
Industry
Textile Manufacturing
Duration
3 months
Team Size
5 people
Overview
Salona Cotspin operates a spinning and knitting facility in Coimbatore, Tamil Nadu, producing cotton yarn and knitted fabrics. To strengthen security and safety monitoring across the facility, Katomaran Technologies deployed AI video analytics across existing camera infrastructure, covering mobile phone usage, unauthorized entry, crowd density, perimeter intrusion, fire and smoke, and vehicle identification. The implementation and deployment were delivered as a complimentary, discounted engagement.
| Parameter | Details |
|---|---|
| Client | Salona Cotspin |
| Industry | Textile Manufacturing (Yarn & Knitted Fabrics) |
| Application | Factory Security & Safety Monitoring |
| Core Technology | AI Video Analytics + VMS |
| Key Use Cases | Mobile Usage, Unauthorized Entry, Crowd, Perimeter Intrusion, Fire & Smoke, ANPR |
| Cameras by Use Case | Mobile Usage, Unauthorized Entry, Crowd, Perimeter, Fire & Smoke, ANPR |
| Deployment Model | Existing camera infrastructure |
Textile Factory Security Challenges
Textile manufacturing facilities like spinning and knitting units typically run continuous production across large floor areas, restricted zones, and boundary perimeters, all of which need consistent monitoring. Relying on manual observation alone across these areas makes it difficult to maintain consistent discipline and respond quickly to safety-related events, which is a common pattern across manufacturing environments Katomaran has worked with.
Typical challenges in this type of facility include:
- Mobile phone usage on the production floor or in restricted zones going unnoticed during regular operations
- Restricted or sensitive areas being accessed without staff being immediately aware
- Unusual gatherings or crowding in specific zones going unmonitored until they become a concern
- Boundary and perimeter areas being vulnerable to intrusion, particularly outside regular working hours
- Fire and smoke incidents, relevant given the presence of cotton and fabric material, requiring early detection to limit escalation
- Vehicles entering and exiting the facility without a consistent way to identify or log them
AI Video Analytics Solution
- Katomaran's approach for Salona Cotspin centered on applying AI video analytics to the facility's existing camera network, converting selected cameras into dedicated monitoring points for specific safety and security use cases. Each use case was mapped to a defined number of cameras based on the areas that needed coverage, rather than applying a single generic layer across the entire facility.
- The result is a set of purpose-built analytics modules — for mobile usage, unauthorized entry, crowd density, perimeter intrusion, fire and smoke, and vehicle identification — working together and feeding into centralized monitoring for the security team.
Mobile Usage Detection

Mobile usage and crowd detection
- Deployed across 13 cameras covering production floor and restricted zones
- Identifies mobile phone usage in areas where it is not permitted during operations
- Supports consistent monitoring of floor discipline without requiring constant manual supervision
- Flags detected usage for the security or operations team to review
Crowd Detection
- Deployed across 5 cameras to monitor gatherings or unusual density in specific zones
- Flags situations where crowd density exceeds expected levels for a given area
- Supports awareness of unusual gatherings that may need attention
- Adds visibility into areas where crowd movement is relevant to safety or operations
Perimeter Intrusion Detection

AI perimeter intrusion detection
- Deployed across 5 cameras covering boundary and perimeter areas
- Detects intrusion attempts along the facility's boundary
- Supports round-the-clock perimeter awareness, including outside working hours
- Strengthens overall boundary security alongside existing physical measures
Fire and Smoke Detection
- Deployed across few cameras positioned for early fire and smoke detection
- Identifies visual indicators of fire or smoke as early as possible
- Supports faster awareness of safety-related incidents, given the material handled at the facility
- Adds a video-based layer of monitoring alongside other fire safety measures already in place
ANPR – Automatic Number Plate Recognition

ANPR vehicle entry monitoring
- Deployed on a camera covering vehicle entry and exit
- Identifies vehicle number plates as vehicles move through the monitored point
- Supports number plate recognition for consistent vehicle identification at the facility
- Adds a vehicle-level record to overall entry and exit monitoring
How AI Video Analytics Works
Once cameras are mapped to their relevant use case, Katomaran's AI video analytics platform takes over the ongoing monitoring across the facility. Each feed is processed against its assigned detection model, and any relevant event triggers an alert without requiring someone to be watching that feed at the time.
- Cameras across the facility are mapped to the specific use case relevant to their location
- The AI video analytics platform processes each feed using the detection model assigned to it mobile usage, unauthorized entry, crowd detection, perimeter intrusion, fire and smoke, or ANPR
- When a use case-specific condition is detected, such as an unauthorized entry or a fire and smoke event, the system generates an alert
- Alerts from all six use cases are handled through the same analytics deployment rather than through separate standalone systems
- Detected events and alerts can be reviewed through a centralized video management system interface, keeping monitoring in one place
- Security teams review detected events and respond according to the type of alert generated
- The existing camera infrastructure supports all six use cases concurrently, without requiring separate dedicated hardware per use case
Key Video Analytics Capabilities
Across the deployment, Katomaran's video analytics platform brings together six distinct use cases relevant to the facility's security and safety needs. Each capability runs on the existing camera network and feeds into the same centralized monitoring setup, giving the team one consistent view instead of separate tools for each function.
- 1Mobile phone usage detection across production floor and restricted zones
- 2Unauthorized entry detection at critical access points
- 3Crowd detection for early awareness of unusual gatherings
- 4Perimeter intrusion detection along the facility boundary
- 5Fire and smoke detection for early safety alerts
- 6ANPR-based vehicle identification at the entry/exit point
- 7Centralized monitoring through an integrated video management system
Outcome
- With AI video analytics layered onto its existing camera infrastructure, Salona Cotspin gained more consistent visibility across its production floor, restricted areas, perimeter, and vehicle entry point. Mobile usage and unauthorized entry are easier to identify as they happen, crowd formations are flagged before they grow into larger concerns, and fire or smoke indicators are picked up automatically rather than through manual observation alone. Centralized monitoring through the VMS gives the security team one point to verify alerts and respond, supporting better day-to-day discipline and safety oversight.
- The approach reflects the same analytics-driven thinking behind Katomaran's work on factory safety and workforce compliance and its broader manufacturing security solutions.
Table of contents
Looking to Strengthen Factory Security with AI Video Analytics?
Katomaran Technologies provides AI video analytics, VMS, ANPR, and automated safety monitoring solutions tailored for manufacturing plants, textile mills, and industrial facilities.
Talk to Our Experts