YOLOv8 Video Analytics System for Security

Imagine a plant with 200 cameras, operators missing 70% of intrusions into dangerous zones. Every shift, people are at risk, and insurance premiums rise. We implemented a video analytics system based on YOLOv8 and ByteTrack — detection at 98% with false alarms below 2%. Over 5 years, we have delivered more than 50 projects at facilities of various scales: from warehouses to oil depots. We guarantee stable operation and post-launch support. The system processes up to 32 streams at 1080p@30fps on a single A100 GPU, and for 100+ cameras, a GPU cluster with load balancing is deployed. All events are timestamped and sent to PSIM or VMS.

How the real-time video analytics system works

IP Cameras (RTSP)
    ↓
Stream Ingestion Layer (GStreamer/FFmpeg)
    ↓
AI Analytics Engine (GPU Cluster)
    ├── Object Detection (YOLOv8)
    ├── Object Tracking (ByteTrack)
    ├── Event Detection (Rules Engine)
    └── Behavior Analysis (ML Models)
    ↓
Event Processing (Kafka/Redis Streams)
    ↓
Alert & Response System
    ├── Security Dashboard
    ├── Mobile Notifications
    ├── Guard Dispatch
    └── Access Control Integration

Each frame is processed in <30 ms. Detection uses YOLOv8n, tracking — ByteTrack with IoU association. To improve accuracy, we apply cascading verification: an object is confirmed over 3 consecutive frames. This reduces false alarms to 2%. This architecture has proven reliable on high-traffic sites — more details in Ultralytics YOLOv8 docs.

Why cascading verification matters

Single-frame detection often gives false alarms due to shadows, animals, or glare. Cascading verification — where the object is confirmed across multiple frames — eliminates 90% of false events. In practice, this means security responds only to real threats. We use three consecutive frames at 0.3-second intervals. This approach is described in ByteTrack: Multi-Object Tracking by Associating Every Detection Box.

Problems we solve

False alarms. Only 5-10% of events are actual security threats. Temporal filtering and cascading zone verification achieve 98% intrusion detection accuracy.

Scaling to 100+ cameras. One NVIDIA A100 handles 32 streams at 1080p@30fps. For larger facilities — a GPU cluster with load balancing. YOLOv8n is 3x faster than predecessors at the same accuracy.

Integration with existing infrastructure. The system connects to PSIM, VMS, access control, and ACS via API. We have integrated with Milestone, Genetec, and Bosch. In one project, we needed to combine 4 different VMS — done in 3 weeks.

How to fine-tune YOLOv8 for a specific site

On a site with unique zones (narrow corridors, specific lighting), the base YOLOv8n gave 85% intrusion accuracy. We collected 5000 labeled frames, fine-tuned the model using LoRA in 2 days. Accuracy rose to 97%. We used PyTorch and Hugging Face Transformers. The result — the system recognizes your specific objects, not generic classes.

What’s included in the work

• Site and threat analysis — identify critical zones and scenarios.
• Architecture design — choose GPUs, cameras, deployment scheme.
• Detection model development — fine-tune YOLOv8 for the site.
• Security zone configuration — polygons, schedules, whitelists.
• Integration with PSIM/VMS/ACS — configure event exchange.
• Staff training — instructions for operators and guards.
• Warranty service — 12 months of technical support.
• Documentation — diagrams, API specs, user guides.

Detectable events

class SecurityZone:
    def __init__(self, name: str, polygon: list, rules: dict):
        self.name = name
        self.polygon = np.array(polygon, dtype=np.int32)
        self.rules = rules

    def is_point_inside(self, point: tuple) -> bool:
        return cv2.pointPolygonTest(self.polygon, (float(point[0]), float(point[1])), False) >= 0

    def is_allowed(self, track_id: int, timestamp: str) -> bool:
        if not self.rules.get('time_restricted'):
            return True
        current_hour = int(timestamp.split(':')[0])
        allowed_hours = self.rules.get('allowed_hours', range(8, 18))
        if current_hour not in allowed_hours:
            return False
        if self.rules.get('whitelist_only'):
            return track_id in self.rules.get('whitelist', set())
        return True

Scaling: processing 100+ cameras

One A100 GPU handles ~32 streams at 1080p@30fps with YOLOv8n. For 100+ cameras — a GPU cluster with load balancing:

class CameraLoadBalancer:
    def __init__(self, gpu_workers: list):
        self.workers = gpu_workers
        self.camera_assignments = {}

    def assign_camera(self, camera_id: str) -> str:
        least_loaded = min(self.workers, key=lambda w: w.load)
        self.camera_assignments[camera_id] = least_loaded
        return least_loaded.worker_id

Integration with physical security

• PSIM — centralized management of all security systems.
• VMS — video storage with forensic search.
• Access Control — automatic door lockdown on alarm.

