Asymmetric score normalization for handwritten word spotting system

A method comprising:

• receiving, by a computer, an image of a handwritten item;
• performing, by said computer, a word segmentation process on said image to produce a sub-image;
• extracting, by said computer, a set of feature vectors from said sub-image;
• computing, by said computer, a first log-likelihood score of said feature vectors using a word model having a first structure comprising a Hidden Markov Model (HMM);
• computing, by said computer, a second log-likelihood score of said feature vectors using a background model having a second structure comprising a Gaussian Mixture Model (GMM);
• computing, by said computer, a final score for said sub-image by subtracting said second log-likelihood score from said first log-likelihood score;
• comparing, by said computer, said final score against a predetermined standard to produce a word identification result;
• and
• outputting, by said computer, said word identification result.

A system comprising:

• an input/output device receiving an image of a handwritten item;
• a word segmenter operatively connected to said input/output device, said word segmenter segmenting said image to produce a sub-image;
• an extractor operatively connected to said word segmenter, said extractor extracting a set of feature vectors from said sub-image;
• a processor operatively connected to said extractor, said processor performing the following:
• computing a first log-likelihood score of said feature vectors using a word model having a first structure;
• computing a second log-likelihood score of said feature vectors using a background model having a second structure different than said first structure;
• and
• computing a final score for said sub-image by subtracting said second log-likelihood score from said first log-likelihood score;
• and
• a comparator operatively connected to said processor, said comparator comparing said final score against a predetermined standard to produce a word identification result,
• said input/output device further being operatively connected to said comparator, said input/output device outputting said word identification result.

A system comprising:

• an input/output device receiving an image of a handwritten item;
• a word segmenter operatively connected to said input/output device, said word segmenter segmenting said image to produce a sub-image;
• an extractor operatively connected to said word segmenter, said extractor extracting a set of feature vectors from said sub-image;
• a processor operatively connected to said extractor, said processor performing the following:
• computing a first log-likelihood score of said feature vectors using a word model having a first structure comprising a Hidden Markov Model (HMM);
• computing a second log-likelihood score of said feature vectors using a background model having a second structure comprising a Gaussian Mixture Model (GMM);
• and
• computing a final score for said sub-image by subtracting said second log-likelihood score from said first log-likelihood score;
• and
• a comparator operatively connected to said processor, said comparator comparing said final score against a predetermined standard to produce a word identification result,
• said input/output device further being operatively connected to said comparator, said input/output device outputting said word identification result.