Facial recognition technology was first developed in the 1960’s and has steadily progressed in its capabilities and accuracy. NeuralVision Technologies is a software company based in Boston MA which has developed novel techniques for leveraging the core technology of facial recognition (“computer vision”) for document recognition and document classification.
Why is this important, you might ask? Well, many documents are scanned images which are difficult to process using text-based methods (the predominant method in the industry). As a result, manual review of these documents has been required in order to find critical records or to classify a large collection of documents by their “record category” or “document type”. Eliminating manual review can save time and money and automate a process which is otherwise a multi-step, labor intensive effort.
NeuralVision Technologies (NVT) generates visual signatures of documents, similar to the way a human sees and remembers what a document looks like. This signature is used to search and find documents which have similar signatures, or used to classify a large collection of documents by document type. NVT doesn’t rely on text nor require any training. Applications of this technology include creating an inventory of document type, in order to slice across collections (such as medical records or MBS loan files), to apply taxonomies for governance purposes, or to support investigations where specific types of records need to be located.
For more information, visit: www.neuralvision.net.