Optical Character Recognition

Unlock the data in your scans! Creekside Digital can run Optical Character Recognition to extract text from your raw images. While the typical output of the OCR process is a searchable PDF file with an invisible layer of text under the original document image, the text content can also be exported as separate “sidecar” files in many other formats including plain text, Excel, and XML. Creekside Digital has OCR’d millions of documents for our customers, and we have the experience and capacity to handle jobs of any size.

Indexing and Data Capture

By default, scans are sequentially named as they come off the scanner (e.g., “0000007.jpg,” “0000008.jpg,” etc. This preserves the sequence in which each document image appeared in the source book or film, which is important information by itself. However, quite often the images we create from source documents need to be indexed or named according to certain fields or data they contain, or using some predefined naming convention. Creekside Digital can perform this service for you to allow easy, rapid identification and retrieval of specific records within a group of documents.

Another common service we provide is the identification and transcription of specific fields within a given document image, such as handwritten entries on a form. This is called Data Capture. Depending on your specific requirements, Creekside Digital can manually enter these values into a spreadsheet and / or use some or all of them to name each image.

Multipage File Assembly

Often, more than one document image makes up the original record. For example, a student record file might consist of six separate pages which appear sequentially on a roll of source microfilm. Creekside Digital can create a single multipage TIFF or PDF file which contains all six document images belonging to that particular student’s record. Typically, such multipage files are named according to some unique characteristic of the record (e.g., student name or Social Security Number). This application is common with personnel records and student transcripts.

Creekside Digital can also create multipage “reader” files for 35mm newspaper microfilm conversions. These allow patrons and staff to easily scroll through multiple pages from a given roll of digitized newspapers. We adjust the number of pages in each multipage “chunk” in order to find a size that opens quickly and easily, as each project is unique.

Latest on Facebook

3 days ago

Creekside Digital

Smithsonian Mass Digitization
New feature at the DPO office: a slideshow displaying close-ups of some of the 8,000 plants from the Living Orchid Collection from Smithsonian Gardens!

Read about our project digitizing the collection at: insider.si.edu/2018/08/see-thousands-of-orchids-in-incredible-detail-in-the-smithsonians-newly-di...

On the technical size: the slideshow is running from a Raspberry Pi Zero (Github: github.com/Smithsonian/Pi-Kiosk/). One new tool we are using to display the objects from the Smithsonian collections.
... See MoreSee Less

2 weeks ago

Creekside Digital

Mass digitization is ongoing with the next phase of our Smithsonian Gardens project. We are now imaging the Archives of American Gardens (AAG) which preserves and provides access to visual resources and records that document the history of gardens in America.

Primarily a photographic archive, AAG includes documentation on over 10,000 historic and contemporary gardens. Photographic formats range from antique stereographs to recent digital images. AAG’s collections document the work of landscape architects, garden designers, and garden photographers as well as a handful of seed companies.

Examples of trade literature generated by these companies in the Archives include seed catalogs with ornately illustrated covers as seen here. Soon all these digital images will be viewable in high resolution online at Smithsonian's collections.si.edu/search/
... See MoreSee Less

Image attachment
Load more
Get Started