Method, apparatus and system for generating a feature vector
A method of generating a feature vector for an image is disclosed. Values are determined from a plurality of points in a region of the image, each of the values being determined using at least two of the plurality of points. A periodic sequence of the determined values is determined based on an order of the plurality of points. The periodic sequence is phase variant to a starting point of the ordered plurality of points, the order of the plurality of points being determined according to a predetermined rule. The feature vector for one of the points is generated from a frequency domain representation of the periodic sequence, the feature vector being invariant to rotation with respect to the plurality of points.
1. A method of generating a feature vector for an image, the method comprising:
determining values from a plurality of points in a region of the image, each of the values being determined using at least two of the plurality of points;
determining a periodic sequence of the determined values based on an order of the plurality of points, the periodic sequence being phase variant to a starting point of the ordered plurality of points, the order of the plurality of points being determined according to a predetermined rule; and
generating the feature vector for one of the points from a frequency domain representation of the periodic sequence, the feature vector being invariant to rotation with respect to the plurality of points.
2. The method according to claim 1 , wherein the feature vector comprises magnitude components of a Fourier transform.
3. The method according to claim 1 , wherein the feature vector represents a second order difference of phase.
4. The method according to claim 3 , further comprising concatenating the second order difference of phase with magnitude values.
5. The method according to claim 1 , wherein the determined values represent areas of normalized triangles.
6. The method according to claim 1 , wherein the points are locations of words.
7. The method according to claim 1 , wherein the points are locations of letter extenders.
8. The method according to claim 1 , wherein the points are locations of letter descenders.
9. The method according to claim 1 , further comprising determining further periodic sequences.
10. The method according to claim 9 , further comprising generating further feature vectors and concatenating the further feature vectors to produce a large feature vector.
11. The method according to claim 1 , further comprising: matching the generated feature vector with one or more other feature vectors; accumulating scores based on matching feature vectors; and identifying a document based on the accumulated scores.
12. An apparatus for generating a feature vector for an image, the apparatus comprising:
module for determining values from a plurality of points in a region of the image, each of the values being determined using at least two of the plurality of points;
module for determining a periodic sequence of the determined values based on an order of the plurality of points, the periodic sequence being phase variant to a starting point of the ordered plurality of points, the order of the plurality of points being determined according to a predetermined rule; and
module for generating the feature vector for one of the points from a frequency domain representation of the periodic sequence, the feature vector being invariant to rotation with respect to the plurality of points.
13. A system for generating a feature vector for an image, the system comprising:
a memory for storing data and a computer program;
a processor coupled to the memory for executing said computer program, said computer program comprising instructions for:
code for determining values from a plurality of points in a region of the image, each of the values being determined using at least two of the plurality of points;
code for determining a periodic sequence of the determined values based on an order of the plurality of points, the periodic sequence being phase variant to a starting point of the ordered plurality of points, the order of the plurality of points being determined according to a predetermined rule; and
code for generating the feature vector for one of the points from a frequency domain representation of the periodic sequence, the feature vector being invariant to rotation with respect to the plurality of points.
14. A non-transitory computer readable medium having a computer program stored thereon for generating a feature vector for an image, the program comprising:
code for determining values from a plurality of points in a region of the image, each of the values being determined using at least two of the plurality of points;
code for determining a periodic sequence of the determined values based on an order of the plurality of points, the periodic sequence being phase variant to a starting point of the ordered plurality of points, the order of the plurality of points being determined according to a predetermined rule; and
code for generating the feature vector for one of the points from a frequency domain representation of the periodic sequence, the feature vector being invariant to rotation with respect to the plurality of points.