File size: 8,406 Bytes
255e6fd |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 |
#!/usr/bin/env python3
"""
OCR Layout Detection API Client
================================
Simple script to interact with the OCR Layout Detection service.
Usage:
python api_client.py <path_to_file>
Examples:
python api_client.py invoice.pdf
python api_client.py document.jpg
python api_client.py signature.png --signature-only
"""
import os
import sys
import json
import argparse
from pathlib import Path
from gradio_client import Client, handle_file
# API Configuration
SPACE_URL = "Ayaan-Sharif/ocr-layout-detection-poc"
HF_TOKEN = os.environ.get("HF_TOKEN") # Read from environment variable if available
def analyze_document(file_path, mode="Fast", enable_ocr=True, enable_tables=True,
detect_signatures=False, signature_conf=0.05):
"""
Analyze a document with layout detection and optional OCR.
Args:
file_path: Path to PDF or image file
mode: "Fast" or "Accurate" processing mode
enable_ocr: Extract text with OCR
enable_tables: Detect and extract tables
detect_signatures: Also detect signatures (slower)
signature_conf: Confidence threshold for signatures (0.01-0.5)
Returns:
dict: Contains visualization, summary, markdown, and JSON outputs
"""
print(f"📄 Analyzing document: {file_path}")
print(f" Mode: {mode} | OCR: {enable_ocr} | Tables: {enable_tables} | Signatures: {detect_signatures}")
try:
client = Client(SPACE_URL, hf_token=HF_TOKEN)
result = client.predict(
file=handle_file(file_path),
mode=mode,
enable_ocr=enable_ocr,
enable_tables=enable_tables,
run_signature_yolo=detect_signatures,
signature_conf=signature_conf,
api_name="/gradio_interface"
)
# result is a tuple: (visualization_image, summary_text, markdown_text, json_text)
visualization, summary, markdown, json_output = result
print("✅ Analysis complete!")
return {
"visualization": visualization,
"summary": summary,
"markdown": markdown,
"json": json_output
}
except Exception as e:
print(f"❌ Error: {e}")
return None
def detect_signatures_only(file_path, multiscale=True, conf=0.03, iou=0.45, augment=True):
"""
Detect signatures only (faster, no OCR or layout analysis).
Args:
file_path: Path to PDF or image file
multiscale: Try multiple scales (1.0, 1.5, 2.0) for better detection
conf: Confidence threshold (0.01-0.5, lower = more detections)
iou: IoU threshold for NMS (0.1-0.9)
augment: Use augmentation (slower but better recall)
Returns:
dict: Contains annotated image, summary, and JSON detections
"""
print(f"✍️ Detecting signatures in: {file_path}")
print(f" Multiscale: {multiscale} | Conf: {conf} | IoU: {iou} | Augment: {augment}")
try:
client = Client(SPACE_URL, hf_token=HF_TOKEN)
result = client.predict(
file=handle_file(file_path),
try_scales=multiscale,
conf=conf,
iou=iou,
augment=augment,
api_name="/signature_only_infer"
)
# result is a tuple: (annotated_image, summary_text, json_detections)
annotated_image, summary, json_output = result
print("✅ Signature detection complete!")
return {
"annotated_image": annotated_image,
"summary": summary,
"json": json_output
}
except Exception as e:
print(f"❌ Error: {e}")
return None
def save_results(results, output_dir="output"):
"""Save API results to files."""
os.makedirs(output_dir, exist_ok=True)
if results is None:
return
# Save visualization/annotated image
if "visualization" in results and results["visualization"]:
viz_path = results["visualization"].get("path")
if viz_path and os.path.exists(viz_path):
import shutil
output_path = os.path.join(output_dir, "visualization.png")
shutil.copy(viz_path, output_path)
print(f"💾 Saved visualization: {output_path}")
if "annotated_image" in results and results["annotated_image"]:
img_path = results["annotated_image"].get("path")
if img_path and os.path.exists(img_path):
import shutil
output_path = os.path.join(output_dir, "signatures_annotated.png")
shutil.copy(img_path, output_path)
print(f"💾 Saved annotated image: {output_path}")
# Save markdown content
if "markdown" in results and results["markdown"]:
markdown_path = os.path.join(output_dir, "content.md")
with open(markdown_path, "w", encoding="utf-8") as f:
f.write(results["markdown"])
print(f"💾 Saved markdown: {markdown_path}")
# Save JSON output
if "json" in results and results["json"]:
json_path = os.path.join(output_dir, "layout.json")
with open(json_path, "w", encoding="utf-8") as f:
f.write(results["json"])
print(f"💾 Saved JSON: {json_path}")
# Save summary
if "summary" in results and results["summary"]:
summary_path = os.path.join(output_dir, "summary.txt")
with open(summary_path, "w", encoding="utf-8") as f:
f.write(results["summary"])
print(f"💾 Saved summary: {summary_path}")
def main():
parser = argparse.ArgumentParser(
description="OCR Layout Detection API Client",
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog="""
Examples:
# Full document analysis with OCR
python api_client.py invoice.pdf
# Accurate mode with signature detection
python api_client.py document.pdf --mode Accurate --detect-signatures
# Signature detection only (faster)
python api_client.py contract.jpg --signature-only
# Custom output directory
python api_client.py file.pdf --output results/
"""
)
parser.add_argument("file", help="Path to document (PDF, JPG, PNG)")
parser.add_argument("--mode", choices=["Fast", "Accurate"], default="Fast",
help="Processing mode (default: Fast)")
parser.add_argument("--no-ocr", action="store_true", help="Disable OCR")
parser.add_argument("--no-tables", action="store_true", help="Disable table detection")
parser.add_argument("--detect-signatures", action="store_true",
help="Also detect signatures in full analysis")
parser.add_argument("--signature-conf", type=float, default=0.05,
help="Signature confidence threshold (default: 0.05)")
parser.add_argument("--signature-only", action="store_true",
help="Only detect signatures (faster, no OCR)")
parser.add_argument("--output", "-o", default="output",
help="Output directory (default: output)")
args = parser.parse_args()
# Validate file exists
if not os.path.exists(args.file):
print(f"❌ Error: File not found: {args.file}")
sys.exit(1)
# Check file type
ext = Path(args.file).suffix.lower()
if ext not in [".pdf", ".jpg", ".jpeg", ".png", ".tiff", ".bmp"]:
print(f"⚠️ Warning: Unsupported file type: {ext}")
print(" Supported: .pdf, .jpg, .jpeg, .png, .tiff, .bmp")
print(f"\n🚀 Starting API call to {SPACE_URL}\n")
# Call appropriate API endpoint
if args.signature_only:
results = detect_signatures_only(args.file)
else:
results = analyze_document(
args.file,
mode=args.mode,
enable_ocr=not args.no_ocr,
enable_tables=not args.no_tables,
detect_signatures=args.detect_signatures,
signature_conf=args.signature_conf
)
# Save results
if results:
print(f"\n📁 Saving results to: {args.output}/")
save_results(results, args.output)
print("\n✨ Done!")
else:
print("\n❌ Failed to process document")
sys.exit(1)
if __name__ == "__main__":
main()
|