I have a set of images of a id card where i need to extract name and date of birth. for creating a dataset i have converted the set of images to graph for that i have used ocr to get each word and its coordinates. in this each word is the node i have made connection to another node based on direct contact to other node by that i have a graph for all the images i have a graph and to creating a adj matrix i used sparse diagonal matrix of each graph adj matrix. for feature vector at each node i used word2vec at words in each node to create a numpy array. and the labels is the tag that i assigned that could be categorical data. so i want to know that am i doing correct or any changes to be needed...?
thank you
I have a set of images of a id card where i need to extract name and date of birth. for creating a dataset i have converted the set of images to graph for that i have used ocr to get each word and its coordinates. in this each word is the node i have made connection to another node based on direct contact to other node by that i have a graph for all the images i have a graph and to creating a adj matrix i used sparse diagonal matrix of each graph adj matrix. for feature vector at each node i used word2vec at words in each node to create a numpy array. and the labels is the tag that i assigned that could be categorical data. so i want to know that am i doing correct or any changes to be needed...?
thank you