IMAGE SIGNATURE EXTRACTION DEVICE

• US 20120027309A1
• Filed: 04/12/2010
• Published: 02/02/2012
• Est. Priority Date: 04/14/2009
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Oami, Ryoma, Iwamoto, Kota
• Granted Patent: US 8,861,871 B2
• Time in Patent Office: Days
• US Class Current: 382/203
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 10/421 by analysing segments intersecting the pattern
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis

Abstract

The image signature extraction device includes an extraction unit and a generation unit. The extraction unit extracts region features from respective sub-regions in an image in accordance with a plurality of pairs of sub-regions in the image, the pairs of sub-regions including at least one pair of sub-regions in which both a combination of shapes of two sub-regions of the pair and a relative position between the two sub-regions of the pair differ from those of at least one of other pairs of sub-regions, and being classified into a plurality of types based on a combination of shapes of two sub-regions and a relative position between the two sub-regions of each of the pairs. The generation unit generates an image signature to be used for identifying the image based on the extracted region features of the respective sub-regions.

Claims

1-83. -83. (canceled)

84. An image signature extraction device, comprising:

a calculation unit that calculates region features from two sub-regions in an image, the two sub-regions being associated with each dimension composing an image signature, the image signature being information for identifying an image; and

a quantization unit that quantizes difference values between the region features, using a parameter defined independently for each type, wherein

the dimension is categorized into one of the types different from a type of at least one other dimension.

85. The image signature extraction device, according to claim 84, wherein

the quantization unit quantizes the difference values, using a threshold defined independently for each type.

86. The image signature extraction device, according to claim 85, further comprising

a determination unit that determines the threshold of each type based on distribution of the difference values associated with dimensions that are categorized in the same type.

87. The image signature extraction device, according to claim 86, wherein

the determination unit sorts absolute values of the difference values associated with dimensions that are categorized in the same type, and determines a value located at a predetermined proportion from top or bottom thereof as the threshold of each type.

88. The image signature extraction device, according to claim 84, wherein

the dimensions are categorized into the types by their patterns of sub-regions, based on a combination of shapes of the two sub-regions and a relative position between them.

89. The image signature extraction device, according to claim 84, wherein

the two sub-regions are a first sub-region and a second sub-region;

and

the types include at least one of;

a first type in which two rectangles, formed by evenly dividing a rectangle of a certain size in the image into two in a vertical direction or a horizontal direction, serve as the first sub-region and the second sub-region;

a second type in which among four rectangles formed by evenly dividing a rectangle of a certain size in the image into two in a vertical direction and into two in a horizontal direction, a region formed by combining an upper left rectangle and a lower right rectangle serves as the first sub-region and a region formed by combining an upper right rectangle and a lower left rectangle serves as the second sub-region;

a third type in which the first sub-region and the second sub-region are rectangles of the same size and are spaced apart in a vertical direction and in a horizontal direction by a distance equal to an integral multiple of one side of the rectangle;

a fourth type in which the first sub-region and the second sub-region are rectangles of the same size but are smaller than those of the third type, and are spaced apart in a vertical direction and in a horizontal direction by a distance equal to an integral multiple of one side of the rectangle;

a fifth type in which of two different sized rectangles sharing the center of gravity, a region obtained by removing the smaller rectangle from the larger rectangle serves as the first sub-region, and the smaller rectangle serves as the second sub-region;

a sixth type in which two rectangles of the same size sharing the center of gravity and placed orthogonally to each other serve as the first sub-region and the second sub-region; and

a seventh type in which among three vertically or horizontally adjacent rectangles of the same size, rectangles at both ends serve as the first sub-region and a rectangle at the center serves as the second sub-region.

90. The image signature extraction device, according to claim 89, wherein

the two sub-regions are formed of one or more blocks among 1024 blocks obtained by dividing the image into 32 in a vertical direction and dividing the image into 32 in a horizontal direction, and

the rectangle of a certain size of the first type is a rectangle constituted of 4 blocks vertically and 4 blocks horizontally;

the rectangle of a certain size of the second type is a rectangle constituted of 8 blocks vertically and 8 blocks horizontally;

the rectangle of the third type is a rectangle constituted of 10 blocks vertically and 10 blocks horizontally;

the rectangle of the fourth type is a rectangle constituted of 6 blocks vertically and 6 blocks horizontally;

the two different sized rectangles of the fifth type are a combination of a rectangle constituted of 10 blocks vertically and 10 blocks horizontally and a rectangle constituted of 4 blocks vertically and 4 blocks horizontally, or a combination of a rectangle constituted of 13 blocks vertically and 13 blocks horizontally and a rectangle constituted of 3 blocks vertically and 3 blocks horizontally;

the two rectangles of the sixth type are a rectangle constituted of 10 blocks vertically and 6 blocks horizontally and a rectangle constituted of 6 blocks vertically and 10 blocks horizontally; and

the rectangle of the seventh type is a rectangle constituted of 4 blocks vertically and 4 blocks horizontally.

