If you have a ton of scanned PDF and other image files saved in cloud storage, how do you find what you’re looking for in a pinch? With Optical Character Recognition (OCR), it’s fast and easy. OCR extracts actual text from scanned images so all you need to do is search for your image using a term or keyword.
Dropbox eliminates the time-consuming task of processing large documents manually. Scan and store receipts, bills, photos, legal contracts, infographics from work or even your next flight itinerary. When you’re ready to access these files from Dropbox, just input a text keyword in the search bar. A short list of words or thumbnail images will pop up so that you can find your docs fast.
Just about any business can benefit from OCR technology for greater productivity. A lawyer, for example, can search for any case file simply by searching for a particular word. Banks, medical offices and advertising and marketing firms can all use OCR to streamline their businesses and virtually eliminate the need to file and store physical paper docs.
OCR stands for Optical Character Recognition or Optical Character Reader. OCR is a technology used to 'recognise' or 'read' text featured in image files. OCR either converts images of typed, handwritten or printed text from the files – including scanned documents, photos of a document, photos with signs or scenes with type or subtitles on an image – into text that machines can process and search through. Currently, Dropbox uses OCR to recognise the text of your PDFs to make them searchable by their contents.
Yes, OCR is based on machine learning. OCR technology replicates the human ability to recognise various text patterns, fonts or styles on images, documents and more. This technology is possible because of the capability of machines to learn. OCR is a prime example of the type of large scale projects involving computer vision and machine learning that engineers at Dropbox are working on all the time.
Even though they sound similar, there are major differences between OCR and ICR. OCR translates scanned images of text and turns these scans into machine-encoded text. ICR is a type of OCR but it’s more specific to text. An ICR can learn different fonts and styles of handwriting rather than just translating text from a scanned image.
OCR uses pattern or feature recognition to define each individual character in different ways, then 'reads' the text and converts it into an indexed file that can be searched.
Easily collaborate, edit and share content with team members
Backup your files and protect against data loss simply and securely
Automatically backup your photos to the cloud from any device
Take the work out of paperwork with the doc scanner app