Our competencies and guarantees

We have certified engineers in YOLOv8, TensorRT, and PyTorch. Experience integrating with PSIM and Access Control — over 50 projects. We guarantee performance: if the system doesn't handle the stated load, we rework it for free. Contact us for a consultation — we'll evaluate your project in 2 business days. Start with a pilot project for 2 cameras to test before full deployment.

How Distribution Shift Kills CV Model Metrics in Industry

On a production line, a camera is installed to control product quality. The model is trained on 10,000 labeled images—test accuracy mAP 0.84. Deployed to production, and in the first week it misses 30% of defects. Lighting on the line changes between shifts; distribution shift nullifies the metrics. This is a classic story with computer vision in industry, where pattern recognition fails without proper drift handling.

Our engineers, with experience from 60+ computer vision projects, know how to eliminate such scenarios. We guarantee stable model performance under real conditions.

Object Detection: YOLO, RT-DETR, and Everything in Between

YOLO is the standard for real-time detection. YOLOv8 and YOLOv11 from Ultralytics are the most used versions in production: simple API, active community, built-in validation, and export to ONNX/TensorRT. For tasks with high accuracy requirements and less critical latency, RT-DETR, a transformer-based architecture without NMS, gives better mAP on COCO at comparable speed to YOLOv8l.

A typical mistake when training a detector: dataset of 8000 images, 3 classes, fine-tune YOLOv8m—F1 0.73 on validation. Look at confusion matrix—one class is almost never detected. Cause: imbalance 1:23. Solution: oversampling rare class, focal loss for objectness, augmentations (Mosaic, MixUp disabled for rare class as they "blur" it). Transfer learning is mandatory: pretrained on COCO weights reduces data requirement by 10 times. Fine-tuning on 500–2000 domain images yields a working model in 1–2 days on a single GPU.

For edge deployment: export to ONNX → TensorRT engine. YOLOv8n in TensorRT FP16 on Jetson AGX Orin gives 150+ FPS at P99 latency < 8 ms—3 times faster than ONNX Runtime without TensorRT. On server A10G: 700+ FPS for YOLOv8n in TensorRT INT8.

How Does Fine-Tuning YOLO Help in Pattern Recognition?

Suppose you need to find micro-defects on a metal surface—a task with high resolution and class imbalance. We use YOLOv8m pretrained on COCO and fine-tune on 2000 proprietary images. Apply augmentations Mosaic, MixUp, random perspective. After 200 epochs, mAP 0.5 reaches 0.93. Key techniques:

• Focal loss for the objectness head—reduces contribution of easily classified examples.
• Class-balanced sampling—equalizes representation of rare classes.
• Test Time Augmentation (TTA)—increases recall by 5–7% through averaging over flips and scales.

Get a consultation on architecture selection for your task—contact us.

Segmentation: SAM, Mask R-CNN, and Instance Segmentation

SAM (Segment Anything Model) from Meta changed the approach to segmentation. SAM 2 works with video, supports object tracking across frames—for interactive object selection by point or bbox, it's the best out-of-the-box choice. For production instance segmentation without interactive prompting, Mask R-CNN or YOLOv8-seg are used. YOLOv8-seg trains like a regular detector with additional masks, convenient in the same pipelines. Semantic segmentation (each pixel is a class) uses SegFormer, DeepLabV3+. SegFormer-B5 provides a good balance of accuracy and speed for satellite imagery or medical segmentation.

Case study: cell segmentation on microscopic images. Dataset of 400 images with manual annotation. Training Mask R-CNN on ResNet-50 backbone gave IoU 0.61—poor. Problem: objects (cells) overlap; standard NMS kills overlapping predictions. Solution: switch to cellpose (specialized architecture for biomedical tasks) + soft-NMS. IoU increased to 0.79.

