from typing import List import pytesseract from PIL import Image import gradio as gr def tesseract_ocr(filepath: str, languages: List[str] = None): if not languages: languages = ['eng'] image = Image.open(filepath) lang_param = '+'.join(languages) return pytesseract.image_to_string(image=image, lang=lang_param) title = "Tesseract OCR" description = "Gradio demo for Tesseract. Tesseract is an open source text recognition (OCR) Engine." article = "

Tesseract documentation | Github Repo

" examples = [ ["examples/weird_unicode_math_symbols.png", []], ["examples/eurotext.png", ["eng"]], ["examples/tesseract_sample.png", ["jpn", "eng"]], ["examples/chi.jpg", ["chi_sim", "chi_tra"]], ] with gr.Blocks(title=title) as demo: gr.Markdown(f'

{title}

') gr.Markdown(description) with gr.Row(): with gr.Column(): image = gr.Image(type="filepath", label="Input") language_choices = pytesseract.get_languages(config='') # <- safer with gr.Accordion("Languages", open=False): languages = gr.CheckboxGroup(language_choices, value=["eng"], label="Language(s)") with gr.Row(): btn_clear = gr.ClearButton([image, languages]) btn_submit = gr.Button(value="Submit", variant="primary") with gr.Column(): text = gr.Textbox(label="Output") btn_submit.click(tesseract_ocr, inputs=[image, languages], outputs=text) btn_clear.add(text) gr.Examples( examples=examples, inputs=[image, languages], outputs=[text], fn=tesseract_ocr ) gr.Markdown(article) if __name__ == '__main__': demo.queue().launch()