I want to load an image with text into OCR Xpress for C/C++/Java/Node.js. What file types are accepted – BMP or multiple file types?
OCR Xpress has three variants: “Java”, “C/C++ Linux”, and “Node.js”.
Of those three, “C/C++ Linux” and “Node.js” require a BMP as the input file type. BMP was made by Microsoft and is the only supported input file type for those two versions.
However, in OCR Xpress Java, multiple file types can be accepted because it doesn’t directly accept a file, but a Java “BufferedImage” object.
Our documentation recommends the use of Java’s Image I/O library to obtain a BufferedImage object from a file. The ImageIO.read(file) command returns a BufferedImage which can be used in OCR Xpress operations. Below is the documentation for ImageIO:
https://docs.oracle.com/javase/7/docs/api/javax/imageio/package-summary.html
Out of the box, this command has support for BMP, JPEG, and PNG.
In order to work around this limitation, you may want to look into one of our image conversion products (ImageGear, ImagXpress) to convert other image file types to BMP.
Are the Linux and Node.js versions multi-threaded toolkits? Can it perform more than one OCR at a time or does it need to do them one at a time?
Yes, the OCR Xpress public API is thread safe. There are some methods that output results to files. As long as multiple threads do not attempt to write to the same file concurrently, there should not be any concerns. All other methods are read-only.
Yes, more than one OCR can be performed at a time.