File size: 8,465 Bytes
a5c15eb | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 | import os
import re
import base64
import logging
import traceback
os.environ['FLAGS_use_onednn'] = '0'
os.environ['PADDLE_PDX_DISABLE_MODEL_SOURCE_CHECK'] = 'True'
import sys
from flask import Flask, request, jsonify
from flask_cors import CORS
import cv2
import numpy as np
from Src.Preprocessing.enhancedpicture import EnhancedPicture
from Src.Model.SVTRmodel import SVTRModel
logging.basicConfig(level=logging.ERROR)
logger = logging.getLogger(__name__)
app = Flask(__name__)
CORS(app, origins="*")
RAW_FOLDER = 'Data/Raw'
PROCESSED_FOLDER = 'Data/Processed'
os.makedirs(RAW_FOLDER, exist_ok=True)
os.makedirs(PROCESSED_FOLDER, exist_ok=True)
BASE_DIR = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
REC_MODEL_DIR = os.path.join(BASE_DIR, 'Src', 'Model', 'Inference')
ai_engine = SVTRModel(rec_model_dir=REC_MODEL_DIR, debug=False)
preprocessor = EnhancedPicture(debug=False)
print(f" Model aktif: {ai_engine.model_source}")
def is_valid_text(text: str, confidence: float) -> bool:
stripped = text.strip()
if len(stripped) <= 2: return False
if confidence < 0.50: return False
if re.match(r'^[^a-zA-Z0-9]+$', stripped): return False
clean = re.sub(r'[^a-zA-Z0-9\s/:\-\.]', '', stripped)
if len(clean) < len(stripped) * 0.5: return False
return True
_DATE_PATTERN = re.compile(
r'\b(\d{1,2}[\s\-/\.]?'
r'(?:JAN|FEB|MAR|APR|MAY|JUN|JUL|AUG|SEP|OCT|NOV|DEC|\d{1,2})'
r'[\s\-/\.]?\d{2,4})\b',
re.IGNORECASE
)
_KODE_PATTERN = re.compile(
r'\b([A-Z]\d+[A-Z]?\s+\d{2}:\d{2}|LOT\s*[A-Z0-9]+|[A-Z]{1,3}\d{3,}[A-Z]?)\b',
re.IGNORECASE
)
def extract_key_info(results: list):
exp_date = kode_produksi = None
for item in results:
text = item['text']
if exp_date is None:
m = _DATE_PATTERN.search(text)
if m:
exp_date = {
"raw_text" : text,
"value" : m.group(0),
"confidence": round(float(item['confidence']), 4),
"status" : item['status'],
}
if kode_produksi is None:
m = _KODE_PATTERN.search(text)
if m:
kode_produksi = {
"raw_text" : text,
"value" : m.group(0),
"confidence": round(float(item['confidence']), 4),
"status" : item['status'],
}
if exp_date and kode_produksi:
break
return exp_date, kode_produksi
def build_response(filtered: list, scan_status: str):
"""
Response JSON flat — langsung di-parse Flutter.
