Method of image analysis of an image in a sequence of images to determine a cross-fade measure
View Patent ↗The present invention relates to a method of and apparatus for image analysis and in particular may relate to the detection of cross-fades in film or video sequences. The invention relates in particular to a method of analysing an image of a sequence of images to determine a cross-fade measure based on determined temporal picture information transitions associated with picture elements of the image. In particular, the cross-fade measure may be determined based on the extent to which the temporal picture information transitions are uniform. The method and apparatus of the invention can provide a measure of likelihood of a cross-fade in a single pass. In addition the described method can be accomplished in real-time or close to real-time. In addition the cross-fade detection results are comparable with, or better than, the results achieved by the prior art methods.
1. A method of image analysis of an image in a sequence of images to determine a cross-fade measure, comprising the steps of:
for a plurality of picture information picture elements representing the image, determining first and second temporal picture information gradients associated with the picture element and comparing the first and second temporal picture information gradients associated with the picture element to determine whether the picture information gradient is substantially uniform;
wherein the first temporal picture information gradient is determined between a current image and a second image of the sequence of images; and
wherein, for each picture element, the second image is selected as the image within a range of candidate second images that has the maximum difference between the picture information of the picture element of said current image and the corresponding picture element of the candidate second image.
2. The method as claimed in claim 1 where the difference in picture information is determined between motion-compensated corresponding picture elements.
3. The method as claimed in claim 1 wherein the second temporal picture information gradient is determined between said second image and a third image of the sequence.
4. The method as claimed in claim 3 wherein, for each picture element, the third image is selected as the image within a range of candidate third images that has the maximum difference between the picture information of the picture element of said second image and the corresponding picture element of the candidate third image.
5. The method as claimed in claim 1 wherein first and second temporal picture information gradients associated with the picture element are compared by taking the difference therebetween.
6. The method of claim 1 wherein a temporal picture information transition is considered to be substantially uniform when the difference between first and second temporal picture information gradients associated with the picture element is below a threshold.
7. The method as claimed in claim 1 further comprising the step of selecting qualifying picture elements, wherein only qualifying picture elements are used in determining a cross-fade measure.
8. The method of image analysis as claimed in claim 7 wherein the step of selecting qualifying picture elements comprises the step of selecting candidate picture elements located in image regions having substantially homogenous picture information as qualifying picture elements.
9. The method of image analysis as claimed in claim 7 wherein the step of selecting qualifying picture elements comprises the step of selecting candidate picture elements located in image regions having substantially homogenous picture information gradients as qualifying picture elements.
10. The method as claimed in claim 7 wherein the determination of the cross-fade measure is also dependent on the number of qualifying picture elements identified in the image.
11. The method of image analysis as claimed in claim 7 wherein the determination of the cross-fade measure is also dependent on how dispersed the qualifying picture elements are throughout the image.
12. The method as claimed in claim 1 wherein the determination of the cross-fade measure is also dependent on the magnitudes of the temporal picture information transitions associated with picture elements.
13. The method of image analysis as claimed in claim 1 further comprising the initial step, for each image of the sequence of images, of filtering the picture information of the image to generate picture information picture elements.
14. The method as claimed in claim 1 wherein the picture information is luminance information.
15. The method as claimed in claim 1 further comprising: establishing a cross-fade indication when the cross-fade measure determined in respect of at least two sequentially considered images indicates that a cross-fade is likely.
16. The method as claimed in claim 1 further comprising metadata generation from the cross-fade measure.
17. A non-transitory program carrier carrying computer-readable instructions implementing the method of image analysis of any of an image in a sequence of images to determine a cross-fade measure, comprising the steps of:
for a plurality of picture information picture elements representing the image, determining first and second temporal picture information gradients associated with the picture element and comparing the first and second temporal picture information gradients associated with the picture element to determine whether the picture information gradient is substantially uniform;
wherein the first temporal picture information gradient is determined between a current image and a second image of the sequence of images; and
wherein, for each picture element the second image is selected as the image within a range of candidate second images that has the maximum difference between the picture information of the picture element of said current image and the corresponding picture element of the candidate second image.