Systems and methods for identifying lights sources at an airport
Systems and methods are provided for identifying light sources at an airport. Light sources are detected in a vehicle camera image. A region of interest (ROI) is determined for each light source based on the location of the light source in the image. A distance and relative angle are determined between each light source and a vehicle camera location. A gray scale version of each ROI is generated based on pre-defined relationships between intensities of red, green, and blue colors in the image and gray color intensities. The gray scale version is compared with pre-defined color histograms to determine a color associated with each light source. Each histogram corresponds to a gray-scale equivalent of an associated color. Context data associated with the image is determined. A light source type is assigned to each of the light sources based on the color of the light source and the context data.
1. A system comprising:
a processor; and
a memory, the memory comprising instructions that upon execution by the processor, cause the processor to:
detect a plurality of light sources in a first image received from a vehicle camera;
determine a region of interest for each of the plurality of light sources based on a location of the light source in the first image;
determine a distance between each of the plurality of light sources and a location of the vehicle camera and a relative angle between each of the plurality of light sources and the location of the vehicle camera;
generate a gray scale version of each region of interest based on pre-defined relationships between intensities of a red color, a green color and a blue color in the first image and gray intensities associated with each of the intensities of the red color, the green color, and the blue color;
compare the gray scale version of each region of interest with pre-defined color specific histograms to determine a color associated with each of the plurality of light sources, wherein each pre-defined color specific histogram corresponds to a gray-scale equivalent of an associated color;
determine context data associated with the first image; and
assign a light source type to each of the plurality of light sources based on the color of the light source and the context data.
2. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
categorize each of a plurality of pixels in the first image into one of a red color channel, a blue color channel, and a green color channel;
detect high pixel intensity regions associated with the red color channel, the blue color channel, and the green color channel;
detect a center pixel for each detected high intensity pixel region;
identify the high pixel intensity regions where a circular pattern of pixel intensity decreases with increasing radius with respect to the associated center pixel; and
define the identified high pixel intensity regions as the plurality of light sources in the first image.
3. The system of claim 2 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to employ a Range Intensity Normalized Gaussian (RING) filter to identify the high pixel intensity regions where the circular pattern of pixel intensity decreases in accordance with a Gaussian probability density function distribution model.
4. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to determine the distance between each of the plurality of light sources and the location of the vehicle camera using a neural network model trained using training camera images including light sources.
5. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to determine the distance between each of the plurality of light sources and the location of the vehicle camera based on distance data received from at least one sensor at a vehicle associated with the vehicle camera.
6. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to determine whether a first one of the plurality of light sources is in one of an ON state or an OFF state based on an analysis of the region of interest associated with the first one of the plurality of light sources in successive images received from the vehicle camera, the successive images including the first image.
7. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to determine a blinking frequency associated with a blinking state of a first one of the plurality of light sources based on a blink rate histogram analysis of the region of interest associated with the first one of the plurality of light sources in successive images received from the vehicle camera, the successive images including the first image.
8. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to determine first context data associated with a first one of the plurality of light sources, wherein the first context data comprises at least one of a vehicle type of a vehicle associated with the vehicle camera, a vehicle location of the vehicle, and a location of the first one of the plurality of light sources.
9. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
determine whether a first one of the plurality of light sources in the first image is associated with a leading vehicle; and
predict a future position of the leading vehicle based on a first distance between the first one of the plurality of light sources and the vehicle camera, a first relative angle between the first one of the plurality of light sources and the location of the vehicle camera, a first light source type of the first one of the plurality of light sources as defined in successive images received from the vehicle camera based on the determination.
10. The system of claim 9 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to generate a recommended vehicle action based on the future position of the leading vehicle, the first light source type of the first one of the plurality of light sources, and whether the first one of the plurality of light sources is in one of an ON state or an OFF state.
11. The system of claim 9 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to implement a vehicle action based on the future position of the leading vehicle, the first light source type of the first one of the plurality of light sources, and whether the first one of the plurality of light sources is in one of an ON state or an OFF state.
12. The system of claim 1 , wherein the vehicle camera is associated with an aircraft.
13. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to provide a recommendation to implement vehicle action based on the context data wherein the context data comprises situational awareness data.
14. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
receive the image from the vehicle camera associated with an aircraft;
identify a subset of the plurality of light sources as being airport light sources;
determine whether an onboard pilot microphone malfunction has occurred; and
auto-initiate communication with an air traffic controller (ATC) based on the determination.
15. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
receive the image from the vehicle camera associated with an aircraft;
identify a subset of the plurality of light sources as a plurality of airfield ground light (AGL) light sources; and
detect a malfunction of a first AGL light source based on an analysis of the identified plurality of AGL light sources in the image.
16. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
receive the image from the vehicle camera associated with an aircraft;
identify a subset of the plurality of light sources as a plurality of precision approach path indicator (PAPI) lights; and
compute an aircraft altitude offset based on locations of the plurality of PAPI lights with respect to a location of the aircraft.
17. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
receive the image from the vehicle camera associated with an aircraft;
identify a subset of the plurality of light sources as a plurality of taxiway lights on a runway; and
implement auto-taxing assistance of the aircraft on the runway based on locations of the plurality of taxiway lights with respect to a location of the aircraft.
18. The system of claim 1 , wherein the memory comprises further instructions that upon execution by the processor, cause the processor to:
identify a subset of the plurality of light sources as a plurality of aircraft navigational lights sources of an aircraft;
identify an aircraft type and an aircraft location of the aircraft based on locations of the plurality of aircraft navigational lights source on the aircraft; and
provide a recommendation to implement vehicle action based the identified aircraft type and the aircraft location.
19. A method comprising:
detecting a plurality of light sources in a first image received from a vehicle camera;
determining a region of interest for each of the plurality of light sources based on a location of the light sources in the first image;
determining a distance between each of the plurality of light sources and a location of the vehicle camera and a relative angle between each of the plurality of light sources and the location of the vehicle camera;
generating a gray scale version of each region of interest based on pre-defined relationships between intensities of a red color, a green color and a blue color in the first image and gray intensities associated with each of the intensities of the red color, the green color, and the blue color;
comparing the gray scale version of each region of interest with pre-defined color specific histograms to determine a color associated with each of the plurality of light sources, wherein each pre-defined color specific histogram corresponds to a gray-scale equivalent of an associated color;
determining context data associated with the first image; and
assigning a light source type to each of the plurality of light sources based on the color of the light source and the context data.
20. The method of claim 1 , further comprising:
categorizing each of a plurality of pixels in the first image into one of a red color channel, a blue color channel, and a green color channel;
detecting high pixel intensity regions associated with the red color channel, the blue color channel, and the green color channel;
detecting a center pixel for each detected high intensity pixel region;
identifying the high pixel intensity regions where a circular pattern of pixel intensity decreases with increasing radius with respect to the associated center pixel; and
defining the identified high pixel intensity regions as the plurality of light sources in the first image.