Identifying Defects in Optical Detector Systems Based on Extent of Stray Light

Abstract

Example embodiments relate to identifying defects in optical detector systems based on extent of stray light. An example embodiment includes a method. The method includes capturing, using an optical detector system, an image of a scene that includes a bright object. The method also includes determining a location of the bright object within the image. Further, the method includes determining, based on the location of the bright object within the image, an extent of stray light from the bright object that is represented in the image. In addition, the method includes determining, by comparing the extent of stray light from the bright object that is represented in the image to a predetermined threshold extent of stray light, whether one or more defects are present within the optical detector system. The predetermined threshold extent of stray light corresponds to an expected extent of stray light.

What is claimed is: 
     
         1 . A method comprising:
 capturing, using a camera, an image of a scene that includes a bright object;   determining a location of the bright object within the image;   determining, based on the location of the bright object within the image, an extent of stray light from the bright object that is represented in the image; and   determining, by comparing the extent of stray light from the bright object that is represented in the image to a predetermined threshold extent of stray light, whether one or more optical imperfections are present on an imaging optic of the camera,   wherein the predetermined threshold extent of stray light corresponds to an expected extent of stray light,   wherein determining the extent of stray light from the bright object that is represented in the image comprises determining how many streaks or stripes of stray light are present within the image, and   wherein determining whether one or more optical imperfections are present on the imaging optic of the camera comprises determining, based on how many streaks or stripes of stray light are present within the image, whether optical imperfections due to a windshield wiper or water dripping across the imaging optic of the camera are present.   
     
     
         2 . The method of  claim 1 , wherein the location of the bright object within the image is determined based on a geographical location of the camera. 
     
     
         3 . The method of  claim 1 , wherein the location of the bright object within the image is determined based on an orientation of the camera. 
     
     
         4 . The method of  claim 1 , wherein the location of the bright object within the image is determined based on a time of day. 
     
     
         5 . The method of  claim 1 , wherein the location of the bright object within the image is determined based on a calendar date. 
     
     
         6 . The method of  claim 1 , wherein the extent of stray light from the bright object that is represented in the image is also determined based on a predetermined angular size of the sun when observed from Earth. 
     
     
         7 . The method of  claim 1 , further comprising cleaning, repairing, recalibrating, replacing, or decommissioning the camera in response to determining that one or more optical imperfections are present on the imaging optic of the camera. 
     
     
         8 . The method of  claim 1 , wherein determining whether one or more optical imperfections are present on the imaging optic of the camera comprises determining whether an imaging optic of the camera is unclean or occluded by debris. 
     
     
         9 . The method of  claim 8 , wherein determining whether the imaging optic of the camera is unclean or occluded by debris comprises determining whether:
 the camera is facing backwards on a vehicle traveling forwards;   the camera is facing forwards on a vehicle traveling forwards;   the camera is facing backwards on a vehicle traveling backwards; or   the camera is facing forwards on a vehicle traveling backwards.   
     
     
         10 . The method of  claim 1 , wherein determining the extent of stray light from the bright object that is represented in the image comprises:
 identifying each pixel within the image that has an intensity value above a predetermined threshold intensity value; and   generating a derivative image by converting all those pixels having an intensity value greater than the predetermined threshold intensity value to white and all those pixels having an intensity value less than or equal to the predetermined threshold intensity value to black.   
     
     
         11 . The method of  claim 10 , wherein determining the extent of stray light from the bright object that is represented in the image further comprises determining a maximum radius of contiguous white pixels within the derivative image relative to a center of the location of the bright object. 
     
     
         12 . The method of  claim 10 , wherein determining the extent of stray light from the bright object that is represented in the image further comprises identifying a secondary region of stray light that is separated by one or more black pixels from a primary region of stray light within the derivative image. 
     
     
         13 . The method of  claim 10 , wherein the predetermined threshold intensity value is based on a rejection ratio relative to an inherent luminance level of the bright object. 
     
     
         14 . The method of  claim 1 , wherein determining how many streaks or stripes of stray light are present within the image comprises applying a machine-learning algorithm that is trained to perform object recognition in images. 
     
     
         15 . The method of  claim 1 ,
 wherein determining the extent of stray light from the bright object that is represented in the image comprises identifying an existence of diffuse veiling glare within the image, and   wherein determining if one or more optical imperfections are present on the imaging optic of the camera comprises identifying, based on the diffuse veiling glare within the image, that dust or water is present on the imaging optic of the camera.   
     
     
         16 . The method of  claim 1 ,
 wherein determining the location of the bright object within the image comprises capturing a baseline image of the scene at a reduced intensity,   wherein the baseline image is captured at a substantially similar perspective relative to the scene as the image, and   wherein the baseline image is captured using a secondary camera that has a different dynamic range than the camera or using the camera with a modified dynamic range.   
     
     
         17 . A non-transitory, computer-readable medium having instructions stored thereon, wherein the instructions, when executed by a processor, cause the processor to execute a method comprising:
 receiving an image of a scene that includes a bright object, wherein the image was captured using a camera;   determining a location of the bright object within the image;   determining, based on the location of the bright object within the image, an extent of stray light from the bright object that is represented in the image; and   determining, by comparing the extent of stray light from the bright object that is represented in the image to a predetermined threshold extent of stray light, whether one or more optical imperfections are present on an imaging optic of the camera,   wherein the predetermined threshold extent of stray light corresponds to an expected extent of stray light,   wherein determining the extent of stray light from the bright object that is represented in the image comprises determining how many streaks or stripes of stray light are present within the image, and   wherein determining whether one or more optical imperfections are present on the imaging optic of the camera comprises determining, based on how many streaks or stripes of stray light are present within the image, whether optical imperfections due to a windshield wiper or water dripping across the imaging optic of the camera are present.   
     
     
         18 . A camera comprising:
 an imaging optic;   an image sensor configured to receive light from a scene via the imaging optic; and   a controller configured to perform a monitoring routine, wherein the monitoring routine comprises:
 capturing, using the camera, an image of a scene that includes a bright object; 
 determining a location of the bright object within the image; 
 determining, based on the location of the bright object within the image, an extent of stray light from the bright object that is represented in the image; and 
 determining, by comparing the extent of stray light from the bright object that is represented in the image to a predetermined threshold extent of stray light, whether one or more optical imperfections are present on the imaging optic, 
 wherein the predetermined threshold extent of stray light corresponds to an expected extent of stray light, 
 wherein determining the extent of stray light from the bright object that is represented in the image comprises determining how many streaks or stripes of stray light are present within the image, and 
 wherein determining whether one or more optical imperfections are present on the imaging optic of the camera comprises determining, based on how many streaks or stripes of stray light are present within the image, whether optical imperfections due to a windshield wiper or water dripping across the imaging optic of the camera are present. 
   
     
     
         19 . The camera of  claim 18 , wherein the controller is configured to perform the monitoring routine in response to a request from a computing device. 
     
     
         20 . The camera of  claim 19 , wherein the computing device is a server computing device configured to monitor states of multiple cameras in a fleet of cameras.