Image coding process and motion detecting process using bidirectional prediction

Claim 2

An image processing apparatus comprising: a) absolute difference value calculating means for calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) first comparison means for comparing the absolute difference value calculated by said absolute difference value calculating means with a first threshold value; c) calculation means for calculating a sum total of comparison results of said first comparison means; d) judging means for judging a field correlation of the image data between the first and second fields in accordance with the sum total calculated by said calculation means; and e) transforming means for orthogonally transforming the image data in accordance with an output of said judging means, wherein said transforming means is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 3

An apparatus according to claim 1 , further comprising data extracting means for extracting a portion of said image data of the first and second fields and supplying the image data portion to said absolute difference value calculating means.

Claim 4

An apparatus according to claim 1 , wherein said transforming mode is selected in accordance with an output of said judging means.

Claim 5

An image processing apparatus comprising: a) absolute difference value calculating means for calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) first comparison means for comparing the absolute difference value calculated by said absolute difference value calculating means with a first threshold value; c) count means for counting comparison results of said first comparison means; and d) judging means for judging a field correlation of the image data between the first and second fields in accordance with a count value of said count means; and e) transforming means for orthogonally transforming the image data in accordance with an output of said judging means, wherein said transforming means is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 6

An image processing apparatus comprising: a) absolute difference value calculating means for calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) first comparison means for comparing the absolute difference value calculated by said absolute difference value calculating means with a first threshold value; c) count means for counting comparison results of said first comparison means; and d) judging means for judging a field correlation of the image data between the first and second fields in accordance with a count value of said count means; and e) transforming means for orthogonally transforming the image data in accordance with an output of said judging means, wherein said transforming means is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 7

An apparatus according to claim 4 , further comprising data extracting means for extracting a portion of said image data of the first and second fields and supplying the image data portion to said absolute difference value calculating means.

Claim 8

An apparatus according to claim 4 , wherein said transforming mode is selected in accordance with an output of said judging means.

Claim 9

An image processing method comprising: a) an absolute difference value calculating step of calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) a first comparison step of comparing the absolute difference value calculated in said absolute difference value calculating step with a first threshold value; c) a calculation step of calculating a sum total of comparison results obtained in said first comparison step; d) a judging step of judging a field correlation of the image data between the first and second fields in accordance with the sum total calculated in said calculation step; and e) a transforming step of orthogonally transforming the image data in accordance with an output of said judging step, wherein said transforming step is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 10

An image processing method comprising: a) an absolute difference value calculating step of calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) a first comparison step of comparing the absolute difference value calculated in said absolute difference value calculating step with a first threshold value; c) a calculation step of calculating a sum total of comparison results obtained in said first comparison step; d) a judging step of judging a field correlation of the image data between the first and second fields in accordance with the sum total calculated in said calculation step; and e) a transforming step of orthogonally transforming the image data in accordance with an output of said judging step, wherein said transforming step is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 11

An image processing method comprising: a) an absolute difference value calculating step of calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) a first comparison step of comparing the absolute difference value calculated in said absolute difference value calculating step with a first threshold value; c) a count step of counting comparison results obtained in said first comparison step; d) a judging step of judging a field correlation of the image data between the first and second fields in accordance with a count value obtained in said count step; and e) a transforming step of orthogonally transforming the image data in accordance with an output of said judging step, wherein said transforming step is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 12

An image processing method comprising: a) an absolute difference value calculating step of calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; b) a first comparison step of comparing the absolute difference value calculated in said absolute difference value calculating step with a first threshold value; c) a count step of counting comparison results obtained in said first comparison step; d) a judging step of judging a field correlation of the image data between the first and second fields in accordance with a count value obtained in said count step; and e) a transforming step of orthogonally transforming the image data in accordance with an output of said judging step, wherein said transforming step is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 13

An image processing apparatus which codes image data through bidirectional prediction, comprising: a) a first memory for storing a reference image in a forward direction; b) first motion vector detecting means for detecting a motion vector by reading image data stored in said first memory; c) a second memory for storing a reference image in a backward direction; d) a third memory for storing a reference image in a forward direction in an expanded range; e) second motion vector detecting means for detecting a motion vector by reading image data stored in said second or third memory; and f) coding means for coding, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detecting means, wherein said coding means includes transforming means for orthogonally transforming image data, and the image data is interlaced image data, and said transforming means is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 14

An image processing apparatus which codes image data through bidirectional prediction, comprising: a) a first memory for storing a reference image in a forward direction; b) first motion vector detecting means for detecting a motion vector by reading image data stored in said first memory; c) a second memory for storing a reference image in a backward direction; d) a third memory for storing a reference image in a forward direction in an expanded range; e) second motion vector detecting means for detecting a motion vector by reading image data stored in said second or third memory; and f) coding means for coding, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detecting means, wherein said coding means includes transforming means for orthogonally transforming image data, and the image data is interlaced image data, and said transforming means is provided with a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 15

An apparatus according to claim 9 , wherein said transforming means includes detection means for detecting a field correlation of the image data, and the transform mode is selected in accordance with a field correlation detected by said detection means.

