We are a small team of technology enthusiasts (with an unfairly large number of PhDs between us) and we are based in Islington, North London, UK.
Our expertise lies in building complex computer-vision software – primarily for PDFs.
Zanran’s first product is its data-search engine (zanran.com).
Utilising the technology we developed for Zanran’s data-search engine, we have managed to create a number of PDF extraction solutions – all of which you can discover here.
Zanran is privately held by its two founders and three angel investors.
Zanran’s unique software ‘understands’ the layout of a PDF page – paragraphs, diagrams and tables
Search the web for data & statistics
Our two founders of Zanran, Jon & Yves, exited their telecoms businesses – but continued to meet. During one of their ‘creative’ sessions, they imagined how one day, a user would be able to enter the X and Y labels of a graph in a search engine and get the resulting pie chart, bar chart, line graph or table. Thus, Zanran.com - the data-search engine - was born!
They understood then that building a numerical search engine capable of this relied on the development of computer-vision algorithms which were capable of filtering images of graphs and tables from other images.
Building the data-search engine, Zanran.com, relied not only on the expertise and brilliance of the team we were working with, but also on a great deal of stubbornness, grit and determination, as well as the timely advent of advent of Amazon Web Services (which made it viable for a small company to run many hundreds of servers at a time).
So why PDFs? PDFs contain large amounts of graphs and tables and it’s the primary format organisations tend to use for major reports and archiving. As a result, we have developed a considerable amount of expertise in developing software for handling PDFs and the content within them – both graphics and text.
At Zanran, we now provide PDF analysis, processing and extraction software to organisations and businesses wanting to process PDFs at large scale. To get a better understanding of the software we provide and the technology we use, click here.
Our software is not conventional – nor do we work with conventional programmers to produce it. As computer-vision is still at a relatively early stage in its development, there can be many blind alleys and false starts in finding good solutions to any vision problem. The remarkable thing is that the brain – human/cat/dog/mouse - is capable of dealing with these problems so fast and so apparently easily. Our aim is to replicate that process – and to that end, we’re constantly improving our unique computer-vision algorithms.