Image retrieval by comparing wavelet encoded database images with input shapes

Claim 2

An image processing apparatus comprising: a storage section arranged to store a plurality of images as encoded data; an input section arranged to input an image shape as information of an image to be retrieved by search; a generation section arranged to generate query data on the basis of the input image shape; and a determination section arranged to determine similarity between the input image shape and an image, which is expected to be obtained by decoding the encoded data, by making a comparison of the query data with the encoded data read out from said storage section, wherein the encoded data is obtained by sequentially encoding transform coefficients, which are obtained by a discrete wavelet transformation of an image and said determination section compares the query data and a synthesized sub-band which is the logical combination of a plurality of sub-bands obtained by the discrete wavelet transformation.

Claim 3

The apparatus according to claim 1 , wherein when the similarity is low, said determination section reads out new encoded data, and compares the new encoded data with the query data.

Claim 4

The apparatus according to claim 1 , further comprising a decoder arranged to reconstruct an image by decoding the encoded data stored in said storage section on the basis of the similarity obtained by said determination section.

Claim 5

The apparatus according to claim 1 , wherein the wavelet transformation of the image uses a predetermined number of bit planes.

Claim 6

The apparatus according to claim 4 , wherein said determination section makes the comparison in units of bit planes read out from the encoded data.

Claim 7

The apparatus according to claim 4 , wherein said determination section makes the comparison in units of data in a predetermined spatial frequency region read out from the encoded data.

Claim 8

The apparatus according to claim 1 , wherein the query data is a binary image indicating the image shape.

Claim 9

The apparatus according to claim 1 , wherein said storage section is a part of an image database.

Claim 10

The image processing apparatus according to claim 1 , wherein the plurality of sub-bands comprises LH, HH, and HL sub-bands and excludes an LL sub-band.

Claim 11

An image processing method for searching a storage section, that stores a plurality of images as encoded data, for a desired image, said method comprising the step of: inputting an image shape as information of an image to be retrieved by search; generating query data on the basis of the input image shape; and determining similarity between the input image shape and an image, which is expected to be obtained by decoding the encoded data, by making a comparison of the query data with the encoded data read out from the storage section, wherein the encoded data is obtained by sequentially encoding transform coefficients, which are obtained by a discrete wavelet transformation of an image, and the determining step compares the query data and a synthesized sub-band which is the logical combination of a plurality of sub-bands obtained by the discrete wavelet transformation.

Claim 12

An image processing method for searching a storage section, that stores a plurality of images as encoded data, for a desired image, said method comprising the step of: inputting an image shape as information of an image to be retrieved by search; generating query data on the basis of the input image shape; and determining similarity between the input image shape and an image, which is expected to be obtained by decoding the encoded data, by making a comparison of the query data with the encoded data read out from the storage section, wherein the encoded data is obtained by sequentially encoding transform coefficients, which are obtained by a discrete wavelet transformation of an image, and the determining step compares the query data and a synthesized sub-band which is the logical combination of a plurality of sub-bands obtained by the discrete wavelet transformation.

Claim 13

The method according to claim 10 , wherein when the similarity is low, the determination step includes reading out new encoded data and comparing the new encoded data with the query data.

Claim 14

The method according to claim 10 , further comprising the step of reconstructing an image by decoding the encoded data stored in the storage on the basis of the similarity.

Claim 15

The method according to claim 10 , wherein the wavelet transformation of the image uses a predetermined number of bit planes.

Claim 16

The method according to claim 13 , wherein the determination step includes making the comparison in units of bit planes read out from the encoded data.

Claim 17

The method according to claim 13 , wherein the determination step includes making the comparison in units of data in a predetermined spatial frequency region read out from the encoded data.

Claim 18

The method according to claim 10 , wherein the query data is a binary image indicating the image shape.

Claim 19

The method according to claim 10 , wherein the storage section is a part of an image database.

Claim 20

The method according to claim 10 , wherein the plurality of sub-bands comprises LH, HH, and HL sub-bands and excludes an LL sub-band.

Claim 21

A computer program product storing a computer readable medium having computer program code, for searching a storage section that stores a plurality of images as encoded data, for a desired image, the product comprising code for: inputting an image shape as information of an image to be retrieved by search; generating query data on the basis of the input image shape; and determining similarity between the input image shape and an image, which is expected to be obtained by decoding the encoded data, by making a comparison of the query data with the encoded data read out from the storage section, wherein the encoded data is obtained by sequentially encoding transform coefficients, which are obtained by a discrete wavelet transformation of an image, and the determining step compares the query data and a synthesized sub-band, which is the logical combination of a plurality of sub-bands obtained by the discrete wavelet transformation.

Claim 22

A computer program product storing a computer readable medium having computer program code, for searching a storage section that stores a plurality of images as encoded data, for a desired image, the product comprising code for: inputting an image shape as information of an image to be retrieved by search; generating query data on the basis of the input image shape; and determining similarity between the input image shape and an image, which is expected to be obtained by decoding the encoded data, by making a comparison of the query data with the encoded data read out from the storage section, wherein the encoded data is obtained by sequentially encoding transform coefficients, which are obtained by a discrete wavelet transformation of an image, and the determining step compares the query data and a synthesized sub-band, which is the logical combination of a plurality of sub-bands obtained by the discrete wavelet transformation.

Claim 23

The computer program product according to claim 19 , wherein the plurality of sub-bands comprises LH, HH, and HL sub-bands and excludes an LL sub-band.