Claim 16

An apparatus according to claim 10 , wherein said detection means comprises: absolute difference value calculating means for calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; first comparison means for comparing the absolute difference value calculated by said absolute difference value calculating means with a first threshold value; calculation means for calculating a sum total of comparison results of said first comparison means; and judging means for judging a field correlation of the image data between the first and second fields in accordance with the sum total calculated by said calculation means.

Claim 17

An apparatus according to claim 10 , wherein said detection means comprises: absolute difference value calculating means for calculating an absolute difference value of each pixel between image data of first and second fields constituting a frame image; first comparison means for comparing the absolute difference value calculated by said absolute difference value calculating means with a first threshold value; count means for counting comparison results of said first comparison means; and judging means for judging a field correlation of the image data between the first and second fields in accordance with a count value of said count means.

Claim 18

An image processing method of coding image data through bidirectional prediction, comprising: a) a first storing step of storing in a first memory, a reference image in a forward direction; b) a first motion vector detecting step of detecting a motion vector by reading image data stored in said first memory; c) a second storing step of storing in a second memory, a reference image in a backward direction; d) a third storing step of storing in a third memory, a reference image in a forward direction in an expanded range; e) a second motion vector detecting step of detecting a motion vector by reading image data stored in said second or third memory; and f) a coding step of coding, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detecting means, wherein said coding step includes a transforming step of orthogonally transforming image data, and wherein the image data is interlaced image data, and said transforming step has a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 19

An image processing method of coding image data through bidirectional prediction, comprising: a) a first storing step of storing in a first memory, a reference image in a forward direction; b) a first motion vector detecting step of detecting a motion vector by reading image data stored in said first memory; c) a second storing step of storing in a second memory, a reference image in a backward direction; d) a third storing step of storing in a third memory, a reference image in a forward direction in an expanded range; e) a second motion vector detecting step of detecting a motion vector by reading image data stored in said second or third memory; and f) a coding step of coding, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detecting means, wherein said coding step includes a transforming step of orthogonally transforming image data, and wherein the image data is interlaced image data, and said transforming step has a field transform mode for transforming in a unit of a field and a frame transform mode for transforming in a unit of a frame.

Claim 20

An image processing apparatus capable of coding image data through bidirectional prediction, comprising: a) a first motion vector detector, arranged to detect a motion vector by using a first reference image in a forward direction; b) a second motion vector detector, arranged to detect a motion vector by selectively using a second reference image in a backward direction and a third reference image in the forward direction, the third reference image being an image on a picture of a same timing as that of the first reference image and having an image of an area different from the first reference image on the picture; and c) an encoder, arranged to encode, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detector, wherein said encoder includes a transformer, arranged to orthogonally transform image data, and the image data is interlace image data, and said transformer is provided with a field transform mode and a frame transform mode.

Claim 21

An image processing apparatus capable of coding image data through bidirectional prediction, comprising: a) a first motion vector detector, arranged to detect a motion vector by using a first reference image in a forward direction; b) a second motion vector detector, arranged to detect a motion vector by selectively using a second reference image in a backward direction and a third reference image in the forward direction, the third reference image being an image on a picture of a same timing as that of the first reference image and having an image of an area different from the first reference image on the picture; and c) an encoder, arranged to encode, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detector, wherein said encoder includes a transformer, arranged to orthogonally transform image data, and the image data is interlace image data, and said transformer is provided with a field transform mode and a frame transform mode.

Claim 22

An apparatus according to claim 14 , wherein said second motion vector detector detects a motion vector for bidirectional prediction by using the second reference image, and detects a motion vector for forward prediction by using the third reference image.

Claim 23

An apparatus according to claim 14 , further comprising a selector, arranged to select either a motion vector detected by said first motion vector detector or a motion vector detected by said second motion vector detector.

Claim 24

An apparatus according to claim 16 , wherein said selector selects a motion vector in accordance with prediction error signals obtained on the basis of a motion vector detected by said first motion vector detector and a motion vector detected by said second motion vector detector.

Claim 25

An apparatus according to claim 14 , wherein said transformer includes a field correlation detector, arranged to detect a field correlation of the image data, and said transform mode is selected in accordance with a field correlation detected by said field correlation detector.

Claim 26

An apparatus according to claim 18 , wherein said field correlation detector comprises: an absolute difference value calculating unit, arranged to calculate an absolute difference value of each pixel between image data of first and second fields constituting a frame image; a first comparison unit, arranged to compare the absolute difference value calculated by said absolute difference value calculating unit with a first threshold value; a. calculation unit, arranged to calculate a sum total of comparison results of said first comparison unit; and a judging unit, arranged to judge a field correlation of the image data between the first and second fields in accordance with the sum total calculated by said calculation unit.

