Add application file

Dockerfile

@@ -0,0 +1,31 @@
+# Use official python slim image
+FROM python:3.10-slim
+
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    libgl1-mesa-glx \
+    libglib2.0-0 \
+    libsm6 \
+    libxext6 \
+    libxrender-dev \
+    ffmpeg \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements.txt and install Python deps
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy app
+COPY app.py .
+
+# Create uploads folder
+RUN mkdir uploads
+
+# Expose port (Hugging Face uses 7860 by default, but FastAPI typically uses 8000)
+EXPOSE 8000
+
+# Use command to run FastAPI with uvicorn on 0.0.0.0 to accept external connections
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -0,0 +1,136 @@
+import os
+import shutil
+import cv2
+import numpy as np
+import matplotlib.pyplot as plt
+import pandas as pd
+from paddleocr import PaddleOCR
+from fastapi import FastAPI, UploadFile, File
+from fastapi.responses import FileResponse, JSONResponse
+
+from fastapi.middleware.cors import CORSMiddleware
+
+app = FastAPI()
+
+origins = [
+    "http://localhost.tiangolo.com",
+    "https://localhost.tiangolo.com",
+    "http://localhost",
+    "http://localhost:8080",
+]
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=['*'],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# Initialize PaddleOCR
+# ocr = PaddleOCR(use_angle_cls=True, lang="en", det_db_box_thresh=0.5)
+
+@app.get("/")
+def read_root():
+    return JSONResponse({
+        "message": "Hello World"
+    })
+
+# Load Image
+@app.post("/process_image")
+def load_image(file: UploadFile = File(...)):
+    ocr = PaddleOCR(use_angle_cls=True, lang='en')
+    image_path = f"uploads/{file.filename}"
+    os.makedirs("uploads", exist_ok=True)
+
+    with open(image_path, "wb") as f:
+        shutil.copyfileobj(file.file, f)
+
+    image = cv2.imread(image_path)
+    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+    height, width, channels = image.shape
+
+    # OCR Processing
+    ocr_results = ocr.ocr(image_path)
+    print(ocr_results)
+
+    extracted_text = []
+    page = ocr_results[0]
+    for block in page:
+        print(block)
+    # Lists of recognized texts and their bounding boxes
+    texts = page['rec_texts']
+    boxes = page['dt_polys']
+    scores = page['rec_scores']
+    print(texts)
+
+    # Zip them together
+    text_and_boxes = list(zip(texts, boxes, scores))
+    final_text = list(zip(texts, scores))
+
+    # Display all results
+    for text, box, score in text_and_boxes:
+        print(f"Text: {text}")
+        print(f"Bounding Box: {box.tolist()}")  # Convert numpy array to regular list
+        print(f"Score: {score}")
+        print("---")
+        extracted_text.append((text, score))
+
+
+    # Print Extracted Text
+    print("🔹 Extracted Text from Invoice:")
+    for text, score in extracted_text:
+        print(f"{text} (Confidence: {score:.2f})")
+
+    # Create a simple dataframe from all OCR text
+    all_text = [text for text, _ in extracted_text]
+    print("\n🔹 Creating a simple data structure from all OCR text")
+    df = pd.DataFrame({'text': all_text})
+    print(df.head())
+    df.to_csv("invoice_extracted_text.csv", index=False)
+
+    # Display Image with OCR Text Overlay
+    plt.figure(figsize=(10, 10))
+    plt.imshow(image)
+
+    # Draw bounding boxes and text annotations
+    for text, box, score in text_and_boxes:
+
+        # y_offset = int(0.03 * height)  # 5% downward shift
+        y_offset = 0
+        print(height)
+        corrected_box = [(x, y + y_offset) for (x, y) in box]
+
+        # Draw bounding box
+        plt.plot(
+            [corrected_box[0][0], corrected_box[1][0], corrected_box[2][0], corrected_box[3][0], corrected_box[0][0]],
+            [corrected_box[0][1], corrected_box[1][1], corrected_box[2][1], corrected_box[3][1], corrected_box[0][1]], 'r-'
+        )
+
+        # Add text annotation
+        csfont = {'fontname': 'Poppins'}
+        plt.text(corrected_box[0][0], corrected_box[0][1], text, color='blue', fontsize=8, **csfont)
+
+    plt.axis("off")
+    plt.tight_layout(pad=2.0)
+    plt.savefig(f"uploads/result.png", bbox_inches='tight')
+    plt.close()
+
+    if os.path.exists(image_path):
+        os.remove(image_path)
+
+    return JSONResponse({
+            "message": "Image processed successfully",
+            "image_path": "result.png",
+            "extracted_text": final_text
+    })
+
+@app.get("/get_image")
+def get_image(input_path: str):
+    if not os.path.exists(f"uploads/{input_path}"):
+        return JSONResponse({
+            "message": "Image not found"
+        })
+    return FileResponse(f"uploads/{input_path}")
+
+print("\n🔹 Processing complete! Annotated image and extracted data saved.")

requirements.txt

@@ -0,0 +1,10 @@
+fastapi==0.104.1
+uvicorn==0.24.0
+python-multipart==0.0.6
+paddleocr==2.7.0
+opencv-python-headless==4.8.1.78
+numpy==1.26.2
+matplotlib==3.8.2
+pandas==2.1.3
+Pillow==10.1.0
+paddlepaddle==2.6.2