Upload 3 files
Browse files- Dockerfile +10 -0
- app.py +53 -0
- requirements.txt +8 -0
Dockerfile
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM python:3.10
|
| 2 |
+
|
| 3 |
+
WORKDIR /app
|
| 4 |
+
|
| 5 |
+
COPY requirements.txt .
|
| 6 |
+
RUN pip install --no-cache-dir -r requirements.txt
|
| 7 |
+
|
| 8 |
+
COPY . .
|
| 9 |
+
|
| 10 |
+
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]
|
app.py
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from fastapi import FastAPI, UploadFile, File
|
| 2 |
+
from fastapi.responses import JSONResponse
|
| 3 |
+
from paddleocr import PaddleOCR
|
| 4 |
+
import numpy as np
|
| 5 |
+
from PIL import Image
|
| 6 |
+
from io import BytesIO
|
| 7 |
+
import requests
|
| 8 |
+
|
| 9 |
+
app = FastAPI(title="PaddleOCR API", description="OCR via URL or File Upload")
|
| 10 |
+
|
| 11 |
+
ocr = PaddleOCR(use_angle_cls=True, lang='en')
|
| 12 |
+
|
| 13 |
+
def run_ocr(image_bytes):
|
| 14 |
+
img = Image.open(BytesIO(image_bytes)).convert("RGB")
|
| 15 |
+
result = ocr.ocr(np.array(img), cls=True)
|
| 16 |
+
|
| 17 |
+
parsed_texts = []
|
| 18 |
+
outputs = []
|
| 19 |
+
|
| 20 |
+
for line in result:
|
| 21 |
+
for box, (text, conf) in line:
|
| 22 |
+
parsed_texts.append(text)
|
| 23 |
+
outputs.append({
|
| 24 |
+
"text": text,
|
| 25 |
+
"confidence": float(conf),
|
| 26 |
+
"bounding_box": [[float(x), float(y)] for x, y in box]
|
| 27 |
+
})
|
| 28 |
+
|
| 29 |
+
final_text = " ".join(parsed_texts)
|
| 30 |
+
|
| 31 |
+
return {
|
| 32 |
+
"results": outputs,
|
| 33 |
+
"parsedText": parsed_texts,
|
| 34 |
+
"finalText": final_text
|
| 35 |
+
}
|
| 36 |
+
|
| 37 |
+
@app.get("/ocr")
|
| 38 |
+
async def ocr_url(url: str):
|
| 39 |
+
try:
|
| 40 |
+
img_data = requests.get(url, timeout=10).content
|
| 41 |
+
data = run_ocr(img_data)
|
| 42 |
+
return JSONResponse({"success": True, **data})
|
| 43 |
+
except Exception as e:
|
| 44 |
+
return JSONResponse({"success": False, "error": str(e)}, status_code=400)
|
| 45 |
+
|
| 46 |
+
@app.post("/ocr")
|
| 47 |
+
async def ocr_buffer(file: UploadFile = File(...)):
|
| 48 |
+
try:
|
| 49 |
+
image_bytes = await file.read()
|
| 50 |
+
data = run_ocr(image_bytes)
|
| 51 |
+
return JSONResponse({"success": True, **data})
|
| 52 |
+
except Exception as e:
|
| 53 |
+
return JSONResponse({"success": False, "error": str(e)}, status_code=400)
|
requirements.txt
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
fastapi
|
| 2 |
+
uvicorn
|
| 3 |
+
paddleocr
|
| 4 |
+
paddlepaddle-lite
|
| 5 |
+
numpy
|
| 6 |
+
Pillow
|
| 7 |
+
requests
|
| 8 |
+
python-multipart
|