A Guide to TopOCR's Graphical User Interface
Click on any of the links below for more information on a particular topic.
TopOCR Image Window
TopOCR Text Window
Document Camera Image Alignment
In order to get the highest possible OCR accuracy, the document camera as well as the documents need to be "aligned", in other words, "to be as straight and parallel with each other as possible".
Aligning the document camera and document is a 2-step process.
First, align the center of the document camera to the center of the Document Camera Rectangle.
Then align your document in the Document Capture Rectangle, keeping the edges as straight as possible.
You can either visually align documents to be straight, by using the the DocCam Image Preview Dialog or you can use software functions to automatically straighten the images.
TopOCR provides two separate functions to correct for commonly encountered image distortion like rotation or page curl.
Neural Warp is a neural network that will take any document camera text image and automatically correct for both 3D and 2D image distortion for perspective, page curl, rotation, lighting and background.
Straighten Columns is a 2D text line tracking function combined with a sophisticated curve fitting function to straighten lines of 2D text.
You can select either of these functions to use automatically every time you scan an image with a document camera with the DocCam Image Preview Dialog These functions automatically create a "corrected" text aligned output image ready for OCR, and can greatly improve OCR accuracy, in some cases by as much as 40%-50%!
Raw Input Image
|Auto-Rotated Raw Binarized Rotated Image||Auto-Rotated Image Straightened With Neural Warp||Auto-Rotated Image Straightened With Column Straighten|
Using TopOCR 58 with a RAM Disk
TopOCR's TAO OCR function can be optimized by use of a RAM Disk.
This can also be a useful feature on flash memory disk file systems used by low cost tablets and stick PCs, because a RAM Disk can reduce the "wear" on a flash file system.
The TopOCR Reader distribution flash drive includes a ramdisk installation directory that is a derivitive of imdisk. It contains a copy of the imdisk toolkit called "ImDiskTk".
This optional step takes less than a minute to complete. First, install ImDiskTk and then run the configuration and formatting tool and create a ram disk with the drive letter of "Z:" and a drive label of "TopOCR". We recommend a 32 MB ram disk. You need to create the ram disk with these parameters in order for it to work correctly with TopOCR, otherwise it will simply default to your hard drive. Please note that when Windows performs an update it may delete your RAM Disk so you will have to reconfigure and format it.
Deleting The Ram Disk
If you want to delete the TopOCR Ram Disk, then in the Windows File Manager, select the "TopOCR (Z:)" drive and and then "right-click" on it with your mouse and then select "Unmount ImDisk Virtual Disk".