Method for detection of film mode or camera mode

1. Method for detection of film mode or camera mode of a sequence comprising at least three consecutive video frames, wherein it comprises the following steps, carried out by a processing unit:

determine a histogram of the video levels of each of the video frames,

calculate, for each of the video frames of the sequence, a value representative of the histogram difference, called histogram difference value, between the histogram of said video frame and the histogram of the next video frame,

compare the histogram difference value of each of the video frames of the sequence with at least one comparison value to generate, for each of said video frames, a parameter, of difference, whose value is representative of the character different or not to that of said video frame with respect to the next video frame, and

compare the pattern formed by the values of difference parameters of video frames of the sequence with a plurality of predetermined patterns to generate mode information identifying the mode, film or camera, of said sequence of video frames.

2. Method according to claim 1 , wherein, for a current video frame, said at least one comparison value belongs to a group comprising:

the histogram difference value of the preceding video frame,

the histogram difference value for the next video frame, and

a predetermined threshold value.

3. Method according to claim 1 , wherein, to calculate the histogram difference value for a current video frame, the following steps are carried out:

calculate, for each video level, the difference between the occurrence value of said video level in the current video frame and the occurrence value of said video level in the next video frame so as to generate an occurrence difference for each of said video levels, and

add-up the absolute values of occurrence differences of all the video levels so as to generate said histogram difference value of the current video frame.

4. Method according to claim 1 , wherein, to calculate the histogram difference value for a current video frame, the following steps are carried out:

calculate, for each video level i comprised between 0 and N and each offset value j comprised between −n and n, the difference between the occurrence value of the video level i in the current video frame and the occurrence value of the video level i+j in the next video frame, so as to generate at most 2n+1 occurrences for each video level i,

select, for each video level i, the lowest absolute value occurrence difference, called minimal occurrence, from among said 2n+1 occurrence differences, and

add-up the minimal occurrence differences of all the video levels so as to generate said histogram difference value of the current video frame.

5. Method according to claim 1 , wherein, generate the difference parameter of a current video frame, the following steps are carried out:

the histogram difference value of said current video frame is compared with a predetermined non-null threshold value, and

a first value (“1”) is assigned to the difference parameter of the current video frame if the histogram difference vale of the current video frame is greater than said threshold value and a second value (“0”) is assigned if not.

6. Method according to claim 1 , wherein, to generate the difference parameter of a current video frame, the following steps are carried out:

said histogram difference value of said current video frame is compared with the histogram difference value of the preceding video frame and with the histogram difference value of the next video frame, and

a first value (“1”) is assigned to the difference parameter of the current video frame if said histogram difference value of said current video frame is greater than both the histogram difference value of the next video frame and the histogram difference value of the preceding video frame and the second value (“0”) is assigned if not.

7. Method according to claim 1 , wherein it also comprises, previous to step of determination of video level histograms, a step of spatial filtering of video frames of the sequence.

8. Method according to claim 1 , wherein it also comprises, previous to the step of comparison of the pattern formed by the difference parameters with the predetermined patters, a step of temporal filtering of said difference parameters pattern.

9. Method according to claim 1 , wherein said mode information identifies the type of film mode of said sequence of video frames.

10. Method according to claim 1 , wherein a mode information (MODE) is generated for each video frame of said sequence, said mode information also comprising a bit indicating if said video frame is a source video frame or not.

11. Device for detection of the film mode or camera mode of a sequence comprising at least three consecutive video frames, characterized in that it comprises:

a first calculation circuit to determine a histogram of video levels of each of the video frames,

a second calculation circuit to calculate, for each of the video frames of the sequence, a value representative of the histogram difference, called histogram difference, between the histogram of said video frame and the histogram of the next video frame,

a first comparison circuit to compare the histogram difference value of each of the video frames of the sequence with at least one comparison value and generate, for each of said video frames, a parameter, called difference, whose value is representative of the character different or not to that of said video frame with respect to the next video frame,

a second comparison circuit to compare the pattern formed by the values of difference parameters of video frames of the sequence with a plurality of predetermined patters to generate mode information identifying the mode, film or camera, of said sequence video frames, and

a control unit ( 160 ) to control said first and second calculation circuits and said first and second comparison circuits.