ARC is often associated with providing print services for the architecture and construction industries. This is to be expected considering ARC’s blueprint roots, but some do not know that ARC has been building and offering technologies to improve construction industry workflows for over a decade.
Powering this vision is ARC’s Chief Technology Officer and a prominent computer engineer, Rahul Roy. Anyone that has seen Roy give a presentation knows that he understands and works with technology at that the level you’d expect from any Silicon Valley tech leaders. Here are 5 highlights from a passionate talk he gave discussing tech in and out of the construction industry in his own words.
OCR (Optical Character Recognition) and The Key To Auto-Hyperlinking
OCR research started 40 years ago. Researchers today still haven’t even scratched the surface. OCR research is bigger than any ocean you can think of.
The most well known OCR engine is—Tesseract, the software Google uses to extract text . But it’s not Google’s product. Tesseract is HP’s product. HP Labs started that in 1984. In 1984, they hired a bunch of PHDs to do a research project, and they funded that in HP Lab. They didn’t know the application of the OCR engine. So in 2004, they made it open source. In 2005, Google started researching and modified the engine. Today, Google has 800 languages from Tesseract.
We knew OCR was the key to extracting information from construction drawing documents. But when we first ran Tesseract to do auto-hyperlinking, it failed everywhere. It would detect all kinds of circles. It cannot detect the graphics inside the circles—either it couldn’t detect the graphics or adjust to the text—but we used all the algorithms and nothing. So then we started to write our own algorithms, detection began to work. It was huge.
The Importance of Automatic Hyperlinking
Simply put, automating the manual hyperlinks saves a lot of time for the architects. As does indexing online documents, which we do.
The problem is that the OCR standards, ABBYY and Tesseract , are unable to extract text from callouts. Therefore, there was no solution for auto-hyperlinking except for doing it manually.
Our engineers built their own engine in-house to overcome this problem. The custom engine allows the technology to dynamically divide character images for accurate information extraction.
Google would never go to these lengths to solve the OCR limitations because they don’t care to that level. We do. We are AEC (architecture, engineering, and construction). We care.
What Makes ARC Different from Other Technology Providers
ARC’s biggest differentiation from other construction technology providers is SKYSITE. The cloud solution uses a Microsoft .NET and open source hybrid foundation. Open source gives you the scalability and the speed. Net for maintainability and interoperability. Mostly our competitors will do either .NET or open source. They won’t do a hybrid, which you need engineers not programmers to build.
The Difference between Programmers and Engineers
Most people don’t understand there’s a difference between an engineer and a programmer. Even some programmers or engineers may not know the difference. An example is the .NET framework. Computer engineers created that. Programmers can use .NET to write an application, but they cannot engineer with that service. The difference is dependent on creating the very tools used to program.
Why Information Technology Matters for the Construction Industry
The value is very simple. We are using this technology so it is easy for people to use. So they don’t have to do brain surgery to figure it out. What our tech does is make your pages move fast. Your pages light. Your speed is there. Your view is clearer. Your content comes quickly inside the page. Security and scalability are there. To bring all that stuff together, you need to have this kind of recipe behind it.