Method and device for identifying an image acquisition feature of a digital image and apparatus for implementing such a device

1. A method for identifying an image acquisition feature of a digital image composed of pixels and oriented in a coordinate system having at least one reference axis (Xo), the method comprising:

(A) detecting the contours of each distinctive element of the image;

(B) forming a list including each contour constituted by a rectilinear segment;

(C) searching, in the list of rectilinear contours, for a pair of rectilinear segments (S 1 , S 2 );

(D) where step (C) is successful, checking for a condition of relative symmetry of the rectilinear segments (S 1 , S 2 ) with respect to the reference axis (Xo); and

(E) producing, respectively in the case where step (D) is successful and in the case where one of steps (C) and (D) fails, a data respectively representative of the presence and absence of perspective in the image acquisition, with respect to the reference axis (Xo), the presence or absence of perspective constituting said image acquisition feature;

wherein step (C) comprises searching in the image for a first rectilinear segment (S 1 ) of maximum length; checking, as a condition for validating step (C), that the length of the first rectilinear segment (S 1 ) is higher than a length lower limit; searching among the remaining rectilinear segments of the image for at least a second rectilinear segment (S 2 ) of acceptable length; and checking, as a condition for validating step (C), that the length of the second rectilinear segment (S 2 ) is at least equal to a fraction of the length of the first rectilinear segment (S 1 );

wherein step (D) comprises determining the slope (θ, θ′) of each of said first and second rectilinear segments (S 1 , S 2 ) with respect to said reference axis (Xo); checking, as a condition for validating step (D), that the slopes of said first and second rectilinear segments are of opposite signs; and checking, as a condition for validating step (D), that the respective absolute values of the slopes of said first and second rectilinear segments exhibit between each other a deviation at the most equal to a maximum permitted slope deviation (Δθ); and

further comprising (F) correcting the orientation of the image by applying to the image a rotation equal to the angle formed, with respect to the reference axis (Xo), by a line having an intermediary slope between the slopes (θ, θ′) of the first and second rectilinear segments (S 1 , S 2 ).

2. The method according to claim 1 , wherein the coordinate system of the image comprises two reference axes (Xo, Yo), and in that steps (D) and (E) are implemented with each of these two axes (Xo, Yo).

3. The method according to claim 1 , wherein step (A) comprises measuring the gradient of the image intensity at each pixel, selecting, as a contour, each set of neighboring pixels having the highest gradient, and forming a contour map containing each detected contour in the image.

4. The method according to claim 3 , wherein step (B) comprises: selecting, as a pole, the origin (0) of the image coordinate system; and considering, as belonging to a same rectilinear segment, the pixels belonging to the contour map and the coordinates x and y of which in the image coordinate system satisfy the following equation: x. cos(θ)+y. sin(θ)=ρ, where θ is the angle that this segment forms with the reference axis, and where ρ is the distance of the pole to this rectilinear segment.

5. The method according to claim 1 , wherein said length lower limit is defined as a fraction of the number of pixels that the image comprises at least along the reference axis (Xo).

6. The method according to claim 1 , wherein said maximum permitted slope deviation (Δθ) is at the most equal to 10 degrees.

7. The method according to claim 1 , further comprising an additional step (F) of correcting the orientation of the image by applying to the image a rotation equal to the angle formed, with respect to the reference axis (Xo), by a line having an intermediary slope between the slopes (θ, θ′) of the first and second rectilinear segments (S 1 , S 2 ).

8. A device for identifying an image acquisition feature of a digital image composed of pixels and oriented in a coordinate system, said device configured to assess the presence or absence of a perspective in the image acquisition based solely on the pixels and the orientation of the image, the presence or absence of perspective constituting said image acquisition feature, the device comprising a computing circuit operative to:

(A) detect the contours of each distinctive element of the image;

(B) form a list including each contour constituted by a rectilinear segment;

(C) search, in the list of rectilinear contours, for a pair of rectilinear segments (S 1 , S 2 );

(D) where step (C) is successful, check for a condition of relative symmetry of the rectilinear segments (S 1 , S 2 ) with respect to the reference axis (Xo); and

(E) produce, respectively in the case where step (D) is successful and in the case where one of steps (C) and (D) fails, a data respectively representative of the presence and absence of perspective in the image acquisition, with respect to the reference axis (Xo), the presence or absence of perspective constituting said image acquisition feature;

wherein step (C) comprises searching in the image for a first rectilinear segment (S 1 ) of maximum length; checking, as a condition for validating step (C), that the length of the first rectilinear segment (S 1 ) is higher than a length lower limit; searching among the remaining rectilinear segments of the image for at least a second rectilinear segment (S 2 ) of acceptable length; and checking, as a condition for validating step (C), that the length of the second rectilinear segment (S 2 ) is at least equal to a fraction of the length of the first rectilinear segment (S 1 );

wherein step (D) comprises determining the slope (θ, θ′) of each of said first and second rectilinear segments (S 1 , S 2 ) with respect to said reference axis (Xo); checking, as a condition for validating step (D), that the slopes of said first and second rectilinear segments are of opposite signs; and checking, as a condition for validating step (D), that the respective absolute values of the slopes of said first and second rectilinear segments exhibit between each other a deviation at the most equal to a maximum permitted slope deviation (Δθ); and

further comprising (F) correcting the orientation of the image by applying to the image a rotation equal to the angle formed, with respect to the reference axis (Xo), by a line having an intermediary slope between the slopes (θ, θ′) of the first and second rectilinear segments (S 1 , S 2 ).

9. A digital image capture apparatus configured to identify an image acquisition feature of a digital image composed of pixels and oriented in a coordinate system, said apparatus configured to assess the presence or absence of a perspective in the image acquisition based solely on the pixels and the orientation of the image, the presence or absence of perspective constituting said image acquisition feature, the device comprising a computing circuit operative to:

(A) detect the contours of each distinctive element of the image;

(B) form a list including each contour constituted by a rectilinear segment;

(C) search, in the list of rectilinear contours, for a pair of rectilinear segments (S 1 , S 2 );

(D) where step (C) is successful, check for a condition of relative symmetry of the rectilinear segments (S 1 , S 2 ) with respect to the reference axis (Xo); and

(E) produce, respectively in the case where step (D) is successful and in the case where one of steps (C) and (D) fails, a data respectively representative of the presence and absence of perspective in the image acquisition, with respect to the reference axis (Xo), the presence or absence of perspective constituting said image acquisition feature;

wherein step (C) comprises searching in the image for a first rectilinear segment (S 1 ) of maximum length; checking, as a condition for validating step (C), that the length of the first rectilinear segment (S 1 ) is higher than a length lower limit; searching among the remaining rectilinear segments of the image for at least a second rectilinear segment (S 2 ) of acceptable length; and checking, as a condition for validating step (C), that the length of the second rectilinear segment (S 2 ) is at least equal to a fraction of the length of the first rectilinear segment (S 1 );

wherein step (D) comprises determining the slope (θ, θ′) of each of said first and second rectilinear segments (S 1 , S 2 ) with respect to said reference axis (Xo); checking, as a condition for validating step (D), that the slopes of said first and second rectilinear segments are of opposite signs; and checking, as a condition for validating step (D), that the respective absolute values of the slopes of said first and second rectilinear segments exhibit between each other a deviation at the most equal to a maximum permitted slope deviation (Δθ); and

further comprising (F) correcting the orientation of the image by applying to the image a rotation equal to the angle formed, with respect to the reference axis (Xo), by a line having an intermediary slope between the slopes (θ, θ′) of the first and second rectilinear segments (S 1 , S 2 ).