feat: media API (video/bbox/thumbnail), UUID unification, dot matrix text, portal fixes, API dictionary V1.3
This commit is contained in:
Warren
@@ -1,243 +1,89 @@
|
||||
#!/opt/homebrew/bin/python3.11
|
||||
"""
|
||||
OCR Processor - Text Recognition with Resume Support
|
||||
Uses EasyOCR (local model)
|
||||
|
||||
Resume Feature:
|
||||
- Auto-detect existing results and resume from last frame
|
||||
- Auto-save at configurable intervals (default: 30 seconds)
|
||||
- Ctrl+C gracefully saves and exits
|
||||
OCR Processor Wrapper
|
||||
Calls Swift Vision Framework OCR (swift_ocr) with fallback to PaddleOCR.
|
||||
"""
|
||||
|
||||
import sys
|
||||
import json
|
||||
import argparse
|
||||
import os
|
||||
import time
|
||||
import subprocess
|
||||
import argparse
|
||||
|
||||
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
|
||||
from resume_framework import ResumeFramework, format_time, print_progress
|
||||
|
||||
SWIFT_OCR_PATH = os.path.join(
|
||||
os.path.dirname(os.path.abspath(__file__)),
|
||||
"swift_processors/.build/debug/swift_ocr"
|
||||
)
|
||||
SWIFT_OCR_ALT = os.path.join(
|
||||
os.path.dirname(os.path.abspath(__file__)),
|
||||
"swift_processors/.build/arm64-apple-macosx/debug/swift_ocr"
|
||||
)
|
||||
|
||||
|
||||
def process_ocr(
|
||||
video_path: str,
|
||||
output_path: str,
|
||||
uuid: str = "",
|
||||
auto_save_interval: int = 30,
|
||||
auto_save_frames: int = 300,
|
||||
force_restart: bool = False,
|
||||
sample_interval: int = 30,
|
||||
):
|
||||
"""Process video for OCR using EasyOCR with resume support"""
|
||||
recognition_level: str = "accurate",
|
||||
) -> dict:
|
||||
swift_bin = SWIFT_OCR_PATH
|
||||
if not os.path.exists(swift_bin):
|
||||
swift_bin = SWIFT_OCR_ALT
|
||||
|
||||
framework = ResumeFramework(
|
||||
output_path=output_path,
|
||||
processor_name="ocr",
|
||||
uuid=uuid,
|
||||
auto_save_interval=auto_save_interval,
|
||||
auto_save_frames=auto_save_frames,
|
||||
force_restart=force_restart,
|
||||
if not os.path.exists(swift_bin):
|
||||
print("[OCR] Swift binary not found, using PaddleOCR", file=sys.stderr)
|
||||
return _fallback(video_path, output_path, uuid, sample_interval)
|
||||
|
||||
cmd = [swift_bin, video_path, output_path,
|
||||
"--sample-interval", str(sample_interval),
|
||||
"--recognition-level", recognition_level,
|
||||
"--uuid", uuid]
|
||||
|
||||
print(f"[OCR] Running Swift OCR", file=sys.stderr)
|
||||
result = subprocess.run(cmd, capture_output=True, text=True, timeout=7200)
|
||||
|
||||
if result.stdout:
|
||||
print(result.stdout.strip(), file=sys.stderr)
|
||||
if result.stderr:
|
||||
print(result.stderr.strip(), file=sys.stderr)
|
||||
|
||||
if result.returncode != 0 or not os.path.exists(output_path):
|
||||
print(f"[OCR] Swift OCR failed, falling back to PaddleOCR", file=sys.stderr)
|
||||
return _fallback(video_path, output_path, uuid, sample_interval)
|
||||
|
||||
with open(output_path) as f:
|
||||
return json.load(f)
|
||||
|
||||
|
||||
def _fallback(video_path, output_path, uuid, sample_interval):
|
||||
"""Fallback to original PaddleOCR implementation"""
|
||||
import importlib
|
||||
spec = importlib.util.spec_from_file_location(
|
||||
"paddle_ocr",
|
||||
os.path.join(os.path.dirname(__file__), "ocr_paddle.py")
|
||||
)
|
||||
|
||||
framework.publish_info("OCR_START")
|
||||
|
||||
try:
|
||||
import easyocr
|
||||
except ImportError:
|
||||
framework.publish_error("easyocr not installed")
|
||||
result = {
|
||||
"metadata": {"status": "error", "error": "easyocr not installed"},
