Method and device for fast detection of repetitive structures in the image of a road scene
View Patent ↗A method of processing images allowing the identification of at least one same point appearing on each image of a sequence of images, the sequence of images including a first image, a second image and a third image generated respectively at a first instant, a second instant, later than the first instant, and a third instant, later than the second instant, by a camera mounted in an automotive vehicle. The method includes detecting a plurality of identical points in each image of the sequence of images, relating to a plurality of repetitive patterns in each image of the sequence of images and of identification, in the second image and in the third image, of the identical points corresponding to the points in space for which the calculated depth is identical, these points in space representing respectively in the second image and the third image one and the same real point.
1. A method of processing images allowing identification of at least one same point appearing on each image of a sequence of images generated by a camera mounted in an automotive vehicle, said method comprising:
detecting a plurality of identical points in each image of the sequence of images, the sequence of images comprising at least a first image generated at a first instant (t 1 ), a second image generated at a second instant (t 2 ) later than the first instant (t 1 ), and a third image generated at a third instant (t 3 ) later than the second instant (t 2 ), the plurality of identical points relating to a plurality of repetitive patterns in each image of the sequence of images,
for each of the plurality of identical points detected in the second image generated at the second instant (t 2 ), calculating a depth (Z A1 (t 1 ), Z A2 (t 1 )) at the first instant (t 1 ) of a point in space corresponding to said each of the plurality of identical points in the second image,
for each of the plurality of identical points detected in the third image generated at the third instant (t 3 ), calculating a depth (Z A3 (t 1 ), Z A4 (t 1 )) at the first instant (t 1 ) of a point in space corresponding to said each of the plurality of identical points in the third image, and
identifying, in the second image and in the third image, respective ones of each of the plurality of identical points corresponding to the points in space for which the calculated depth at the first instant (t 1 ) is identical, these points in space representing respectively in the second image and in the third image one and the same real point.
2. The method as claimed in claim 1 , in which the depth (Z A1 (t 1 ), Z A2 (t 1 )) at the first instant of the point in space corresponding to said each of the plurality of identical points in the second image is calculated on the basis of a rotation matrix and of a translation vector between the first image and the second image.
3. The method as claimed in claim 1 , in which the depth (Z A3 (t 1 ), Z A4 (t 1 )) at the first instant of the point in space corresponding to said each of the plurality of identical points in the third image is calculated on the basis of a rotation matrix and of a translation vector between the first image and the third image.
4. The method as claimed in claim 1 , comprising storing, after the calculating, of the depth (Z A1 (t 1 ), Z A2 (t 1 )) for each of the plurality of identical points detected in the second image of the sequence of images.
5. The method as claimed in claim 1 , comprising storing, after the calculating, of the depth (Z A3 (t 1 ), Z A4 (t 1 )) for each of the plurality of identical points detected in the third image of the sequence of images.
6. An image processing module allowing identification of at least one same point appearing on each image of a sequence of images generated by a camera mounted in an automotive vehicle, said image processing module configured to:
detect a plurality of identical points in each image of the sequence of images, the sequence of images comprising at least a first image generated at a first instant (t 1 ), a second image generated at a second instant (t 2 ) later than the first instant (t 1 ), and a third image generated at a third instant (t 3 ) later than the second instant (t 2 ), the plurality of identical points relating to a plurality of repetitive patterns in each image of the sequence of images,
calculate, for each of the plurality of identical points detected in the second image generated at the second instant (t 2 ), a depth (Z A1 (t 1 ), Z A2 (t 1 )) at the first instant (t 1 ) of a point in space corresponding to said each of the plurality of identical points in the second image,
calculate, for each of the plurality of identical points detected in the third image generated at the third instant (t 3 ), a depth (Z A3 (t 1 ), Z A4 (t 1 )) at the first instant (t 1 ) of a point in space corresponding to said each of the plurality of identical points in the third image, and
identify in the second image and in the third image, respective ones of each of the plurality of identical points corresponding to the points in space for which the calculated depth at the first instant (t 1 ) is identical, these points in space representing respectively in the second image and in the third image one and the same real point.
7. The image processing module as claimed in claim 6 , said image processing module being configured to calculate the depth (Z A1 (t 1 ), Z A2 (t 1 )) at the first instant of the point in space corresponding to said each of the plurality of identical points in the second image on the basis of a rotation matrix and of a translation vector between the first image and the second image.
8. The image processing module as claimed in claim 6 , said image processing module being configured to calculate the depth (Z A3 (t 1 ), Z A4 (t 1 )) at the first instant of the point in space corresponding to said each of the plurality of identical points in the third image on the basis of a rotation matrix and of a translation vector between the first image and the third image.
9. The image processing module as claimed in claim 6 , said image processing module being configured to store the depth (Z A1 (t 1 ), Z A2 (t 1 )) for each of the plurality of identical points detected in the second image of the sequence of images and the depth (Z A3 (t 1 ), Z A4 (t 1 )) for each of the plurality of identical points detected in the third image of the sequence of images.
10. An automotive vehicle comprising an image processing module as claimed in claim 6 .
11. The image processing module as claimed in claim 7 , said image processing module being configured to calculate the depth (Z A3 (t 1 ), Z A4 (t 1 )) at the first instant of the point in space corresponding to said each of the plurality of identical points in the third image on the basis of a rotation matrix and of a translation vector between the first image and the third image.
12. The method as claimed in claim 2 , in which the depth (Z A3 (t 1 ), Z A4 (t 1 )) at the first instant of the point in space corresponding to said each of the plurality of identical points in the third image is calculated on the basis of a rotation matrix and of a translation vector between the first image and the third image.