of the portable visually impaired assistance system of the present invention through gestures or voice.

As shown in FIG. 3 , the input unit 80 is used for the user to input a search command, and the control unit 30 is used for receiving the search command. The processing unit uses the first image sensing unit 20 and the comparison program 42 to search for objects in the environment image that match the search command, and converts the search command into a command guide. The output unit 50 receives the command guide and transmits the command guide to the user through the main body 10. Specifically, the input unit 80 is a microphone, the search command is a voice, and the microphone is used to receive the voice. The second image sensing unit 70 can also be used for the user to input a search command, and the search command is a gesture of the user. Therefore, the voice or gesture may be the user asking whether the object is an object or an obstacle. If the processing unit 40 uses the first image sensing unit 20 and the comparison program 42 to search for an object matching the search command in the environment image as the target object, the output unit 50 transmits the voice or sound as a command guide to the user through the main body 10 . If the processing unit 40 uses the first image sensing unit 20 and the comparison program 42 to search out the matching The object of the search command is an obstacle, and the output unit 50 transmits vibration or micro current or as a command guide to the user through the main body 10 .

As shown in FIG. 3 , the first image sensing unit 20 is used to sense a user's reaction behavior, the control unit 30 is used to receive the reaction behavior, and the reaction behavior is processed by the processing unit 40 through the comparison program 42 to compare the image database 41 to obtain shape information. Specifically, the reaction behavior refers to the user’s reaction when encountering an object. The first image sensing unit 20 is used to sense the user’s reaction when encountering an object. The processing unit 40 compares the image database 41 through the processing unit 40 through the comparison program 42. The deep learning algorithm of the comparison program 42 will self-learn and judge the specific form of the object and automatically classify its abstract category. For example, the user generally sits on a chair and leans on the table to operate the keyboard and mouse. The first image sensing unit 20 will sense the above-mentioned reaction behavior, and the control unit 30 is used to receive the above-mentioned reaction behavior. The above-mentioned reaction behavior is processed by the processing unit 40. During the process of comparing the image database 41 through the comparison program 42, the deep learning algorithm of the comparison program 42 will self-learn and judge the specific form of the object as a chair, a table, a keyboard, and a mouse, and automatically classify the abstract categories.

FIG. 5 is a schematic diagram of a second embodiment of the portable visually impaired assistance system of the present invention. As shown in FIG. 5 , the difference between the second embodiment and the first embodiment is that the processing unit 40 , the operation interface 60 and the input unit 80 are all arranged in a portable electronic device 1 , the processing unit 40 is connected to the control unit 30 , and the operation interface 60 and the input unit 80 are electrically connected to the processing unit 40 . The processing unit 40 can be a microprocessor or an application program built in the portable electronic device 1 . The operation interface 60 can be a touch screen or buttons of the portable electronic device 1 . The portable electronic device 1 (for example, a smart phone or a tablet computer) can be loaded with a high-capacity lithium battery, and the lithium battery provides the required power for relatively power-consuming components (for example, the processing unit 40), and the processing unit 40 configured therein usually has high-performance computing capabilities to perform more complex and more power-consuming programs (for example, analysis and comparison programs and learning programs). Processing in Portable Electronic Device 1 The unit 40 can also exchange messages with the control unit 30 in the main body 10 . Therefore, the number of components configured on the body 10 is reduced, making it lighter and less burdening on the user.

To sum up, the portable visually impaired assistance system of the present invention can assist the visually impaired to know the object information of the objects in the surrounding environment, and is easy to wear on the visually impaired, so that the visually impaired can understand the situation of the surrounding environment, and engage in general daily activities, and even exercise.

The above are only preferred embodiments for explaining the present invention, and are not intended to limit the present invention in any form. Therefore, any modifications or changes made under the same spirit of the present invention should still be included in the intended protection scope of the present invention.

10: Ontology

20: The first image sensing unit

30: Control unit

40: Processing unit

41: Image database

42: Comparison program

50: output unit

60: Operation interface

70: The second image sensing unit

80: input unit

Claims (10)

A portable assistive system for the visually impaired, comprising: a body, used to be worn on the body of a user; a first image sensing unit, disposed on the body, and used to sense an environmental image; a control unit, disposed on the body, electrically connected to the first image sensing unit, and used to receive the environmental image; a processing unit, disposed on the control unit, or disposed on a portable electronic device and connected to the control unit, the processing unit processes the environmental image to obtain object information of an object existing in the environmental image; And an output unit, arranged on the body, electrically connected to the control unit, and used to receive the object information, the output unit converts the object information into a cognitive guide, and conveys the cognitive guide to the user through the main body. The portable assistive system for the visually impaired as described in claim 1 further includes a second image sensing unit disposed on the main body, electrically connected to the control unit, and used for sensing a gesture of a user to obtain an image of a gesture. The portable assistive system for the visually impaired as described in Claim 1, wherein the processing unit further includes an image database and a comparison program; the object information includes azimuth information, a distance information, and a shape information; the object obtains the shape information through the comparison program through the comparison program to obtain the shape information, and outputs the corresponding cognitive guidance through the output unit. The portable visually impaired assistance system as described in Claim 3, wherein the comparison program further includes a deep learning algorithm, the body information is calculated by the deep learning algorithm, and the corresponding cognitive guidance is output through the output unit. The portable assistive system for the visually impaired as described in claim 3 further includes an operation interface, the operation interface is disposed on the body and electrically connected to the control unit, or disposed on the portable electronic device and electrically connected to the processing unit, and the processing unit further obtains a target object information, a travel information and an obstacle information of the object through the user setting through the operation interface. The portable visually impaired assistance system as described in claim 3 further includes an input unit, which is arranged on the main body and is electrically connected to the control unit, or is arranged on a portable electronic device and is electrically connected to the processing unit, for the user to input a search command; the processing unit uses the first image sensing unit and the comparison program to search for the object in the environment image that matches the search command, and converts the search command into a command guide, and the output unit receives the command guide and transmits the command guide to the user through the body. The portable assistive system for the visually impaired as described in Claim 3, wherein the first image sensing unit is used to sense a reaction behavior of the user, the control unit is used to receive the reaction behavior, and the processing unit compares the reaction behavior with the image database through the comparison program to obtain the shape information. In the portable assistive system for the visually impaired as described in claim 3, wherein the body information further includes an abstract category of the object. The portable visually impaired assistance system according to claim 1, wherein the first image sensing unit senses the environment image by using optical radar technology, structured light technology, indirect time-of-flight technology, direct time-of-flight technology or image recognition technology. The portable aiding system for the visually impaired as described in Claim 1, wherein the cognitive guidance is a microcurrent, sound, voice or vibration, and the output unit adjusts the amplitude or frequency of the microcurrent, sound, voice or vibration and the volume of the sound or voice according to the object information.

Images