TWI710749B - Indoor positioning method with improved accuracy and mobile device using the same - Google Patents

Abstract

An indoor positioning method with improved accuracy and a mobile device using the same are provided. The indoor positioning method for improving the accuracy includes the following steps. At least one live photo is captured by the mobile device. An electronic compass value and a gravity sensor value are captured. Whether the electronic compass value satisfies a first preset condition is determined. Whether the gravity sensor value satisfies a second preset condition is determined. According to the local photo, an indoor position of the mobile device is identified through an identification model. If the electronic compass value satisfies the first preset condition and the gravity sensor value satisfies the second preset condition, the indoor position is displayed.

Description

Indoor positioning method for improving accuracy and mobile device using the same

The present disclosure relates to an indoor positioning method, and in particular to an indoor positioning method that improves accuracy and a mobile device using the same.

With the advancement of technology, a geographic location positioning technology has been developed. The geographic location positioning technology uses the global positioning system (GPS) for positioning, for example. However, the global positioning system cannot be used indoors, and indoor positioning technology cannot be developed.

At present, indoor positioning technology is trying to use Wi-Fi signals, but it has problems such as poor accuracy, additional equipment costs, construction costs, adjustment costs, difficult installation, and excessive power consumption. universal.

However, if you want to use image recognition technology to perform indoor positioning, it is difficult to recognize or recognize errors due to the different shooting angles and directions. Therefore, researchers are working on this aspect of research, with a view to smoothly promoting indoor positioning technology.

This disclosure relates to an indoor positioning method with improved accuracy and a mobile device using the same. It uses orientation verification technology to appropriately remind the user to turn or tilt to obtain the correct orientation of the spot photo, thereby improving the identification accuracy Accuracy rate.

According to the first aspect of this disclosure, an indoor positioning method with improved accuracy is proposed. The indoor positioning method to improve accuracy includes the following steps. Take at least one spot photo with a mobile device. Capture an electronic compass value and a gravity sensor value. Determine whether the value of the electronic compass meets a first preset condition. Determine whether the value of the gravity sensor meets a second preset condition. According to the local photos, the indoor location of a mobile device is identified through a recognition model. If the electronic compass value meets the first preset condition and the gravity sensor value meets the second preset condition, the indoor position is displayed.

According to the second aspect of this disclosure, an indoor positioning method with improved accuracy is proposed. The mobile device includes an image capturing unit, an electronic compass, a gravity sensor, a processing unit and a display unit. The imaging unit is used for taking at least one spot photo. The electronic compass is used to capture an electronic compass value. The gravity sensor is used to capture a value of the gravity sensor. The processing unit is used to determine whether the value of the electronic compass meets a first preset condition, and to determine whether the value of the gravity sensor meets a second preset condition. The processing unit recognizes an indoor location of the mobile device through a recognition model based on the local photos. If the value of the electronic compass meets the first preset condition and the value of the gravity sensor meets the second preset condition, the processing unit controls the display unit to display the indoor position.

In order to have a better understanding of the above and other aspects of the present disclosure, the following embodiments are specially cited, and the accompanying drawings are described in detail as follows:

100: mobile device

110: Acquisition unit

120: electronic compass

130: Gravity Sensor

140: Processing Unit

160: display unit

900: Indoor positioning training system

911, 912, 913, 914: imaging unit

920: electronic compass

930: Gravity Sensor

940: Training Unit

950: database

BD1: Building

CF11: Recognition confidence

D1: First direction

D2: second direction

D3: Third party

D4: Fourth direction

DR11: Rotation direction

DT11: Tilt direction

EC1, EC11, EC2, EC3, EC4: electronic compass values

GS1, GS11, GS2, GS3, GS4: gravity sensor value

LC11: Indoor location

MD1: Identification model

P11: Local photos

R1, R2, R3, R4, R5: partition

S110, S110’, S120, S120’, S130, S130’, S140, S150, S160, S170, S180, S190, S240, S240’, S250, S250’: steps

Figure 1 shows a schematic diagram of the interior of a building according to an embodiment.

Figure 2 shows a block diagram of an indoor positioning training system according to an embodiment

FIG. 3 is a schematic diagram of an image capturing unit, an image capturing unit, an image capturing unit, and an image capturing unit according to an embodiment.

Figure 4 is a schematic diagram of a mobile device according to an embodiment.

Fig. 5 shows a flowchart of an indoor positioning method for improving accuracy according to an embodiment.

FIG. 6 shows a flowchart of an indoor positioning method for improving accuracy according to another embodiment.

FIG. 7 is a flowchart of an indoor positioning method for improving accuracy according to another embodiment.

FIG. 8 is a flowchart of an indoor positioning method for improving accuracy according to another embodiment.

Figure 9 shows a timing diagram of capturing live photos, electronic compass values, and gravity sensor values.