{
"scan_status" : "OK" | "RE_CAPTURE",
"exp_date" : { value, confidence, status, raw_text } | null,
"kode_produksi" : { value, confidence, status, raw_text } | null,
"confidence_score": float 0–1,
"confidence_pct" : int 0–100,
"total_detected" : int,
"all_text" : "baris1\nbaris2\n...",
"detections" : [ { text, confidence, status }, ... ]
}
"""
exp_date, kode_produksi = extract_key_info(filtered)
best = max(filtered, key=lambda x: x['confidence'])
return {
"scan_status" : scan_status,
"exp_date" : exp_date,
"kode_produksi" : kode_produksi,
"confidence_score": round(float(best['confidence']), 4),
"confidence_pct" : round(float(best['confidence']) * 100),
"total_detected" : len(filtered),
"all_text" : '\n'.join(r['text'] for r in filtered),
"detections" : [
{"text": r['text'], "confidence": round(float(r['confidence']), 4), "status": r['status']}
for r in filtered
],
}
def _empty_response(error_msg: str):
return {
"scan_status" : "RE_CAPTURE",
"exp_date" : None,
"kode_produksi" : None,
"confidence_score": 0.0,
"confidence_pct" : 0,
"total_detected" : 0,
"all_text" : "",
"detections" : [],
"error" : error_msg,
}
@app.route('/', methods=['GET'])
def home():
return jsonify({
"message" : "SVTR OCR API — Server aktif",
"version" : "2.1.0",
"model_source": ai_engine.model_source,
"endpoints" : {"health": "GET /health", "scan": "POST /api/scan"},
})
@app.route('/health', methods=['GET'])
@app.route('/cek-koneksi', methods=['GET'])
def health():
return jsonify({
"status" : "OK",
"model_source": ai_engine.model_source,
"message" : "SVTR OCR API aktif dan siap menerima gambar",
}), 200
@app.route('/api/scan', methods=['POST'])
def api_scan():
data = request.get_json(silent=True)
if not data:
return jsonify({"error": "Request body harus JSON"}), 400
if not data.get('image'):
return jsonify({"error": "Field 'image' tidak ada atau kosong"}), 400
try:
img_bytes = base64.b64decode(data['image'])
nparr = np.frombuffer(img_bytes, np.uint8)
image = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
if image is None:
return jsonify({"error": "Gambar tidak valid"}), 400
except Exception as e:
return jsonify({"error": f"Gagal decode gambar: {str(e)}"}), 400
tmp_path = os.path.join(RAW_FOLDER, 'tmp_scan.jpg')
try:
cv2.imwrite(tmp_path, image)
except Exception as e:
return jsonify({"error": f"Gagal menyimpan sementara: {str(e)}"}), 500
try:
results = ai_engine.predict_from_file(tmp_path)
if not results:
results = ai_engine.predict(image)
if not results:
return jsonify(_empty_response(
"Teks tidak terdeteksi. Coba foto lebih dekat."
)), 200
filtered = [r for r in results if is_valid_text(r['text'], r['confidence'])]
if not filtered:
return jsonify(_empty_response("Teks tidak jelas. Silakan foto ulang.")), 200
best = max(filtered, key=lambda x: x['confidence'])
scan_status = "OK" if best['confidence'] >= 0.85 else "RE_CAPTURE"
return jsonify(build_response(filtered, scan_status)), 200
except Exception:
logger.error(traceback.format_exc())
return jsonify({"error": "Kesalahan internal server. Cek log terminal."}), 500
@app.route('/predict', methods=['POST'])
def predict():
if 'image' not in request.files:
return jsonify({"status": "error", "message": "File tidak ditemukan"}), 400
file = request.files['image']
if not file.filename:
return jsonify({"status": "error", "message": "Nama file kosong"}), 400
ext = file.filename.rsplit('.', 1)[-1].lower()
if ext not in {'jpg', 'jpeg', 'png'}:
return jsonify({"status": "error", "message": f"Format tidak didukung: .{ext}"}), 400
raw_path = os.path.join(RAW_FOLDER, file.filename)
file.save(raw_path)
try:
results = ai_engine.predict_from_file(raw_path) or ai_engine.predict(cv2.imread(raw_path))
if not results:
return jsonify({"status": "RE_CAPTURE", "message": "Teks tidak terdeteksi", "data": None}), 200
filtered = [r for r in results if is_valid_text(r['text'], r['confidence'])]
if not filtered:
return jsonify({"status": "RE_CAPTURE", "message": "Teks tidak jelas", "data": None}), 200
best = max(filtered, key=lambda x: x['confidence'])
scan_status = "OK" if best['confidence'] >= 0.85 else "RE_CAPTURE"
return jsonify({"status": scan_status, "message": "Proses selesai",
"data": build_response(filtered, scan_status)}), 200
except Exception:
logger.error(traceback.format_exc())
return jsonify({"status": "error", "message": "Kesalahan server"}), 500
if __name__ == '__main__':
port = int(os.environ.get('PORT', 5000))
print("=" * 55)
print(" SVTR OCR Backend v2.1")
print(f" Model: {ai_engine.model_source}")
print("=" * 55)
app.run(host='0.0.0.0', port=port, debug=False) |