Systems, apparatuses, and methods for dynamic filtering of high intensity broadband electromagnetic waves from image data from a sensor coupled to a robot
Systems, apparatuses, and methods for dynamic filtering of high intensity broadband electromagnetic waves in image data from a sensor of a robot are disclosed herein. According to at least one non-limiting exemplary embodiment, sunlight or light emitted from nearby fluorescent lamps may cause a robot to generate false positives of objects nearby the robot as the light may be of high intensity and large bandwidth. These false positives may cause a robot to get stuck or navigate without use of a camera sensor, which may be unsafe.
1. A robotic system for traveling along a trajectory, comprising:
a memory having computer readable instructions stored thereon; and
at least one processor device configurable to execute the computer readable instructions to,
obtain at least one image captured by a sensor coupled to the robotic system at a first time instance at a first location of a route traveled by the robotic system;
detect at least one region of pixels within the respective at least one image comprising a mean intensity value exceeding a dynamic lighting intensity threshold;
apply at least one mask to the at least one region such that a false positive is not detected as the robotic system travels the route at the first location at a subsequent second time instance; and
maneuver the robotic system past the first location with an application of the at least one mask such that the false positive is no longer detected at the first location.
2. The robotic system of claim 1 , wherein the at least one processor device is configurable to execute the computer readable instructions to add the at least one image to an image frame queue such that the image frame queue comprises a plurality of images.
3. The robotic system of claim 1 , wherein the dynamic light intensity threshold is based on a mean light intensity value of a plurality of images captured by the sensor in real-time, the plurality of images are present in an image frame queue, the mean light intensity value is configurable to adjust in real-time as additional images are captured by the sensor as the robotic system travels the route, adjusting of the mean light intensity value corresponds to adjusting of the dynamic light intensity threshold.
4. The robotic system of claim 1 , wherein the at least one processor device is configurable to execute the computer readable instructions to store the at least one image in an image frame queue without applying the at least one mask if the mean light intensity value does not exceed the dynamic lighting intensity threshold.
5. The robotic system of claim 1 , wherein,
the at least one mask comprises a pixel-wise determination of the false positive based on the dynamic light intensity threshold, the false positive comprising of a plurality of pixels, and
the determined plurality of pixels of the false positive are masked to eliminate the false positive from the at least one image.
6. The robotic system of claim 1 , wherein the at least one processing device is coupled to an image frame queuing unit, the image frame queuing unit configurable to determine intensity values of respective images of the plurality of images within an image frame queue.
7. The robotic system of claim 6 , wherein the image frame queuing unit comprises a respective intensity value corresponding to a respective image of the plurality of images stored in the image frame queue.
8. The robotic system of claim 7 , wherein the image frame queuing unit comprises the mean intensity value that is an average of the respective intensity value corresponding to the respective image of the plurality of images.
9. The robotic system of claim 1 , wherein the at least one processing device is configurable to execute the computer readable instructions to remove the at least one mask after the robotic system travels past the first location.
10. The robotic system of claim 1 , wherein the false positive detected is a representation of high intensity broadband electromagnetic waves captured by the at least one sensor.
11. A method for traveling along a trajectory by a robot, comprising:
at obtaining at least one image captured by a sensor coupled to the robotic system at a first time instance at a first location of a route traveled by the robot;
detecting at least one region of pixels within the respective at least one image comprising a mean intensity value exceeding a dynamic lighting intensity threshold;
applying at least one mask to the at least one region such that a false positive is not detected as the robot travels the route at the first location at a subsequent second time instance; and
maneuvering the robot past the first location with the at least one mask such that the false positive is no longer detected at the first location.
12. The method of claim 11 , further comprising:
adding the at least one image to an image frame queue such that the image frame queue comprises a plurality of images.
13. The method of claim 11 , wherein the dynamic light intensity threshold is based on a mean light intensity value of a plurality of images captured by the sensor in real-time, the plurality of images present in an image frame queue, the mean light intensity value is configurable to adjust in real-time as additional images are captured by the sensor as the robot travels the route, adjusting of the mean light intensity value corresponds to adjusting of the dynamic light intensity threshold.
14. The method of claim 11 , further comprising:
storing the at least one image in an image frame queue without applying the at least one mask if the mean light intensity value does not exceed the dynamic lighting intensity threshold.
15. The method of claim 11 , wherein,
the at least one mask comprises a pixel-wise determination of the false positive based on the dynamic light intensity threshold, the false positive comprises of a plurality of pixels, and
the determined plurality of pixels of the false positive is masked to eliminate the false positive from the at least one image.
16. The method of claim 11 , further comprising:
determining intensity values of respective images of the plurality of images within an image frame queue.
17. The method of claim 16 , further comprising:
a respective intensity value corresponding to a respective image of the plurality of images stored in the image frame queue.
18. The method of claim 17 , wherein the mean intensity value that is an average of the respective intensity value corresponding to the respective image of the plurality of images.
19. The method of claim 11 , further comprising:
removing the at least one mask after the robot travels past the first location.
20. The method of claim 11 , wherein the false positive detected is a representation of high intensity broadband electromagnetic waves captured by the at least one sensor.
21. A non-transitory computer readable medium having computer readable instructions stored thereon that when executed by at least one processor device configure the at least one processor device to,
maneuver, at a first time instance, a robot along a first trajectory along a path;
detect, at the first time instance, a high intensity broadband electromagnetic wave by a sensor coupled to the robot while the robot maneuvers along the first trajectory; and
maneuver, at the first time instance, the robot along a second trajectory along the path upon identifying a false positive along the first trajectory due to detection of the high intensity broadband electromagnetic wave, the second trajectory being different from the first trajectory.
22. The non-transitory computer readable medium of claim 21 , wherein the at least one processor device is further configurable to execute the computer readable instructions to,
disregard, at a second time instance, the false positive identified along the first trajectory during the first time instance, and
maneuver, at the second time instance, the robot along the first trajectory such that the false positive identified along the first trajectory during the first time instance is disregarded.
23. The non-transitory computer readable medium of claim 21 , wherein the at least one processor device is further configurable to execute the computer readable instructions to,
apply at least one mask such that the false positive is not detected as the robot travels the first trajectory along the path.
24. The non-transitory computer readable medium of claim 21 , wherein the high intensity broadband electromagnetic wave corresponds to either sunlight, flashlight, or floodlight.
25. The non-transitory computer readable medium of claim 21 , wherein the identifying of the false positive is based on the sensor detecting the high intensity broadband electromagnetic wave.