Please refer to FIGS. 1 and 2. FIG. 1 is a schematic diagram of the interior of a building BD1 according to an embodiment, and FIG. 2 is a block diagram of an indoor positioning training system 900 according to an embodiment. The interior of the building BD1 cannot be positioned through the global positioning system, and the indoor positioning training system 900 can allow the user to know which partition R1, R2, R3, R4, R5,...

The indoor positioning training system 900 includes an image capturing unit 911, an image capturing unit 912, an image capturing unit 913, an image capturing unit 914, an electronic compass 920, a gravity sensor 930, a training unit 940, and a database. 950.

The sample photos P1, P2, P3, and P4 obtained by the image capturing unit 911, the image capturing unit 912, the image capturing unit 913, and the image capturing unit 914 can be input to the training unit 940. The training unit 940 uses the deep learning technology of artificial intelligence to build a recognition model MD1. The identification model MD1 is stored in the database 950.

Please refer to FIG. 3, which illustrates a schematic diagram of an image capturing unit 911, an image capturing unit 912, an image capturing unit 913, and an image capturing unit 914 according to an embodiment. For example, the imaging unit 911 faces the first direction D1, the imaging unit 912 faces the second direction D2, for example, the imaging unit 913 faces the third direction D3, and the imaging unit 914 faces the fourth direction D4, for example. The first direction D1 has a specific electronic compass value EC1 and a specific gravity sensor value GS1. The second direction D2 has a specific electronic compass value EC2 and a specific gravity sensor value GS2. The third direction D3 has a specific electronic compass value EC3 and a specific gravity sensor value GS3. The fourth direction D4 has a specific electronic compass value EC4 and a specific gravity sensor value GS4. In other words, the above-mentioned sample photos P1, P2, P3, and P4 are taken according to the specific first direction D1, second direction D2, third direction D3, and fourth direction D4.

Please refer to FIG. 4, which shows a schematic diagram of a mobile device 100 according to an embodiment. The mobile device 100 includes an image capturing unit 110, an electronic compass 120, a gravity sensor 130, a processing unit 140, and a display unit 150. Wherein, the image capturing unit 110 is, for example, a camera or a camera with photography and photographing functions. The processing unit 140 is, for example, a circuit, a chip, a circuit board, or a storage device that stores an array of program codes with calculation, processing, and control functions.

The imaging unit 110 can take a spot photo P11, and the processing unit 140 uses the recognition model MD1 provided by the indoor positioning training system 900 to perform indoor positioning. If the on-site photo P11 taken by the user is too different from the first direction D1, the second direction D2, the third direction D3, and the fourth direction D4, the problem of inaccurate recognition may occur. The mobile device 100 of this embodiment has the technology of position checking and reminding to improve the recognition accuracy. The following is a flowchart to explain in detail the indoor positioning method to improve accuracy.

Please refer to FIG. 5, which illustrates a flowchart of an indoor positioning method for improving accuracy according to an embodiment. In step S110, the imaging unit 110 of the mobile device 100 is used to take a spot photo P11.

Next, in step S120, the processing unit 140 recognizes the indoor position LC11 of the mobile device 100 through the recognition model MD1 according to the local photo P11. In this step, the recognition model MD1 will simultaneously generate the indoor position LC11 and the recognition confidence CF11. Factors that may affect the recognition confidence CF11 include the imaging orientation, exposure parameters of the imaging unit 110, shutter parameters, wide-angle parameters, and on-site personnel and objects. The higher the recognition confidence CF11, the higher the reliability of the indoor location LC11.

Then, in step S130, the electronic compass 120 captures an electronic compass value EC11, and the gravity sensor 130 captures a gravity sensor value GS11.

Next, in step S140, the processing unit 140 determines whether the electronic compass value EC11 satisfies a first preset condition. The first preset condition is, for example, a preset interval of ±5% or ±10% of the electronic compass values EC1, EC2, EC3, and EC4. If the electronic compass value EC11 meets the first preset condition, go to step S160; if the electronic compass value EC11 does not meet the first preset condition, go to step S150.

In step S150, the processing unit 140 controls the display unit 160 to prompt a rotation direction DR11. For example, the display unit 160 may prompt "turn right, turn left, turn clockwise, turn counterclockwise, turn right 20 degrees, turn left 20 degrees, turn clockwise 20 degrees, Turn 20 degrees counterclockwise" and so on.

In step S160, the processing unit 140 determines whether the gravity sensor value GS11 satisfies a second preset condition. The second preset condition is, for example, falling within a preset interval of ±5% or ±10% of the gravity sensor values GS1, GS2, GS3, and GS4. If the gravity sensor value GS11 meets the second preset condition, then step S180 is entered; if the gravity sensor value GS11 does not meet the second preset condition, then step S170 is entered.

