Non-linear quantization and similarity matching methods for retrieving image data
First Claim
1. A method for constructing a database having image information representing a plurality of images, the method comprising the steps of:
- a) calculating L×
5 number of normalized edge histogram bins to generate L number of edge histograms of a target image, wherein said each edge histogram has five normalized edge histogram bins and represents a spatial distribution of five reference edges in a sub-image and L is a positive integer, wherein the reference edges include four directional edges and a non-directional edge;
b) non-linearly quantizing the L×
5 number of normalized edge histogram bins to generate L×
5 number of quantization index values for the target image;
c) storing the L×
5 number of quantization index values to the database; and
d) repeating the steps a) to c) until all of the stored imaged are processed to thereby construct the database having the image information.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for constructing a database having image information representing a plurality of images including the steps of: a) calculating L×5 number of normalized edge histogram bins to generate L number of edge histograms of a target image, wherein each edge histogram has five normalized edge histogram bins and represents a spatial distribution of five reference edges in a sub-image and L is a positive integer, wherein the reference edges include four directional edges and a non-directional edge; b) non-linearly quantizing the L×5 number of normalized edge histogram bins to generate L×5 number of quantization index values for the target image; c) storing the L×5 number of quantization index values to the database; and d) repeating the steps a) to c) until all of the stored images are processed to thereby construct the database having the image information.
98 Citations
40 Claims
-
1. A method for constructing a database having image information representing a plurality of images, the method comprising the steps of:
-
a) calculating L×
5 number of normalized edge histogram bins to generate L number of edge histograms of a target image, wherein said each edge histogram has five normalized edge histogram bins and represents a spatial distribution of five reference edges in a sub-image and L is a positive integer, wherein the reference edges include four directional edges and a non-directional edge;b) non-linearly quantizing the L×
5 number of normalized edge histogram bins to generate L×
5 number of quantization index values for the target image;c) storing the L×
5 number of quantization index values to the database; andd) repeating the steps a) to c) until all of the stored imaged are processed to thereby construct the database having the image information. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
-
20. A method for retrieving corresponding images in response to a query image based on a database, the method comprising the steps of:
-
a) calculating L number of local edge histograms of the query image as an image descriptor for the query image, wherein each local edge histogram represents a spatial distribution of five reference edges in a sub-image and L is a positive integer, wherein the reference edges include four directional edges and a non-directional edge; b) extracting a plurality of image descriptors for target images based on image information from the database, wherein the image descriptor for each target image includes L number of local edge histograms for each target image; c) comparing the image descriptor for the query image with each image descriptor for the target images to generate a comparison result; and d) retrieving at least one target image similar to the query image based on the comparison result. - View Dependent Claims (21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37)
-
-
38. A method for constructing a bit-representation of an edge histogram descriptor for a target image, comprising the steps of:
-
a) partitioning the target image into 16 sub-images, wherein each sub-image is further partitioned into S×
K number of image-blocks, S and K being positive integers;b) assigning one of five reference edges to each image-block to thereby generate 16 edge histograms for the target image, wherein the edge histogram includes the 80 edge histogram bins and the reference edges include four directional edges and a non-directional edge; c) normalizing the edge histogram bins contained in each edge histogram by S×
K to thereby generate 80 normalized edge histogram bins, each normalized edge histogram bin having a real number between 0 and 1; and
d) non-linearly quantizing 80 normalized edge histogram bins to generate 80 quantization index values as a second image descriptor for the target image. - View Dependent Claims (39, 40)
-
Specification