18358800. OBJECT DETECTION IN DYNAMIC LIGHTING CONDITIONS simplified abstract (QUALCOMM Incorporated)
OBJECT DETECTION IN DYNAMIC LIGHTING CONDITIONS
Organization Name
Inventor(s)
Louis Joseph Kerofsky of San Diego CA (US)
Shihao Shen of San Jose CA (US)
OBJECT DETECTION IN DYNAMIC LIGHTING CONDITIONS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18358800 titled 'OBJECT DETECTION IN DYNAMIC LIGHTING CONDITIONS
Simplified Explanation
- The patent application describes systems and methods for detecting objects in dynamic lighting conditions using image processing techniques. - The method involves obtaining two images of an object in different positions in an environment and determining the movement of the object between the images using an optical flow engine trained on augmented training data. - The augmented training data includes noise associated with low ambient lighting conditions, motion blur due to exposure of an image sensor in low lighting conditions, and brightness variations. - The object and/or the computing device may move between the two images.
Potential Applications
- Surveillance systems - Autonomous vehicles - Robotics - Augmented reality
Problems Solved
- Detection of objects in challenging lighting conditions - Accurate tracking of moving objects - Reduction of false positives in object detection
Benefits
- Improved object detection in dynamic lighting conditions - Enhanced accuracy in tracking object movement - Increased reliability of surveillance and monitoring systems
Original Abstract Submitted
Disclosed are systems, apparatuses, processes, and computer-readable media to detect objects in dynamic lighting conditions. A method of processing image data includes obtaining, at a computing device, a first image of an object at a first position in an environment, obtaining, at the computing device, a second image of the object at a second position in the environment, and determining, at the computing device, movement of the object in the first image and the second image at least in part using an optical flow engine, wherein the optical flow engine is trained based on augmented training data generated using at least one of noise associated with low ambient lighting conditions, noise associated with motion blur due to exposure of an image sensor in low ambient lighting conditions, or brightness variations. The object and/or the computing device may move between the first image and the second image.