In step S170, the processing unit 140 controls the display unit 160 to prompt a tilt direction DT11. For example, the display unit 160 may prompt “tilt up, tilt down, tilt up 20 degrees, tilt down 20 degrees” and so on.

After prompting the user to turn the direction DR11 or the tilt direction DT11 in step S150 and step S170, return to step S110 to take the spot photo again The picture P11 is used to obtain a more correct location picture P11, thereby increasing the recognition accuracy.

After entering step S180, it is confirmed that the electronic compass value EC11 meets the first preset condition and the gravity sensor value GS11 meets the second preset condition. In step S180, the processing unit 140 controls the display unit 160 to display the indoor position LC11. In this step, the processing unit 140 may further control the display unit 160 to display information about the correct shooting direction to inform the user that the shooting direction can correctly identify the position.

In step S190, the processing unit 140 further controls the display unit 160 to display the recognition confidence CF11. In this way, the mobile device 100 can correct the orientation of the local photo P11 by prompting the user to rotate the direction DR11 or the tilt direction DT11, so that the recognition accuracy can be greatly improved.

In one embodiment, the above-mentioned steps S140 and S160 can be executed in an exchange order or simultaneously. Please refer to FIG. 6, which shows a flowchart of an indoor positioning method for improving accuracy according to another embodiment. In step S240 in FIG. 6, the processing unit 140 simultaneously determines whether the electronic compass value EC11 meets the first preset condition and the gravity sensor value GS11 meets the second preset condition. If the first electronic compass value EC11 does not meet the first preset condition or the gravity sensor value GS11 does not meet the second preset condition, step S250 is entered. If the first electronic compass value EC11 satisfies the first preset condition and the gravity sensor value GS11 satisfies the second preset condition, then step S180 is entered.

In step S250, the processing unit 140 controls the display unit 160 to prompt the rotation direction DR11 and the tilt direction DT11. In other words, the processing unit 140 The electronic compass value EC11 and the gravity sensor value GS11 can be comprehensively judged, and the suitable rotation direction DR11 and the tilt direction DT11 can be directly given.

In another embodiment, the action of recognizing the indoor position LC11 can be performed after the orientation is correct, so as to avoid waste of computing resources. Please refer to FIG. 7, which shows a flowchart of an indoor positioning method for improving accuracy according to another embodiment. In the embodiment of Fig. 7, step S120 is performed after step S160, and the indoor position identification is performed only after the electronic compass value EC11 meets the first preset condition and the gravity sensor value GS11 meets the second preset condition Action of LC11.

In another embodiment, the image capturing unit 110 can shoot a video to obtain multiple shots of spot photos P11, and continuously capture the electronic compass value EC11 and the gravity sensor value GS11 as the spot photos P11 are obtained. Please refer to Figures 8 and 9. Figure 8 shows a flowchart of an indoor positioning method for improving accuracy according to another embodiment. Figure 9 shows a spot photo P11, an electronic compass value EC11, and a gravity sensor The timing diagram of the value GS11 captured. In step S110' in Fig. 8, the imaging unit 110 of the mobile device 100 continuously takes the spot photo P11. In step S120' in Fig. 8, the processing unit 140 recognizes a plurality of corresponding indoor locations LC11 through the recognition model MD1 based on the spot photos P11.

In step S130', the electronic compass 120 continuously captures the electronic compass value EC11, and the gravity sensor 130 continuously captures the gravity sensor value GS11. As shown in Figure 9, the capture of the spot photo P11, the capture of the electronic compass value EC11, and the capture of the gravity sensor value GS11 are performed simultaneously. The spot photo P11, the electronic compass value EC11 and the gravity sensor value GS11 at the same time point are grouped into the same group of data.

In step S240', the processing unit 140 determines whether one of the electronic compass value EC11 and the gravity sensor value GS11 satisfies the first preset condition and the second preset condition at the same time. If none of the electronic compass value EC11 and the gravity sensor value GS11 satisfies the first preset condition and the second preset condition at the same time, go to step S250'; the electronic compass value EC11 and the gravity sensor value GS11 If there is a group that satisfies both the first preset condition and the second preset condition, then step S180 is entered.

In step S250', the processing unit 140 controls the display unit 160 to prompt the rotation direction DR11 and the tilt direction DT11. The processing unit 140 can comprehensively determine the electronic compass value EC11 and the gravity sensor value GS11 of the last group, and directly give the appropriate rotation direction DR11 and the tilt direction DT11.

In step S180, the processing unit 140 controls the display unit 160 to display the indoor position LC11. In this step, according to the group of the electronic compass value EC11 and the gravity sensor value GS11 satisfying both the first preset condition and the second preset condition, the indoor position LC11 corresponding to the group is extracted.