OCR: When Tesseract Fails

Tesseract is a starting point for simple tasks: printed text, good lighting, straight layout. As soon as there are handwritten elements, non-standard fonts, perspective distortions, or multi-column layouts, Tesseract degrades quickly.

PaddleOCR is a production-grade solution: text block detection + recognition + structural analysis. Works out of the box for 80+ languages, including Russian. Supports tables and complex document structures. TrOCR (Microsoft) is a transformer OCR with strong results on handwritten text. For Russian handwritten text, fine-tuning is needed: the base model is trained mostly on Latin script.

What to Do When Tesseract Cannot Handle Pattern Recognition on Documents?

For tasks like "extract data from invoices/contracts/passports," we use LayoutLMv3 or Donut—these models understand document layout, not just text. Integration via Hugging Face Transformers, fine-tuning on 200–500 annotated documents. Typical pipeline:

1. Preprocessing: deskew, denoising, binarization via OpenCV.
2. Text block detection: PaddleOCR detection or CRAFT.
3. Recognition: PaddleOCR recognition or TrOCR.
4. Post-processing: normalization, validation via regex or LLM for structured fields.

For documents with fixed structure, template matching + OCR by coordinates is often more reliable than an end-to-end solution.

Face Recognition: Identification and Verification

Face recognition = detection + alignment + embedding + matching. Each stage matters.

Detection: RetinaFace or InsightFace for accurate face localization and keypoints. MTCNN is older but reliable. Embedding: ArcFace (InsightFace) is state-of-the-art for face recognition embeddings. Models iresnet50/iresnet100 pretrained on MS1MV3 (5M identities). Embedding vector 512 float32, comparison by cosine similarity. Threshold tuning: decision threshold is a critical parameter. At threshold 0.6, typical FPR on LFW benchmark is 0.001, TPR is 0.985. In production, threshold must be calibrated to the real distribution: people in masks, with changed appearance, different lighting conditions. Liveness detection is mandatory: MiniFASNet—lightweight model on CPU; FaceX-Zoo contains several pretrained liveness detectors.

Video Analytics

Video is a sequence of frames plus a temporal dimension. A naive approach—detecting on every frame—is expensive.

Tracking: ByteTrack and BoT-SORT are the standard for multi-object tracking. They work on top of any detector, adding persistent IDs to objects across frames—enabling object counting, motion tracking, velocity.

Optimization: not every frame needs processing. For static scenes, detect every 5–10 frames, with tracking in between. For event detection (person entering a zone), background subtraction (OpenCV MOG2) serves as a lightweight pre-filter before neural detection. Action recognition: SlowFast, VideoMAE for action classification. Heavy models—for production use ONNX export + TensorRT or offline processing.

How to Measure Pattern Recognition Model Quality in Production?

Quality monitoring is key to MLOps. We track:

• Prediction confidence distribution.
• Share of low-confidence predictions (indicator of OOD data).
• Drift of input images via feature distribution (embeddings from backbone).

A drop in average confidence from 0.87 to 0.71 over a week is an early signal of distribution shift. NVIDIA Triton Inference Server recommends tracking these metrics via Prometheus. Our certified engineers set up monitoring and guarantee SLA for inference quality.

Deployment of CV Models

For online inference, we use Triton Inference Server (NVIDIA)—production standard for serving CV models. Supports TensorRT, ONNX, PyTorch, dynamic batching, multiple instances. REST and gRPC API. We guarantee stable operation under load.

Edge deployment: ONNX Runtime on ARM/x86 CPU. TensorFlow Lite for mobile devices. OpenVINO for Intel CPU/GPU/VPU—gives 2–3× speedup on Intel hardware compared to ONNX Runtime. After deployment, we hand over the model with documentation and train personnel.

What Is Included in the Work

Deadlines and Cost

A prototype detector on existing data takes 1–2 weeks. Production system with optimization for target hardware takes 4–8 weeks. Full cycle including data labeling (1000–5000 images) takes 2–4 months. Cost is calculated individually for each task. Typical savings from implementing a quality control system can be significant per production line.

We have been in the market for over 5 years and completed 60+ computer vision projects. We will evaluate your project end-to-end—request a consultation to get a quote and technical proposal.