About ten years ago, video surveillance was simple and straightforward. Manufacturers, organizations, law enforcement bodies, etc. installed cameras as both a deterrent as well as a means of reactive control.

Advancements in technology have revolutionized the area of video surveillance. Today, industrial smart cameras have Edge Analytics capabilities and can connect to the Cloud for advanced Analytics using company’s existing IP network, over both wired and wireless connections. This expands the range and reduces the cost of new deployments by allowing cameras to be positioned at any location where connectivity is available. A report titled The Physical Security Business 2018 to 2023 sized up the world market for video surveillance products in 2018 at $17.57 billion and stated that it will grow to reach $32.64 billion by 2023.

Video Analytics
Smart Camera Analytics is one of the key technology advancements in the area of camera development and it can be done on the Edge as well as the Cloud as per requirement. Some video analytics use cases in manufacturing industry that can be enabled with next-generation cameras include:

• Motion detection
• Object movement detection
• Facial recognition
• Crowd detection
• Geofence violation detection
• Object or person counting
• Tampering detection
• Keyword search
• Fire recognition

Edge Analytics Cameras and the Industrial Sector
While the video surveillance industry is undergoing a radical change, there is a massive digital transformation effort underway in the manufacturing industry as well. The Industry 4.0 trend has manufacturers turning to sensors to gather data from a wide variety of industrial activities, processing the data on the Edge/Cloud, and enabling multiple use cases such as Remote Monitoring, Predictive Maintenance, Asset tracking, etc. Manufacturers can digitize their operational processes by integrating data from Industrial IoT sensors with data from other systems to achieve a specific digital flow. IoTizing a production line will yield significant results in improving the efficiency and productivity of manufacturing operations. In future, production line supervisors may view IoT deployments as their eyes, ears, and hands across the entire factory floor to monitor efficiency and productivity.

Industrial Smart Cameras will play an important role in Industry 4.0. While they function as real-time video sensors, creating data streams that can be fed directly into an organization’s digital streaming pipeline, they also add another important data source to Industry 4.0 enablement. This will enable manufacturers to actively monitor their factory floor and take corrective action by acting upon the information received. Today, most cameras are equipped with built-in sensors for humidity, audio, temperature, etc. that provide additional data points.

Enhanced video surveillance can enable a multitude of use cases for a manufacturing environment. Some of them include:

• Detecting motion in a restricted area, such as server closet, high-voltage equipment, hazardous materials, etc. with industrial camera monitoring system.
• Monitoring the movement of valuable assets, such as molds and identifying line-crossing events, such as a factory worker reaching an unsafe zone, etc. through construction site video monitoring.
• Regulating the HVAC temperature based on the presence of employees in a particular area.
• Using facial recognition technology to differentiate between employees and partners on the factory floor.
• Read license plates to identify the vehicles entering and leaving the facility.
• Inspect quality of raw materials, work-in progress and finished goods using industrial security camera system.

Edge Analytics Cameras and Smart Building
Smart building interacts with multiple systems, people, and external elements. They use sensors to collect data from different systems (heating, ventilation, air conditioning, lighting, security, etc.) and manage it according to different business functions. It helps owners, operators, and facility managers to improve asset reliability and performance, which reduces energy usage, optimizes space usage, and minimizes the environmental impact of buildings. Industrial Smart Cameras can be the key data source to achieve the above objectives and enable use cases similar to ones enabled in a smart factory setup.

The Speed and Accuracy Trade-offs
The speed and accuracy of the above Industrial and Building Automation use cases is very critical. The response time will be delayed if the video feed is analyzed on the Cloud, but it will be much faster if the video feed is analyzed on the Edge. The hardware processing capacity is also critical. A tradeoff needs to be made between the hardware processing capability and the use cases which can be enabled on the Edge with the required speed and accuracy.

An optimally designed camera should be able to support multiple use case on the Edge while giving the best accuracy and speed at the same time and Cloud Analytics can be leveraged for advanced use cases.

Sasken has extensive experience in Hardware, Embedded and Digital engineering and has leveraged its Machine Vision expertise to develop Industrial Smart Cameras enabled with Edge as well as Cloud Analytics capabilities for use cases in Building Automation, Manufacturing, and Automotive with the required accuracy and speed. Sasken has invested in solution accelerators to enable certain use cases such as Object, Face, and Crowd Recognition in the manufacturing sector. For example, an industrial camera monitoring system developed by Sasken, for one of the largest North American manufacturers of vision systems, is being used for quality defects detection in finished goods.