System and method for detecting and tracking an object
Inventors
Jafek, Benjamin • Tumuluru, Samvruta
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11961274-B2
Publication Date
2024-04-16
Expiration Date
Abstract
A method includes receiving a first image that is captured at a first time. The method also includes detecting a location of a first object in the first image. The method also includes determining a region of interest based at least partially upon the location of the first object in the first image. The method also includes receiving a second image that is captured at a second time. The method also includes identifying the region of interest in the second image. The method also includes detecting a location of a second object in a portion of the second image that is outside of the region of interest.
Core Innovation
The invention addresses object detection and tracking using a camera on an aircraft by identifying a first image captured at a first time and detecting a location of a first moving object in the first image. A region of interest is determined based at least partially upon the location of the first moving object, and the region of interest comprises a plurality of pixels having a probability greater than a predetermined threshold for the first moving object to be located therein at a second time. The region of interest guides processing of a second image captured at the second time.
The invention identifies the region of interest in the second image and detects the location of a second moving object in a portion of the second image that is outside of the region of interest. This constrains where a moving object is searched, such that detection within the region of interest is used for the first moving object while a second moving object is detected outside the region of interest. The document further specifies that the second time is after the first time.
The invention further extends to using a hidden Markov model for detection and to avoid re-detection. A hidden Markov model is used to detect the location of the first object and to detect a first object in the region of interest in the second image, while the location of a second object is detected in the portion of the second image outside the region of interest using the hidden Markov model to avoid re-detection of the first object. The invention also predicts a trajectory of the first and second objects based on the detected locations across the first and second images and causes the first aircraft to navigate based at least partially upon the predicted trajectory.
In an aircraft-in-flight context, the invention includes representing the first object as five or fewer pixels in the first image and detecting the first object as a second aircraft in flight. The region of interest is again defined with pixel probabilities greater than a predetermined threshold for the object at a second time after the first time. Trajectory prediction and navigation are based at least partially upon the location of the first object in the first image, the location of the first object in the second image, and the location of the second object in the second image.
Claims Coverage
The independent claims cover a method and computing systems that perform time-separated image identification, probability-based region-of-interest constrained detection inside the region of interest for the first object and outside the region of interest for a second object, hidden Markov model-based detection, and trajectory prediction to cause aircraft navigation. Each independent claim includes core inventive features tied to region-of-interest probability and outside-ROI detection.
Probability-based region of interest for time-separated object prediction
Determining a region of interest based at least partially upon the location of the first moving object in the first image, wherein the region of interest comprises a plurality of pixels in the first image, and wherein each pixel in the region of interest has a probability that is greater than a predetermined threshold that the first moving object will be located therein at a second time.
Outside-ROI detection to detect a second moving object
Detecting a location of a second moving object in a portion of the second image that is outside of the region of interest.
Hidden Markov model detection for ROI and outside-ROI regions
Detecting a location of a first object in the first image using a hidden Markov model; and detecting the location of the second object in a portion of the second image that is outside of the region of interest using the hidden Markov model to avoid re-detection of the first object.
Trajectory prediction and aircraft navigation based on detected objects
Predicting a trajectory of the first and second objects based at least partially upon the location of the first object in the first image, the location of the first object in the second image, and the location of the second object in the second image; and causing the first aircraft to navigate based at least partially upon the trajectory of the first and second objects.
Object represented as five or fewer pixels in the first image
Detecting a location of a first object in the first image using a hidden Markov model, wherein the first object is a second aircraft in flight, and wherein the first object is represented as five or fewer pixels in the first image.
Across the independent claims, the inventive coverage centers on constraining tracking using a probability-based region of interest across first and second images, detecting additional moving objects in portions of the second image outside the region of interest to separate detections and avoid re-detection, and using a hidden Markov model for detection while predicting object trajectories to cause aircraft navigation.
Stated Advantages
Avoid re-detection of the first object by detecting a second object outside the region of interest using the hidden Markov model.
Documented Applications
Aircraft-centric object detection and tracking to support trajectory prediction and navigating an aircraft based on the predicted trajectories.
Detecting objects represented as aircraft in flight using a camera on a first aircraft, including region-of-interest constrained detection and navigation based on predicted trajectories.
Interested in licensing this patent?