Program for manga image processing by splitting or concatenating.
The main function is optical character recognition (OCR) division with layers to crop images up to 60,000 pixels tall while preserving the text separately in the images.
-h, --help Print help information
-i, --input path to folder Path to the images to be processed
-o, --output path to folder Path to where the images will be saved
-o, --output2 path to folder Optional output path, functional only in OCR mode. Address where non-text images will be saved.
-m, --mode {s, c, ocr} Processing mode:
split (s): Dividing the image vertically by the width results in images with dimensions of width x width.
concatenate (c): Vertically concatenate images. If the height limit of a single image is exceeded, several images will be created.
split-with-ocr (ocr): The software detects the text within the image and separates it vertically into individual images without causing any harm to the text.
-l, --detect-lang { jaided.ai/easyocr/ } Language to detect (ko, en, es, ...), default = English (en)