OCR API — 快速擷取圖片中的文字

# pip install httpx
import httpx

resp = httpx.post(
    "https://api.parsejet.com/v1/parse/image/ocr",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    files={"file": open("receipt.jpg", "rb")},
    data={"language": "en"},     # optional hint: ko, ru, ar, hi, th, el ...
    timeout=60,
)
resp.raise_for_status()
result = resp.json()
print(result["text"])
if result["metadata"]["ocr_confidence"] < 0.8:
    print("low confidence — send for manual review")

# Batch a folder of scans
from pathlib import Path
for img in Path("scans").glob("*.png"):
    r = httpx.post(
        "https://api.parsejet.com/v1/parse/image/ocr",
        headers={"Authorization": "Bearer YOUR_API_KEY"},
        files={"file": (img.name, img.read_bytes(), "image/png")},
    )
    print(img.name, r.json()["text"][:80])

// npm install parsejet
import { ParseJet } from "parsejet";
import { readFile } from "node:fs/promises";

const client = new ParseJet({ apiKey: process.env.PARSEJET_API_KEY });
const image = await readFile("receipt.jpg");
const result = await client.parse.ocr(image, "receipt.jpg");
console.log(result.text);

// Or plain fetch, in any runtime:
const form = new FormData();
form.append("file", new Blob([image]), "receipt.jpg");
const res = await fetch("https://api.parsejet.com/v1/parse/image/ocr", {
  method: "POST",
  headers: { Authorization: `Bearer ${process.env.PARSEJET_API_KEY}` },
  body: form,
});
const { text, metadata } = await res.json();

# No API key: 3 requests/day for testing
curl -X POST https://api.parsejet.com/v1/parse/image/ocr \
  -F "[email protected]"

# With a key (300 free requests/month), Korean text
curl -X POST https://api.parsejet.com/v1/parse/image/ocr \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" -F "language=ko"

# Response
# {
#   "text": "...",
#   "title": "menu",
#   "source_type": "image_ocr",
#   "metadata": { "width": 1280, "height": 960, "ocr_lines": 14, "ocr_confidence": 0.96 }
# }