91. A matching device that performs matching using an image signature output by the image signature extraction device according to claim 84.

92. An identifying device that performs identification using an image signature output by the image signature extraction device according to claim 84.

93. An image signature extraction method, comprising:

calculating region features from two sub-regions in an image, the two sub-regions being associated with each dimension composing an image signature, the image signature being information for identifying an image; and

quantizing difference values between the region features, using a parameter defined independently for each type, wherein

the dimension is categorized into one of the types different from a type of at least one other dimension.

94. The image signature extraction method, according to claim 93, wherein

the quantizing includes quantizing the difference values, using a threshold defined independently for each type.

95. The image signature extraction method, according to claim 94, further comprising

determining the threshold of each type based on distribution of the difference values associated with dimensions that are categorized in the same type.

96. The image signature extraction method, according to claim 95, wherein

the determining includes sorting absolute values of the difference values associated with dimensions that are categorized in the same type, and determining a value located at a predetermined proportion from top or bottom thereof as the threshold of each type.

97. The image signature extraction method, according to claim 93, wherein

the dimensions are categorized into the types by their patterns of sub-regions, based on a combination of shapes of the two sub-regions and a relative position between them.

98. The image signature extraction method, according to claim 93, wherein

the two sub-regions are a first sub-region and a second sub-region;

and

the types include at least one of;

a first type in which two rectangles, formed by evenly dividing a rectangle of a certain size in the image into two in a vertical direction or a horizontal direction, serve as the first sub-region and the second sub-region;

a second type in which among four rectangles formed by evenly dividing a rectangle of a certain size in the image into two in a vertical direction and into two in a horizontal direction, a region formed by combining an upper left rectangle and a lower right rectangle serves as the first sub-region and a region formed by combining an upper right rectangle and a lower left rectangle serves as the second sub-region;

a third type in which the first sub-region and the second sub-region are rectangles of the same size and are spaced apart in a vertical direction and in a horizontal direction by a distance equal to an integral multiple of one side of the rectangle;

a fourth type in which the first sub-region and the second sub-region are rectangles of the same size but are smaller than those of the third type, and are spaced apart in a vertical direction and in a horizontal direction by a distance equal to an integral multiple of one side of the rectangle;

a fifth type in which of two different sized rectangles sharing the center of gravity, a region obtained by removing the smaller rectangle from the larger rectangle serves as the first sub-region, and the smaller rectangle serves as the second sub-region;

a sixth type in which two rectangles of the same size sharing the center of gravity and placed orthogonally to each other serve as the first sub-region and the second sub-region; and

a seventh type in which among three vertically or horizontally adjacent rectangles of the same size, rectangles at both ends serve as the first sub-region and a rectangle at the center serves as the second sub-region.

99. The image signature extraction method, according to claim 98, wherein

the two sub-regions are formed of one or more blocks among 1024 blocks obtained by dividing the image into 32 in a vertical direction and dividing the image into 32 in a horizontal direction, and

the rectangle of a certain size of the first type is a rectangle constituted of 4 blocks vertically and 4 blocks horizontally;

the rectangle of a certain size of the second type is a rectangle constituted of 8 blocks vertically and 8 blocks horizontally;

the rectangle of the third type is a rectangle constituted of 10 blocks vertically and 10 blocks horizontally;

the rectangle of the fourth type is a rectangle constituted of 6 blocks vertically and 6 blocks horizontally;

the two different sized rectangles of the fifth type are a combination of a rectangle constituted of 10 blocks vertically and 10 blocks horizontally and a rectangle constituted of 4 blocks vertically and 4 blocks horizontally, or a combination of a rectangle constituted of 13 blocks vertically and 13 blocks horizontally and a rectangle constituted of 3 blocks vertically and 3 blocks horizontally;

the two rectangles of the sixth type are a rectangle constituted of 10 blocks vertically and 6 blocks horizontally and a rectangle constituted of 6 blocks vertically and 10 blocks horizontally; and

the rectangle of the seventh type is a rectangle constituted of 4 blocks vertically and 4 blocks horizontally.

100. A matching method for performing matching using an image signature extracted by the image signature extraction method according to claim 93.

101. An identifying method for performing identification using an image signature extracted by the image signature extraction method according to claim 93.

102. A computer-readable medium storing a program comprising instructions for causing a computer to function as:

a calculation unit that calculates region features from two sub-regions in an image, the two sub-regions being associated with each dimension composing an image signature, the image signature being information for identifying an image; and

a quantization unit that quantizes difference values between the region features, using a parameter defined independently for each type, wherein

the dimension is categorized into one of the types different from a type of at least one other dimension.