Image processing system

Claim 2

An image processing system for combining a plurality of images, which include at least a time-varying image, said system comprising: mask data memory means for storing mask data designating area(s) of the time-varying image in the plurality of images to be combined; first image memory means for storing a present image of the time-varying image; second image memory means for storing a preceding image of the time-varying image; detection means for detecting a motion of the time-varying image based on the present image and the preceding image stored in said first and second image memory means respectively; updating means for updating mask data stored in said mask data memory means based on the motion of the time-varying image detected by said detection means; and combining means for combining the present image of the time-varying image with other images(s) of the plurality of images on a basis of mask data updated by said updating means.

Claim 3

The image processing system according to claim 1, further comprising: third image memory means for storing a present combined image combined by said combining means; fourth image memory means for storing a preceding combined image combined by said combining means; second mask data memory means for storing a preceding mask data corresponding to the preceding combined image; discrimination means for discriminating which original image of the plurality of images is included in an area of the combined image based on the mask data; second detection means for detecting a motion of the area of the combined image by referring to both areas of the preceding combined image and a present combined image including a same original image; and coding means for coding an area of the combined image by motion compensation predictive coding by using the motion detected by said second detection means.

Claim 4

The image processing system according to claim 1, further comprising: motion memory means for storing the motion of the time-varying image detected by said detection means; third image memory means for storing a present combined image combined by said combining means; fourth image memory means for storing a preceding combined image combined by said combining means; second mask data memory means for storing a preceding mask data corresponding to the preceding combined image; discrimination means for discriminating whether an area of the combined image includes the time-varying image based on the mask data; second detection means for detecting a motion of an area of the present image discriminated not to include the time-varying image; and coding means for coding the combined image by motion compensation predictive coding by using the motion stored in said motion memory means or the motion detected by said second detection means.

Claim 5

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of: storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; updating the mask data stored in the mask data memory based on the detected motion of the time-varying image; and combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data.

Claim 6

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of: storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; updating the mask data stored in the mask data memory based on the detected motion of the time-varying image; and combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data.

Claim 7

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of: storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; creating a present mask data from a preceding mask data stored in the mask data memory based on the detected motion of the time-varying image; and combining the present time-varying image with other image(s) of the plurality of images on a basis of the present mask data.

Claim 8

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of: storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; creating a present mask data from a preceding mask data stored in the mask data memory based on the detected motion of the time-varying image; and combining the present time-varying image with other image(s) of the plurality of images on a basis of the present mask data.

Claim 9

The image processing method according to claim 5, wherein the detection is executed by referring, as the preceding time-varying image, only to a portion designated to be combined by the mask data.

Claim 10

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of: storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; updating mask data stored in the mask data memory based on the detected motion of the time-varying image; combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data; cutting out an area to be encoded from the combined image; judging whether the cut out area contains the time-varying image with reference to the mask data; determining an encoding method according to the judging; and encoding the cut out area by the determined encoding method.

Claim 11

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of: storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; updating mask data stored in the mask data memory based on the detected motion of the time-varying image; combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data; cutting out an area to be encoded from the combined image; judging whether the cut out area contains the time-varying image with reference to the mask data; determining an encoding method according to the judging; and encoding the cut out area by the determined encoding method.

Claim 12

The image processing method according to claim 7, wherein when the area to be encoded is judged to contain the time-varying image, an encoding method is selected which comprises the steps of searching an area in a preceding combined image corresponding to the area to be encoded, with limiting a search range within the area containing the time-varying image in reference to the mask data; detecting a motion based on the searched area and the area to be encoded; getting a difference between the searched area and the area to be encoded; and encoding the motion and the difference.

Claim 13

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of; storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; updating mask data stored in the mask data memory based on the detected motion of the time-varying image; combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data; cutting out an area to be encoded from the combined image; judging whether the cut out area contains the time-varying image with reference to the mask data; and encoding the cut out area based on the detected motion when the area is judged to contain the time-varying image.

Claim 14

An image processing method for combining a plurality of images, which include at least one time-varying image, said method comprising the steps of; storing mask data designating area(s) of the time-varying image in the plurality of images to be combined in a mask data memory; storing a present time-varying image and a preceding time-varying image in an image memory; detecting a motion of the time-varying image based on the present and the preceding time-varying image; updating mask data stored in the mask data memory based on the detected motion of the time-varying image; combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data; cutting out an area to be encoded from the combined image; judging whether the cut out area contains the time-varying image with reference to the mask data; and encoding the cut out area based on the detected motion when the area is judged to contain the time-varying image.

Claim 15

The image processing method according to claim 9, wherein the encoding step comprises the steps of: determining an area in the preceding combined image corresponding to the area to be encoded based on the detected motion; getting a difference between the determined area and the area to be encoded; and encoding the motion and the difference.

Claim 16

An image processing method for combining a plurality of images, which include at least one encoded time-varying image, said method comprising the steps of: extracting a motion data from the encoded time-varying image; decoding the encoded time-varying image by using the motion data; storing a present decoded time-varying image in an image memory; storing mask data designating areas of each of the plurality of images to be combined in a mask data memory; updating a mask data stored in the mask data memory based on the extracted motion data; combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data; cutting out an area to be encoded from the combined image; judging whether the cut out area contains the time-varying image with reference to the mask data; and encoding the cut out area based on the extracted motion when the area is judged to contain the time-varying image.

Claim 17

An image processing method for combining a plurality of images, which include at least one encoded time-varying image, said method comprising the steps of: extracting a motion data from the encoded time-varying image; decoding the encoded time-varying image by using the motion data; storing a present decoded time-varying image in an image memory; storing mask data designating areas of each of the plurality of images to be combined in a mask data memory; updating a mask data stored in the mask data memory based on the extracted motion data; combining the present time-varying image with other image(s) of the plurality of images on a basis of the updated mask data; cutting out an area to be encoded from the combined image; judging whether the cut out area contains the time-varying image with reference to the mask data; and encoding the cut out area based on the extracted motion when the area is judged to contain the time-varying image.