app.py

import atexit
import base64
import os
import shutil
import tempfile
import time

import fitz
import gradio as gr
import spaces
import torch
from PIL import Image, ImageEnhance
from transformers import AutoModel, AutoTokenizer

model_name = "ucaslcl/GOT-OCR2_0"
device = "cuda" if torch.cuda.is_available() else "cpu"

tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
model = AutoModel.from_pretrained(model_name, trust_remote_code=True, device_map=device)
model = model.eval().to(device)

# 创建一个持久的临时目录
TEMP_DIR = tempfile.mkdtemp()


def cleanup():
    """清理临时目录"""
    shutil.rmtree(TEMP_DIR, ignore_errors=True)


# 确保在程序退出时清理临时目录

atexit.register(cleanup)


def pdf_to_images(pdf_path):
    images = []
    pdf_document = fitz.open(pdf_path)
    for page_num in range(len(pdf_document)):
        page = pdf_document.load_page(page_num)
        zoom = 10  # 增加缩放比例到10
        mat = fitz.Matrix(zoom, zoom)
        pix = page.get_pixmap(matrix=mat, alpha=False)
        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)

        # 增对比度
        enhancer = ImageEnhance.Contrast(img)
        img = enhancer.enhance(1.5)  # 增加50%的对比度

        images.append(img)
    pdf_document.close()
    return images


@spaces.GPU()
def convert_pdf_to_images(file):
    if file is None:
        return "错误:未提供文件", None

    try:
        if not file.name.lower().endswith(".pdf"):
            return "错误:请上传PDF文件", None

        images = pdf_to_images(file.name)
        image_paths = []

        for i, image in enumerate(images):
            img_path = os.path.join(TEMP_DIR, f"page_{i+1}.png")
            image.save(img_path, "PNG")
            image_paths.append(img_path)

        return "PDF转换为图片成功", image_paths
    except Exception as e:
        return f"错误: {str(e)}", None


@spaces.GPU()
def ocr_process(image, got_mode, ocr_color="", ocr_box="", progress=gr.Progress()):
    if image is None:
        return "错误:未选择图片"

    try:
        progress(0, desc="开始处理...")

        # 模拟OCR处理的不同阶段
        progress(0.2, desc="图像预处理...")
        time.sleep(0.5)

        progress(0.4, desc="文字识别中...")
        time.sleep(0.5)

        progress(0.6, desc="后处理...")
        time.sleep(0.5)

        result = process_single_image(image, got_mode, ocr_color, ocr_box)

        progress(0.8, desc="生成结果...")
        time.sleep(0.5)

        progress(1, desc="处理完成")
        return result
    except Exception as e:
        return f"错误: {str(e)}"


def process_single_image(image_path, got_mode, ocr_color, ocr_box):
    result_path = f"{os.path.splitext(image_path)[0]}_result.html"

    if "plain" in got_mode:
        if "multi-crop" in got_mode:
            res = model.chat_crop(tokenizer, image_path, ocr_type="ocr")
        else:
            res = model.chat(tokenizer, image_path, ocr_type="ocr", ocr_box=ocr_box, ocr_color=ocr_color)
        return res
    elif "format" in got_mode:
        if "multi-crop" in got_mode:
            res = model.chat_crop(tokenizer, image_path, ocr_type="format", render=True, save_render_file=result_path)
        else:
            res = model.chat(tokenizer, image_path, ocr_type="format", ocr_box=ocr_box, ocr_color=ocr_color, render=True, save_render_file=result_path)

        if os.path.exists(result_path):
            with open(result_path, "r", encoding="utf-8") as f:
                html_content = f.read()
            encoded_html = base64.b64encode(html_content.encode("utf-8")).decode("utf-8")
            data_uri = f"data:text/html;charset=utf-8;base64,{encoded_html}"
            preview = f'<iframe src="{data_uri}" width="100%" height="600px"></iframe>'
            download_link = f'<a href="{data_uri}" download="result.html">下载完整结果</a>'
            return f"{download_link}\n\n{preview}"

    return "错误: 未知的OCR模式"


with gr.Blocks() as demo:
    gr.Markdown("# OCR 图像识别")

    with gr.Row():
        with gr.Column(scale=1):
            pdf_input = gr.File(label="上传PDF文件")
            convert_button = gr.Button("转换PDF为图片")

        with gr.Column(scale=2):
            image_gallery = gr.Gallery(label="图片预览", columns=3)

    with gr.Row():
        with gr.Column(scale=1):
            selected_image = gr.State(value=None)  # 使用 gr.State 来存储选中的图片路径
            preview_image = gr.Image(label="选中的图片", type="filepath")
            got_mode = gr.Dropdown(
                choices=[
                    "plain texts OCR",
                    "format texts OCR",
                    "plain multi-crop OCR",
                    "format multi-crop OCR",
                    "plain fine-grained OCR",
                    "format fine-grained OCR",
                ],
                label="OCR模式",
                value="plain texts OCR",
            )
            ocr_color = gr.Textbox(label="OCR颜色 (仅用于fine-grained模式)")
            ocr_box = gr.Textbox(label="OCR边界框 (仅用于fine-grained模式)")
            ocr_button = gr.Button("开始OCR识别")

        with gr.Column(scale=2):
            ocr_output = gr.HTML(label="识别结果")

    def select_image(evt: gr.SelectData, gallery):
        selected = gallery[evt.index]
        return selected, selected

    image_gallery.select(select_image, image_gallery, [selected_image, preview_image])
    convert_button.click(convert_pdf_to_images, inputs=[pdf_input], outputs=[gr.Textbox(visible=False), image_gallery])
    ocr_button.click(ocr_process, inputs=[selected_image, got_mode, ocr_color, ocr_box], outputs=ocr_output)

if __name__ == "__main__":
    demo.launch()