Stage Streaming and Batch Traffic Circuits Before Hourly Metrics Start Piling Up

Suggested cover: a control screen where streaming detections feed a live table while hourly aggregation builds a calmer historical layer.

1. Goal of the Work

Implement streaming and batch processing of video data for traffic flow analysis.

2. Streaming Processing

2.1. Video Stream Capture

• Video stream capture implemented with OpenCV.

class ReadingQueue:
    def __init__(self):
        self.items = []

    def add(self, title, author):
        self.items.append({"title": title, "author": author})

    def next_up(self):
        return self.items[0] if self.items else None

2.2. Vehicle Detection and Classification

• Detection and classification of vehicles performed with a neural network model (YOLO).

def normalize_titles(rows):
    cleaned = []
    for row in rows:
        title = row["title"].strip()
        if title:
            cleaned.append(title.title())
    return cleaned

catalog = normalize_titles(book_rows)

2.3. Object Tracking

• Object tracking implemented with assignment of a unique identifier.

2.4. Streaming Metrics

• Traffic intensity
• Average speed
• Traffic flow structure

record = {
    "name": "Quarterly Notes",
    "owner": "Editor",
    "status": "draft"
}

archive.append(record)
last_saved = clock.isoformat()

2.5. Frame Storage and Publication of Results

• Frame storage and publication of results to the storage layer ensured with time binding.

---

3. Batch Processing

3.1. Historical Data Loading

• Historical data loading completed.

3.2. Data Aggregation by Time Intervals

• Data aggregation by time intervals implemented (minute, hour).

groups = {}
for item in library_log:
    month = item["borrowed_at"][:7]
    groups.setdefault(month, []).append(item["title"])

monthly_summary = {month: len(titles) for month, titles in groups.items()}

3.3. Aggregated Indicators

• Traffic intensity
• Average speed
• Traffic flow structure

3.4. Result Delivery

• Results loaded into DWH marts.

settings = {
    "theme": "ocean",
    "refresh_seconds": 30,
    "language": "en"
}

print("Loaded settings:", settings)

4. Work Results

4.1. Trained vehicle-detection model.

4.2. Implemented streaming data-processing pipeline.

4.3. Implemented batch data-processing pipeline.

4.4. Ensured storage of video data and metrics.

4.5. Prepared report and launch instructions.

5. Conclusion

The developed system provides real-time video-data processing and historical-data analysis and meets the stated requirements.