Slicing tool for tesseract.
Sometimes tesseract can give sub-optimal results for multi-column text. This tool is designed for the user to select custom regions of an image and feed them to tesseract, so a more specific output can be attained. The ocr output ends up in the clipboard.
Designed and tested on Linux. Presumably it will also work on Windows if tesseract is in the path.
Click "Browse" to select an image. Click "Load" to load the image into the program. Once loaded, select a rectangle with mouse's left and right buttons, where each button corresponds to a corner of the rectangle. Click "Slice to clipboard" to run tesseract on the selected region; the output of tesseract will be placed in the clipboard. Paste into your favorite editor. The clipboard output usually has a trailing invisibile character, so hit backspace once after paste.