In step S190, the processing unit 140 further controls the display unit 160 to display the recognition confidence CF11. In this step, according to the group of the electronic compass value EC11 and the gravity sensor value GS11 satisfying both the first preset condition and the second preset condition, the recognition confidence CF11 corresponding to the group is obtained.

In other words, the user can directly shoot continuous images, and obtain multiple on-site photos P11, and directly perform subsequent judgments and identifications based on them, thus speeding up indoor positioning.

According to the various embodiments described above, the mobile device 100 is equipped with an orientation verification technology, which can appropriately remind the user of the rotation direction or the tilt direction to obtain the correct orientation of the spot photo P11, thereby improving the recognition accuracy.

To sum up, although the present disclosure has been disclosed as above by embodiments, it is not intended to limit the present disclosure. Those with ordinary knowledge in the technical field to which this disclosure belongs can make various changes and modifications without departing from the spirit and scope of this disclosure. Therefore, the protection scope of this disclosure shall be subject to the scope of the attached patent application.

S110, S120, S130, S140, S150, S160, S170, S180, S190: steps

Claims (22)

An indoor positioning method for improving accuracy includes: taking at least one spot photo with a mobile device; capturing an electronic compass value and a gravity sensor value; determining whether the electronic compass value meets a first preset condition; Whether the value of the gravity sensor meets a second preset condition; according to the spot photo, an indoor position of the mobile device is identified through a recognition model; and if the value of the electronic compass meets the first preset condition and the gravity When the sensor value meets the second preset condition, the indoor position is displayed. For example, the indoor positioning method for improving accuracy described in item 1 of the scope of patent application further includes: if the value of the electronic compass does not meet the first preset condition, prompting a rotation direction. For example, the indoor positioning method for improving accuracy described in item 1 of the scope of patent application further includes: if the value of the gravity sensor does not meet the second preset condition, prompting a tilt direction. As described in item 1 of the scope of patent application, the indoor positioning method for improving accuracy includes: If the electronic compass value does not meet the first preset condition, a rotation angle is prompted. For example, the indoor positioning method for improving accuracy described in item 1 of the scope of patent application further includes: if the value of the gravity sensor does not meet the second preset condition, prompting an inclination angle. As described in item 1 of the scope of patent application, the indoor positioning method for improving the accuracy rate further includes: outputting a recognition confidence. The indoor positioning method for improving accuracy as described in item 1 of the scope of patent application, wherein the mobile device continuously takes pictures of the spot. The indoor positioning method for improving accuracy as described in item 1 of the scope of patent application, wherein the spot photo is presented on a display unit of the mobile device. As described in item 1 of the scope of patent application, the indoor positioning method for improving accuracy further includes: if the electronic compass value satisfies the first preset condition and the gravity sensor value satisfies the second preset condition, prompt Information about the correct shooting direction. In the indoor positioning method for improving accuracy described in item 1 of the scope of patent application, the first preset condition is a preset numerical interval. In the indoor positioning method for improving accuracy as described in item 1 of the scope of patent application, the second predetermined condition is a predetermined numerical interval. A mobile device includes: an image capturing unit for taking at least one spot photo; an electronic compass for capturing an electronic compass value; a gravity sensor for capturing a gravity sensor value; a processing Unit for judging whether the value of the electronic compass satisfies a first preset condition, and judging whether the value of the gravity sensor satisfies a second preset condition, and the processing unit recognizes the image through a recognition model based on the spot photo An indoor location of the mobile device; and a display unit, if the electronic compass value meets the first preset condition and the gravity sensor value meets the second preset condition, the processing unit controls the display unit to display the indoor position. For example, in the mobile device described in claim 12, if the value of the electronic compass does not meet the first preset condition, the processing unit controls the display device to prompt a rotation direction. For example, in the mobile device described in claim 12, if the value of the gravity sensor does not meet the second preset condition, the processing unit controls the display unit to prompt an inclined direction. For example, in the mobile device described in item 12 of the scope of patent application, if the value of the electronic compass does not meet the first preset condition, the processing unit controls the display unit to prompt a rotation angle. For example, in the mobile device described in claim 12, if the value of the gravity sensor does not meet the second preset condition, the processing unit controls the display unit to prompt an inclination angle. For the mobile device described in claim 12, the processing unit further controls the display unit to output a recognition confidence. For the mobile device described in item 12 of the scope of patent application, the image capturing unit continuously captures the spot photos. In the mobile device described in item 12 of the scope of patent application, the local photo is presented on the display unit. The mobile device described in item 12 of the scope of patent application, wherein if the value of the electronic compass meets the first preset condition and the value of the gravity sensor meets For the second preset condition, the processing unit controls the display unit to prompt information about the correct shooting direction. For the mobile device described in item 12 of the scope of patent application, the first predetermined condition is a predetermined numerical interval. For the mobile device described in item 12 of the scope of patent application, the second predetermined condition is a predetermined numerical interval.

Images