Apple patent is for an automobile imaging system for object detection

Let the Apple Car rumors roll on. Apple has been granted a patent (number 10,397,497) for a “solar invariant imaging system for object detection” in an automobile.

In the patent data, the tech giant notes that vehicle guidance systems, autonomous control systems, and passengers viewing a surrounding environment while operating a vehicle can all rely on imaging systems to improve vehicle operation. Imaging systems can identify objects and track movement of the objects that may intersect an intended vehicle path. 

Depending on the dynamic range of the imaging system and the optical set up used to capture images of the surrounding environment (e.g., lens set, housing, protective window, type of image sensor, etc.), some types of radiation can blur or blind groups of pixels on an image sensor associated with the imaging system. Blurred or blinded groups of pixels can result in bright spots or blind spots within captured images, for example, when the imaging systems operate solely within a visible spectrum, making identification of objects within these regions difficult. Apple makes to change this — though, as I’ve said before, I don’t believe the company will actually start manufacturing vehicles, but will work with current car manufacturers on future automobiles.

Here’s the summary of the patent: “An object detection system includes an infrared light source and an imaging system that generates an image from a portion of the infrared spectrum characterized by full absorption of solar radiation. A control system detects an object using the image, determines a command based on a location of the object, and sends a command to one or more vehicle systems. 

“Another object detection system includes an imaging system that generates a first image based on a visible spectrum and a second image based an infrared spectrum. A control system receives a disparity indication associated with object detection and sends a command to one or more vehicle systems to implement a disparity response based on the disparity indication. The disparity indication includes information that an object is not detected within the first image and that the object is detected within the second image.”