System and method of character recognition using fully convolutional neural networks with attention
Embodiments of the present disclosure include a method that obtains a digital image. The method includes extracting a word block from the digital image. The method includes processing the word block by evaluating a value of the word block against a dictionary. The method includes outputting a prediction equal to a common word in the dictionary when a confidence factor is greater than a predetermined threshold. The method includes processing the word block and assigning a descriptor to the word block corresponding to a property of the word block. The method includes processing the word block using the descriptor to prioritize evaluation of the word block. The method includes concatenating a first output and a second output. The method includes predicting a value of the word block.
1. A system, comprising:
at least one processor; and
memory including instructions that, when executed by the at least one processor, cause the system to:
receive a digitized object, the digitized object corresponding to an image including at least one symbol;
form digitized line blocks from the digitized object by determining whitespace separators in the digitized object;
transform the line blocks into one or more word blocks separated by whitespace, the one or more word blocks having a value associated with at least one symbol forming the one or more word blocks;
resize each word block of the one or more word blocks into a canonical representation of a fixed height based off a predicted number of symbols of the image;
evaluate each word block of the one or more word blocks with a dictionary, the dictionary containing a collection of common words;
output a first word prediction when one or more of the word blocks corresponds to a word in the collection;
assign one or more qualitative descriptors to the one or more word blocks when the one or more word blocks do not correspond to the collection;
perform probabilistic correction on the one or more word blocks; and
output a second word prediction, the second word prediction corresponding to the value of the word block.
2. The system of claim 1 , wherein the instructions when executed further cause the at least one processor to assign a steering factor to the one or more word blocks when the one or more word blocks do not correspond to the collection, the steering factor being used to prioritize further evaluation of the one or more word blocks.
3. The system of claim 2 , wherein the steering factors are weight factors comprising attention weights, attention vectors, Gaussian distributions, or a combination thereof.
4. The system of claim 1 , wherein:
a first neural network evaluates the one or more word blocks against the dictionary and assigns the one or more descriptors to the one or more word blocks; and
a second neural network receives the one or more descriptors, the one or more descriptors used by the second neural network to predict the value of the one or more word blocks.
5. The system of claim 1 , wherein the one or more descriptors include handwriting style, slant, skew, rotation, blur, marker type, paper type, image fidelity, background threshold, or a combination thereof.
6. The system of claim 1 , wherein the instructions when executed further cause the at least one processor to resize each world block of the one or more word blocks into a canonical width based upon a prediction of a number of symbols in the world block.