Back to results
US 10,650,553 B2Granted
Method of image processing and image processing device
AssigneeIntel IP Corporation
InventorsKoba Natroshvili, Okan Köse
Life of this patent
Estimating the term…
Abstract
A method of image processing is provided. The method may include: determining a candidate tuple from at least two images that are taken at different times, wherein the candidate tuples are determined using at least odometry sensor information. The couple of subsequent images have been detected by a moving image sensor moved by a vehicle. The odometry sensor information is detected by a sensor moved by the vehicle. The method may further include classifying the candidate tuples into a static tuple or a dynamic tuple. The static tuple represents a static object within the couple of subsequent images, and the dynamic tuple represents a moving object within the couple of subsequent images.
Claims19
A method of image processing, the method comprising:
- determining a candidate tuple from at least two images that are taken at different times, wherein the candidate tuples are determined using at least odometry sensor information, and wherein the couple of subsequent images have been detected by a moving image sensor moved by a vehicle, wherein the odometry sensor information is detected by a sensor moved by the vehicle;
- and
- classifying the candidate tuples into a static tuple or a dynamic tuple, wherein the static tuple represents a static object within the couple of subsequent images, and the dynamic tuple represents a moving object within the couple of subsequent images, and wherein the candidate tuple is classified into a static tuple or a dynamic tuple by comparing a first angle of translation of the vehicle as estimated from the odometry sensors in the vehicle with a second angle of translation of the vehicle as estimated from the candidate tuples.
An image processing device, comprising:
- a determiner configured to determine a candidate tuple from at least two images that are taken at different times, wherein the candidate tuples are determined using at least odometry sensor information, and wherein the couple of subsequent images have been detected by a moving image sensor moved by a vehicle, wherein the odometry sensor information is detected by a sensor moved by the vehicle;
- and
- a classifier configured to classify the candidate tuples into a static tuple or a dynamic tuple, wherein the static tuple represents a static object within the couple of subsequent images, and the dynamic tuple represents a moving object within the couple of subsequent images, and wherein the classifier is further configured to classify the candidate tuple into a static tuple or a dynamic tuple by comparing a first angle of translation of the vehicle as estimated from the odometry sensors in the vehicle, with a second angle of translation of the vehicle as estimated from the candidate tuples.
An image processing device, comprising:
- at least one image sensor coupled to a vehicle, wherein the at least one image sensor is configured to take successive pictures of a pre-determined region;
- at least one odometer sensor coupled to the vehicle and configured to record information on the vehicle movements;
- at least one antenna configured to receive communication about other objects and vehicles in the predetermined area, wherein the information about objects and vehicles includes position information and direction of motion information;
- a memory configured to store at least one map of the predetermined region indicating the position of objects and vehicles in the predetermined region;
- a processor configured to
- detect at least one object using information from the image sensors, wherein the information comprises at least two images that are taken at different times, each of the at least two successive images including a respective a candidate tuple corresponding to the at least one object, wherein the candidate tuples are determined using at least odometry sensor information provided by the at least one odometer sensor coupled to the vehicle, and wherein the at least two successive images have been detected by the at least one image sensor coupled to the vehicle;
- determine whether the object is a static object or a dynamic object by classifying the candidate tuples into a static tuple or a dynamic tuple, wherein the static tuple represents a static object within the at least two images, and the dynamic tuple represents a moving object within the at least two successive images, and wherein the candidate tuple is classified into a static tuple or a dynamic tuple by comparing a first angle of translation of the vehicle as estimated from the at least one odometer sensor coupled to the vehicle with a second angle of translation of the vehicle as estimated from the candidate tuples;
- compute direction of the object if the object is determined as a dynamic object;
- fuse the information received from the at least one image sensor and from the at least one antenna;
- generate a map of the predetermined region including information about the static object, dynamic object, and the direction of dynamic object.
7 more elements
Dependencies are inferred from the claim text.