Feature extraction and use with a probability density function (PDF) divergence metric

摘要

An image of real world is processed to identify blocks as candidates to be recognized. Each block is subdivided into sub-blocks, and each sub-block is traversed to obtain counts, in a group for each sub-block. Each count in the group is either of presence of transitions between intensity values of pixels or of absence of transition between intensity values of pixels. Hence, each pixel in a sub-block contributes to at least one of the counts in each group. The counts in a group for a sub-block are normalized, based at least on a total number of pixels in the sub-block. Vector(s) for each sub-block including such normalized counts may be compared with multiple predetermined vectors of corresponding symbols in a set, using any metric of divergence between probability density functions (e.g. Jensen-Shannon divergence metric). Whichever symbol has a predetermined vector that most closely matches the vector(s) is identified and stored.

主权项

1. A method of symbol recognition, the method comprising:
receiving a portion of an image of a scene of real world; traversing at least the portion of the image to:
identify changes in intensities of pixels in the image; andgenerate a group of counts based on the changes, without encoding positions at which the changes occur in the image; wherein an intensity of each pixel in the portion of the image is used to generate at least one count in the group of counts; wherein a size of the group of counts is predetermined; wherein at least a first count in the group of counts is incremented, when an intensity change between two pixels, in a direction of traversal, exceeds a predetermined threshold; wherein at least a second count in the group of counts is incremented, when the intensity change is positive and the intensity change does not exceed the predetermined threshold; using a measure of difference between two probability distributions, to compare a vector that is based on at least the group of counts with a plurality of predetermined vectors of corresponding symbols in a set, to identify a specific symbol therein; and storing the specific symbol in a memory, as being recognized in the image; wherein the memory is coupled to one or more processors; and wherein the traversing and the using are performed by the one or more processors.