init

Dockerfile

@@ -0,0 +1,32 @@
+# read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
+# you will also find guides on how best to write your Dockerfile
+
+FROM ubuntu:22.04
+
+ARG DEBIAN_FRONTEND=noninteractive
+
+RUN useradd -m -u 1000 user
+
+RUN apt-get update && apt-get install -y \
+    git \
+    curl \
+    software-properties-common \
+    python3.10 \
+    python3.10-dev \
+    && rm -rf /var/lib/apt/lists/* \
+    && apt-get remove -y --purge python3-blinker
+
+RUN apt-get update && apt-get install -y python3-opencv
+
+WORKDIR /code
+
+COPY --chown=user ./requirements.txt /code/requirements.txt
+
+RUN curl -sS https://bootstrap.pypa.io/get-pip.py | python3.10 \
+   && python3.10 -m pip install --no-cache-dir -r /code/requirements.txt 
+
+COPY --chown=user . .
+
+USER user
+
+CMD ["python3.10", "main.py"]

main.py

@@ -0,0 +1,54 @@
+import gradio as gr
+from paddleocr import PaddleOCR, draw_ocr
+import json
+import os
+import cv2
+import numpy as np
+from PIL import Image
+
+ocr_en = PaddleOCR(use_angle_cls=True, lang="en")
+ocr_ch = PaddleOCR(use_angle_cls=True, lang='ch')
+ocr_ru = PaddleOCR(use_angle_cls=True, lang='cyrillic')
+ocr_ar = PaddleOCR(use_angle_cls=True, lang='arabic')
+#ocr_ch = PaddleOCR(det_model_dir="models/det/ch/ch_PP-OCRv4_det_infer", rec_model_dir="models/rec/ch/ch_PP-OCRv4_rec_infer", cls_model_dir="models/cls/ch_ppocr_mobile_v2.0_cls_infer", rec_char_dict_path="models/dict/ppocr_keys_v1.txt", lang="ch")
+
+
+def perform_ocr(img):
+    
+    lang = "Russian"
+    if lang == "English":
+        ocr = ocr_en
+    elif lang == "Chinese (Simplified)":
+        ocr = ocr_ch
+    elif lang == "Russian" or lang == "Ukrainian":
+        ocr = ocr_ru
+    elif lang == "Arabic" or lang == "Persian":
+        ocr = ocr_ar
+
+    result = ocr.ocr(img, cls=True)
+    final_result = ""
+ 
+    image = Image.open(img).convert('RGB')
+    boxes = [line[0] for line in result]
+    txts = [line[1][0] for line in result]
+    scores = [line[1][1] for line in result]
+    im_show = draw_ocr(image, boxes, txts, scores, font_path='fonts/simfang.ttf')
+    im_show = Image.fromarray(im_show)
+    return [im_show, final_result]
+
+demo = gr.Blocks()
+
+with demo:
+    gr.Markdown("# Multilingual OCR")
+    with gr.Row():
+        with gr.Column():
+            input_image = gr.Image(source="upload", type="filepath")
+            input_radio = gr.Radio(["English", "Chinese (Simplified)", "Russian", "Ukrainian", "Arabic", "Persian"], label="Languages"),
+            input_button = gr.Button("Run!")
+        with gr.Column():
+            output_image = gr.Image()
+            output_text = gr.Textbox(label="Results")
+
+    input_button.click(fn=perform_ocr, inputs=[input_image], outputs=[output_image, output_text])
+
+demo.launch(server_name="0.0.0.0", server_port=7860)

requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+uvicorn
+gradio
+paddleocr==2.7.0.2
+paddlepaddle==2.4.2