|
||||
"frames": {},
|
||||
}
|
||||
with open(output_path, "w") as f:
|
||||
json.dump(result, f, indent=2)
|
||||
framework.publish_progress(0, 0, "0 frames")
|
||||
return result
|
||||
|
||||
framework.publish_info("OCR_LOADING_MODEL")
|
||||
|
||||
reader = easyocr.Reader(["en"], gpu=False, verbose=False)
|
||||
|
||||
framework.publish_info("OCR_MODEL_LOADED")
|
||||
|
||||
import cv2
|
||||
|
||||
cap = cv2.VideoCapture(video_path)
|
||||
|
||||
if not cap.isOpened():
|
||||
print(f"Error: Cannot open video: {video_path}")
|
||||
return {"metadata": {"status": "error"}, "frames": {}}
|
||||
|
||||
fps = cap.get(cv2.CAP_PROP_FPS)
|
||||
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
|
||||
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
|
||||
total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
|
||||
total_duration = total_frames / fps if fps > 0 else 0
|
||||
cap.release()
|
||||
|
||||
framework.publish_info(f"fps={fps}, frames={total_frames}")
|
||||
|
||||
existing_data, last_checkpoint = framework.load_existing_data()
|
||||
resume_mode = existing_data is not None and last_checkpoint > 0 and not force_restart
|
||||
|
||||
if resume_mode:
|
||||
print(f"\nFound existing data: {output_path}")
|
||||
print(f"Last processed frame: {last_checkpoint}")
|
||||
print(f"Will resume from frame {last_checkpoint + 1}")
|
||||
|
||||
if resume_mode and existing_data:
|
||||
ocr_data = existing_data
|
||||
frame_count = last_checkpoint
|
||||
processed_frames = set(int(k) for k in existing_data.get("frames", {}).keys())
|
||||
cap = cv2.VideoCapture(video_path)
|
||||
cap.set(cv2.CAP_PROP_POS_FRAMES, frame_count)
|
||||
else:
|
||||
ocr_data = {
|
||||
"metadata": framework.init_metadata(
|
||||
video_path=video_path,
|
||||
fps=fps,
|
||||
width=width,
|
||||
height=height,
|
||||
total_frames=total_frames,
|
||||
total_duration=total_duration,
|
||||
extra={"sample_interval": sample_interval},
|
||||
),
|
||||
"frames": {},
|
||||
}
|
||||
frame_count = 0
|
||||
processed_frames = set()
|
||||
cap = cv2.VideoCapture(video_path)
|
||||
|
||||
framework.set_data(ocr_data)
|
||||
|
||||
start_time = time.time()
|
||||
framework.last_save_time = start_time
|
||||
|
||||
print(f"\nProcessing video: {total_frames} frames @ {fps:.2f} fps")
|
||||
print(f"Auto-save every {auto_save_interval}s or {auto_save_frames} frames")
|
||||
print(f"Resume from frame {frame_count + 1 if resume_mode else 1}")
|
||||
print()
|
||||
|
||||
while True:
|
||||
ret, frame = cap.read()
|
||||
if not ret:
|
||||
break
|
||||
|
||||
frame_count += 1
|
||||
current_time = (frame_count - 1) / fps if fps > 0 else 0
|
||||
|
||||
if frame_count in processed_frames:
|
||||
continue
|
||||
|
||||
if frame_count % sample_interval != 0:
|
||||
continue
|
||||
|
||||
frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
|
||||
|
||||
try:
|
||||
detections = reader.readtext(
|
||||
frame_rgb, text_threshold=0.5, low_text=0.3, link_threshold=0.3
|
||||
)
|
||||
except Exception as e:
|
||||
framework.publish_error(f"Frame {frame_count}: {e}")
|
||||
detections = []
|
||||
|
||||
texts = []
|
||||
for detection in detections:
|
||||
det: tuple = tuple(detection)
|
||||
bbox = list(det[0])
|
||||
text: str = str(det[1])
|
||||
confidence: float = float(det[2])
|
||||
|
||||
x = int(min(float(p[0]) for p in bbox))
|
||||
y = int(min(float(p[1]) for p in bbox))
|
||||
w = int(max(float(p[0]) for p in bbox) - x)
|
||||
h = int(max(float(p[1]) for p in bbox) - y)
|
||||
|
||||
if text.strip():
