Method for noise reduction in an image sequence
A method for processing a frame in a sequence of successively acquired frames in dynamic digital radiography includes the multi-scale representation of frames being subjected to temporal filtering by adding at least one correction image to a corresponding detail image(s) in the multi-scale representations of a frame of interest, the correction image being computed by combining clipped difference images obtained as the difference between the multi-scale representation of the frame of interest and the multi-scale representations of a selection of other frames in said sequence.
1. A method for processing a frame of interest in a sequence of successively acquired frames of digital images of an object in digital radiation imaging without requiring motion estimation, the method comprising the steps of:
generating a multi-scale representation of the successively acquired frames of the digital images of the object, the multi-scale representation including detail images of the object at different scales;
subjecting the multi-scale representation of the frame of interest to temporal filtering by adding at least one correction image of the object to a detail image or detail images of the object in the multi-scale representation corresponding to the frame of interest in order to reduce noise in the frame of interest; wherein
the at least one correction image of the object is computed by combining clipped difference images of the object obtained as a difference between the multi-scale representation of the frame of interest and the multi-scale representations of a selection of other frames of digital images of the object in the sequence.
2. The method according to claim 1 , wherein the other frames of the digital images of the object in the sequence are a predefined number of image frames of the digital images of the object preceding an image of the object to be processed in the sequence.
3. The method according to claim 2 , wherein the other frames of the digital images of the object in the sequence preceding the frame of interest are successive frames in the sequence.
4. The method according to claim 1 , wherein the other frames of the digital images of the object in the sequence are a predefined number of frames preceding and a predefined number of frames of the digital images of the object following the frame of interest.
5. The method according to claim 2 , wherein the predefined number of frames of the digital images of the object is a function of noise characteristics and/or a desired amount of noise reduction in the frame of interest.
6. The method according to claim 4 , wherein the predefined number of frames of the digital images of the object is a function of noise characteristics and/or a desired amount of noise reduction in the frame of interest.
7. The method according to claim 1 , wherein the clipped difference images of the object are obtained by clipping difference images of the object with a clipping bound that is defined per scale.
8. The method according to claim 1 , wherein the clipped difference images of the object are obtained by clipping difference images of the object with a clipping bound that depends on a value of a noise level computed in the frame of interest.
9. The method according to claim 1 , wherein the temporal filtering is only applied to the detail images of the object up to a predefined scale s max .
10. A non-transitory computer readable medium comprising computer executable program code adapted to carry out the method of claim 1 .