This extracts the lines of text identified in the input image and saves them to the lines directory. This is useful for debugging text line recognition by exporting the images that will be passed to the line recognition model. Note that these images are not currently rescaled to 64px high as the actual recognition model inputs are.
This extracts the lines of text identified in the input image and saves them to the
lines
directory. This is useful for debugging text line recognition by exporting the images that will be passed to the line recognition model. Note that these images are not currently rescaled to 64px high as the actual recognition model inputs are.