|
||||
texts.append(
|
||||
{
|
||||
"text": text,
|
||||
"x": x,
|
||||
"y": y,
|
||||
"width": w,
|
||||
"height": h,
|
||||
"confidence": confidence,
|
||||
}
|
||||
)
|
||||
|
||||
if texts:
|
||||
ocr_data["frames"][str(frame_count)] = {
|
||||
"frame_number": frame_count,
|
||||
"time_seconds": round(current_time, 3),
|
||||
"time_formatted": format_time(current_time),
|
||||
"texts": texts,
|
||||
}
|
||||
processed_frames.add(frame_count)
|
||||
|
||||
if frame_count % 500 == 0:
|
||||
elapsed = time.time() - start_time
|
||||
print_progress(frame_count, total_frames, elapsed, f"{len(texts)} texts")
|
||||
framework.publish_progress(frame_count, total_frames, f"frame {frame_count}")
|
||||
|
||||
if framework.should_auto_save(frame_count):
|
||||
framework.save_progress(frame_count, silent=True)
|
||||
|
||||
cap.release()
|
||||
|
||||
total_processed = len(processed_frames)
|
||||
|
||||
framework.finalize(
|
||||
total_processed=total_processed,
|
||||
extra_metadata={"sample_interval": sample_interval},
|
||||
)
|
||||
|
||||
print(f"\nOCR completed: {total_processed} frames processed")
|
||||
print(f"Frames with text: {len(ocr_data['frames'])}")
|
||||
|
||||
return ocr_data
|
||||
if spec is None:
|
||||
print("[OCR] No fallback available, returning empty result", file=sys.stderr)
|
||||
return {"frame_count": 0, "fps": 0, "frames": []}
|
||||
paddle = importlib.util.module_from_spec(spec)
|
||||
spec.loader.exec_module(paddle)
|
||||
return paddle.process_ocr(video_path, output_path, uuid, sample_interval=sample_interval)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
parser = argparse.ArgumentParser(description="OCR Text Recognition with Resume Support")
|
||||
parser.add_argument("video_path", help="Path to video file")
|
||||
parser.add_argument("output_path", help="Output JSON path")
|
||||
parser.add_argument("--uuid", "-u", help="UUID for Redis progress", default="")
|
||||
parser.add_argument(
|
||||
"--auto-save-interval",
|
||||
"-a",
|
||||
help="Auto-save interval in seconds",
|
||||
type=int,
|
||||
default=30,
|
||||
)
|
||||
parser.add_argument(
|
||||
"--auto-save-frames",
|
||||
"-f",
|
||||
help="Auto-save interval in frames",
|
||||
type=int,
|
||||
default=300,
|
||||
)
|
||||
parser.add_argument(
|
||||
"--force-restart",
|
||||
"-r",
|
||||
help="Force restart (ignore existing data)",
|
||||
action="store_true",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--sample-interval",
|
||||
"-s",
|
||||
help="Frame sample interval",
|
||||
type=int,
|
||||
default=30,
|
||||
)
|
||||
parser = argparse.ArgumentParser(description="OCR Processor (Swift Vision)")
|
||||
parser.add_argument("video_path")
|
||||
parser.add_argument("output_path")
|
||||
parser.add_argument("--uuid", "-u", default="")
|
||||
parser.add_argument("--sample-interval", type=int, default=30)
|
||||
parser.add_argument("--recognition-level", choices=["fast", "accurate"], default="accurate")
|
||||
args = parser.parse_args()
|
||||
|
||||
process_ocr(
|
||||
args.video_path,
|
||||
args.output_path,
|
||||
args.uuid,
|
||||
args.auto_save_interval,
|
||||
args.auto_save_frames,
|
||||
args.force_restart,
|
||||
args.sample_interval,
|
||||
)
|
||||
result = process_ocr(args.video_path, args.output_path, args.uuid,
|
||||
args.sample_interval, args.recognition_level)
|
||||
|
||||
with open(args.output_path, "w") as f:
|
||||
json.dump(result, f, indent=2)
|
||||
print(f"OCR: {len(result.get('frames', []))} frames with text")
|
||||
|
||||
Reference in New Issue
Block a user