Method and system for modifying a beacon light source for use in a light based positioning system
In one aspect, the present disclosure related to a method for modifying a beacon light source for use in a light-based positioning system. In some embodiments, the method includes selecting a modulation scheme for the light source, determining a duty cycle for the light source based on the modulation scheme, the duty cycle having a proportion of time the light source is in an on state and a corresponding proportion of time the light source is in an off state, the proportion of time the light source is in an off state resulting in reduced luminosity of the light source, and supplying additional power to the light source to compensate for the reduced luminosity of the light source by the duty cycle.
1. A method comprising:
in response to light illuminating of a scene modulated by a plurality of light sources, acquiring an image of the illuminated scene by an image sensor of a mobile device, wherein:
pixels of the image are: (i) dependent on illumination of the plurality of light sources; and (ii) scene dependent such that at least two pixels in a same row of each image exposed to the illumination have a different brightness due to the scene of the image;
processing the image to isolate the plurality of light sources, wherein processing the image includes:
applying a background subtraction algorithm to the image such that background of the scene is removed leaving only a stripe image of the modulated illumination that is independent of the scene; and
after background subtraction of the image, calculating a fast Fourier transform (FFT) of the stripe image such that a tone modulating the illumination is detected by performing a frequency analysis of the stripe image to recover the tone; and identifying a physical location of the mobile device based on the processed image.
2. The method of claim 1 , wherein performing the frequency analysis of the stripe image includes:
applying a peak frequency detector to the calculated FFT to identify the tone as a frequency of a periodic component in the calculated FFT that corresponds to the detected tone.
3. The method of claim 2 , further comprising:
when applying the peak frequency detector to the calculated FFT, applying a high pass filter to the calculated FFT to remove a direct current (DC) offset component to detect the tone frequency.
4. The method of claim 3 , further comprising:
when applying the peak frequency detector to the calculated FFT, applying a low pass filter to remove noise in the calculated FFT to detect the tone frequency.
5. The method of claim 1 , wherein applying the background subtraction algorithm includes:
generating an averaged image by averaging a plurality of frames such that a banding effect due to the modulated illumination of the plurality of light sources is reduced to a constant value added on to all pixels of the averaged image; and
subtracting the averaged image from the image to reveal the stripe image that is independent of the scene.
6. The method of claim 5 , further comprising:
(a) after background subtraction but before calculating the FFT, averaging together each pixel in a given row of the averaged image to compute an average pixel row value; and
(b) assigning the average pixel row value to every pixel in the given row.
7. The method of claim 6 , further comprising:
creating a row averaged image with a column that includes the average pixel row value of each row by repeating (a) and (b) for each row; and
assigning the row averaged image to the stripe image.
8. The method of claim 1 , wherein applying the background subtraction algorithm further includes:
applying phase correlation to detect scene motion between a plurality of frames;
upon detecting the scene motion between the plurality of frames, transforming the plurality of frames by shifting each of the plurality of frames to cause each frame to overlie a previous frame;
generating an averaged image by averaging a plurality of frames such that a banding effect due to the time-varying illumination of the VLC light sources is reduced to a constant value added on to all pixels of the averaged image; and
subtracting the averaged image from the image to reveal the stripe image that is independent of the scene.
9. The method of claim 1 , wherein applying the background subtraction algorithm further includes:
applying a motion compensator selected from a group consisting of a kernel density estimator, a mean-shift based estimator, and an eigenbackground.
10. The method of claim 1 , wherein the stripe image includes alternating black/white stripes that represent a discrete time-domain representation of a transmitted signal from the plurality of light sources.
11. The method of claim 1 , wherein identifying the physical location of the mobile device includes:
correlating the detected tone to a light source identifier associated with one of the plurality of light sources.
12. The method of claim 11 , further comprising:
calculating an estimate of the physical location of the mobile device based on location of the correlated light source identifier.
13. The method of claim 12 , wherein the location of the correlated VLC light source identifier is global positioning system (GPS) coordinates mapped into an indoor space.
14. The method of claim 1 , wherein identifying the physical location of the mobile device includes:
transmitting the detected tone to a server; and
in response to transmitting the detected tone, receiving a calculation of the physical location of the mobile device based on the detected tone from the server.
15. The method of claim 1 , wherein the modulated illumination of the light sources is imparted by controlling a frequency of a periodic tone driving the plurality of light sources to distinguish a respective light source from other nearby light sources.
16. There method of claim 1 :
wherein the image sensor is a rolling shutter complementary metal-oxide semiconductor (CMOS) image sensor; and
wherein the scene is an area within view of the image sensor of the mobile device.
17. The method of claim 1 , further comprising:
acquiring the image of the illuminated scene by the image sensor only upon detecting a change in an accelerometer reading of the mobile device indicating the mobile device is in view of modulated illumination of the plurality of light sources.