app.py

import streamlit as st
import numpy as np
import pandas as pd
import easyocr
from PIL import Image
from streamlit_drawable_canvas import st_canvas

def rectangle(image, result):
    """Draw rectangles on image based on predicted coordinates using PIL."""
    from PIL import ImageDraw
    draw = ImageDraw.Draw(image)
    for res in result:
        top_left = tuple(res[0][0])
        bottom_right = tuple(res[0][2])
        draw.rectangle((top_left, bottom_right), outline="blue", width=2)
    return image

def main():
    # Set up Streamlit page
    st.set_page_config(
        page_title="OCR App",
        page_icon=":mag:",
        layout="centered",
        initial_sidebar_state="auto",
    )

    st.title("Optical Character Recognition (OCR) with EasyOCR")
    
    st.markdown("""
    Upload an image or use the canvas to draw, and the app will recognize and extract text from it.
    Supported languages for recognition include English and Japanese.
    """)
    
    # Upload image or draw
    col1, col2 = st.columns(2)
    with col1:
        file = st.file_uploader("Upload Image", type=['png', 'jpg', 'jpeg'])
    with col2:
        stroke_width = st.slider("Stroke Width: ", 1, 25, 3)
        canvas_result = st_canvas(
            fill_color="rgba(255, 165, 0, 0.3)",
            stroke_width=stroke_width,
            stroke_color="#ffffff",
            background_color="#000000",
            update_streamlit=True,
            width=400,
            height=400,
            drawing_mode="freedraw",
            key="canvas",
        )

    # Process uploaded image or drawing
    if file is not None:
        image = Image.open(file).convert('RGB')
    elif canvas_result.image_data is not None:
        image = Image.fromarray(np.array(canvas_result.image_data).astype('uint8'), 'RGBA').convert('RGB')
    else:
        st.warning("Please upload an image or use the canvas to draw.")
        st.stop()

    st.image(image, caption='Uploaded Image', use_column_width=True)
    
    # OCR
    reader = easyocr.Reader(['en', 'ja'], gpu=False)
    result = reader.readtext(np.array(image))
    
    # Display results
    if result:
        image = rectangle(image, result)
        st.image(image, caption='Processed Image with Detected Text', use_column_width=True)
        
        textdic_easyocr = {idx: {'text': res[1], 'confidence': res[2]} for idx, res in enumerate(result)}
        df = pd.DataFrame.from_dict(textdic_easyocr, orient='index')
        st.dataframe(df)
    else:
        st.info("No text detected.")

if __name__ == "__main__":
    main()