Claim 27

An apparatus according to claim 18 , wherein said field correlation detector comprises: an absolute difference value calculating unit, arranged to calculate an absolute difference value of each pixel between image data of first and second fields constituting a frame image; a first comparison unit, arranged to compare the absolute difference value calculated by said absolute difference value calculating unit with a first threshold value; a count unit, arranged to count comparison results of said first comparison unit; and a judging unit, arranged to judge a field correlation of the image data between the first and second fields in accordance with a count value of said count unit.

Claim 28

An image processing method capable of coding image data through bidirectional prediction, comprising: a) a first motion vector detecting step of detecting a motion vector by using a first reference image in a forward direction, by using a first motion vector detector; b) a second motion vector detecting step of detecting a motion vector by selectively using a second reference image in a backward direction and a third reference image in the forward direction, by using a second motion vector detector, the third reference image being an image on a picture of a same timing as that of the first reference image and having an image of an area different from the first reference image on the picture; and c) an encoding step of encoding, through motion compensation prediction, input image data by using a motion vector detected in said first or second motion vector detecting steps, wherein said encoder step includes a transforming step of orthogonally transforming image data, and the image data is interlaced image data, and said transforming step is provided with a field transform mode and a frame transform mode.

Claim 29

An image processing method capable of coding image data through bidirectional prediction, comprising: a) a first motion vector detecting step of detecting a motion vector by using a first reference image in a forward direction, by using a first motion vector detector; b) a second motion vector detecting step of detecting a motion vector by selectively using a second reference image in a backward direction and a third reference image in the forward direction, by using a second motion vector detector, the third reference image being an image on a picture of a same timing as that of the first reference image and having an image of an area different from the first reference image on the picture; and c) an encoding step of encoding, through motion compensation prediction, input image data by using a motion vector detected in said first or second motion vector detecting steps, wherein said encoder step includes a transforming step of orthogonally transforming image data, and the image data is interlaced image data, and said transforming step is provided with a field transform mode and a frame transform mode.

Claim 30

An image processing apparatus capable of coding image data through bidirectional prediction, comprising: a) a storage unit, arranged to store a reference image in a forward direction, in a first memory area, a reference image in a backward direction, in a second memory area, and a reference image in a forward direction in an expanded range, in a third memory area; b) a first motion vector detector, arranged to detect a motion vector by reading image data stored in said first memory area; c) a second motion vector detector, arranged to detect a motion vector by selectively reading image data stored in said second and third memory areas; and d) an encoder, arranged to encode, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detector, wherein said encoder includes a transformer, arranged to orthogonally transform image data, and the image data is interlaced image data, and said transformer is provided with a field transform mode for transforming in field units and a frame transform mode for transforming in frame units.

Claim 31

An image processing apparatus capable of coding image data through bidirectional prediction, comprising: a) a storage unit, arranged to store a reference image in a forward direction, in a first memory area, a reference image in a backward direction, in a second memory area, and a reference image in a forward direction in an expanded range, in a third memory area; b) a first motion vector detector, arranged to detect a motion vector by reading image data stored in said first memory area; c) a second motion vector detector, arranged to detect a motion vector by selectively reading image data stored in said second and third memory areas; and d) an encoder, arranged to encode, through motion compensation prediction, input image data by using a motion vector detected by said first or second motion vector detector, wherein said encoder includes a transformer, arranged to orthogonally transform image data, and the image data is interlaced image data, and said transformer is provided with a field transform mode for transforming in field units and a frame transform mode for transforming in frame units.

Claim 32

An image processing method capable of coding image data through bidirectional prediction, comprising: a) a storing step of storing a reference image in a forward direction, in a first memory area, a reference image in a backward direction, in a second memory area, and a reference image in a forward direction in an expanded range, in a third memory area; b) a first motion vector detecting step of detecting a motion vector by reading image data stored in said first memory area; c) a second motion vector detecting step of detecting a motion vector by selectively reading image data stored in said second and third memory areas; and d) an encoding step of encoding, through motion compensation prediction, input image data by using a motion vector detected in said first or second motion vector detecting steps, wherein said encoding step includes a transforming step of orthogonally transforming image data, and the image data is interlaced image data, and said transforming step has a field transform mode for transforming in field units and a frame transform mode for transforming in frame units.

Claim 33

An image processing method capable of coding image data through bidirectional prediction, comprising: a) a storing step of storing a reference image in a forward direction, in a first memory area, a reference image in a backward direction, in a second memory area, and a reference image in a forward direction in an expanded range, in a third memory area; b) a first motion vector detecting step of detecting a motion vector by reading image data stored in said first memory area; c) a second motion vector detecting step of detecting a motion vector by selectively reading image data stored in said second and third memory areas; and d) an encoding step of encoding, through motion compensation prediction, input image data by using a motion vector detected in said first or second motion vector detecting steps, wherein said encoding step includes a transforming step of orthogonally transforming image data, and the image data is interlaced image data, and said transforming step has a field transform mode for transforming in field units and a frame transform mode for transforming in frame units.