US 12,249,080 B2
Method for detecting and tracking orbiting objects by on-board processing of images acquired by a space platform
Fabio Curti, Rome (IT); Daniele Luchena, Rome (IT); Marco Moriani, Rome (IT); and Dario Spiller, Rome (IT)
Assigned to Arca Dynamics Societa' A Responsabilita' Limitata Semplificata, Rome (IT)
Appl. No. 17/614,825
Filed by Arca Dynamics Societa' A Responsabilita' Limitata Semplificata, Rome (IT)
PCT Filed May 29, 2020, PCT No. PCT/IB2020/055127
§ 371(c)(1), (2) Date Nov. 29, 2021,
PCT Pub. No. WO2020/240501, PCT Pub. Date Dec. 3, 2020.
Claims priority of application No. 102019000007509 (IT), filed on May 29, 2019.
Prior Publication US 2022/0245833 A1, Aug. 4, 2022
Int. Cl. G06T 7/246 (2017.01); G06T 7/00 (2017.01); B64G 1/36 (2006.01)
CPC G06T 7/248 (2017.01) [G06T 7/97 (2017.01); B64G 1/361 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/20221 (2013.01)] 11 Claims
OG exemplary drawing
 
1. Method (1) for detecting and tracking space objects, comprising:
acquiring pairs of images by means of an optical sensor installed on a space platform in orbit around the earth, wherein each pair of images comprises a respective first image and a respective second image acquired immediately after said respective first image; and,
for each new pair of images acquired, carrying out a merging operation and a tracking operation;
wherein the merging operation includes:
identifying, in the first image of the new pair of images, first clusters of pixels related to potential space objects imaged in said first image;
identifying, in the second image of the new pair of images, second clusters of pixels related to potential space objects imaged in said second image; and
detecting pairs of clusters of pixels each comprising a respective first cluster of pixels and a respective second cluster of pixels both related to one and the same potential space object imaged in the new pair of images, wherein said respective first and second clusters of pixels are merged together;
wherein the tracking operation includes detecting, from among the pairs of clusters of pixels merged together, pairs of clusters of pixels related to space objects detected in one or more previous pairs of images, on the basis of stored tracking information items indicative of last positions and average velocities of space objects detected in one or more previous pairs of images; and
wherein the merging operation further includes computing:
a respective centroid of each first cluster of pixels and of each second cluster of pixels; and
a displacement velocity of the centroids of the pairs of clusters of pixels merged together;
wherein the stored tracking information items are indicative of a last centroid position and of an average velocity related to a given space object detected in one or more previous pairs of images;
wherein the tracking operation further includes:
estimating a new centroid position for the given space object based on
the last centroid position and the average velocity related to said given space object, and
a given time interval elapsed between acquisition of the image for which said last centroid position was computed, and acquisition of the first image of the new pair of images;
detecting, from among the pairs of clusters of pixels merged together, a pair of clusters of pixels related to the given space object for which the centroid of the respective first cluster of pixels is within a search area that is centered on the new centroid position estimated and that has predefined dimensions; and
updating/integrating the stored tracking information items based on the displacement velocity and the positions of the centroids of the pair of clusters of pixels related to the given space object.