Given a reservoir located in a mountainous area with an automated water level monitoring system. Human observers would notice if the water temperature rises and when, if a forest fire has broken out nearby, or if catastrophic events such as an earthquake could damage the dam structure. An IoT system that is only designed to perform an automated function - in this case water level measurement - does not do this per se.

In order to be able to dispense with a human operator for reservoir operation, the IoT system also requires various networked sensors that monitor factors such as temperature, humidity and contamination. Each sensor would have to be fitted individually - a challenge for the designer. Multifunctional environmental sensors such as the '2JCIE-BL01' from Omron make this task easier. The fully integrated sensor incorporates various additional environmental factors: UV index (UVI), air pressure, noise level as well as the acceleration of an object (triggered by movement, gravity or even vibrations). With the help of a combined environmental sensor, the user can track the values in the environment and set threshold values that alert the user to abnormal sensor results. To stay with the water reservoir example: In response to a detected temperature rise, an alarm and/or automated response could be triggered, for example for an additional water feed to compensate for increased evaporation.

Seismic sensors

There is a constant risk of earthquakes in many regions of the world. An average of 50 quakes a day add up to around 20,000 events per year. Many are weak enough to go unnoticed. Others, however, lead to disasters. This is a serious problem in unattended IoT systems: a supervised system can be shut down immediately by its human operator in an emergency. A decentrally controlled system, on the other hand, cannot: it is vulnerable as long as the problem is not detected. Here too, sensor technology provides support: an intelligent seismic sensor such as the 'D7S' from Omron, for example, minimizes the risk of secondary damage after earthquakes by stopping and shutting down potentially dangerous devices and production lines. The sensor also helps with damage assessment by recording the seismic intensity and thus providing information on destroyed buildings.

The sensor uses a globally patented algorithm technology to calculate the spectral intensity (SI). For example, if the factory to be monitored is located near railroad tracks, the sensor uses its triaxial accelerometer and the algorithms to calculate the SI value to distinguish between real seismic activity caused by an earthquake and other sources of origin.

Dust sensors

As a result of increased vigilance with regard to vehicle emissions, industrial smoke and cigarette smoke, air pollution in the home and workplace is also increasingly coming to the attention of the public. The increase in particles in the atmosphere means a higher risk of exposure. Using IoT technology, a dust sensor can, for example, automatically send a warning message to activate the air conditioning system to take appropriate countermeasures. An Omron solution detects 0.5 micron tiny particles - six times the air flow rate of conventional alternatives - making this dust sensor much more sensitive to environmental changes.

Sensors for facial recognition

The 'HVC-P2 B5T' from Omron offers ten essential image capture modes. Embedded developers can integrate face recognition functions into their systems or require the optical setup.

© Omron Electronic Components

IoT technology not only automates 'unmanned systems'. Thanks to IoT, modern sensors are also able to recognize people in a wide variety of environments. For example, a facial recognition sensor can be used to distinguish people from animals or to check whether a person is legitimately present in a particular building. In the latter case, security applications are involved, ranging from access to buildings, computers and telephones to restricting access or even denying access to certain people with different clearance values. Face and various gesture recognition can be easily integrated into a design without the need for knowledge of the underlying algorithms or the optical structure. The 'HVC-P2 B5T' from Omron, for example, offers ten essential image capture functions: Detection of a person's face, hand, body, facial recognition, gender recognition, age estimation, state of mind estimation, head posture, gaze direction and blink frequency. In each case, the module outputs a value along with a degree of probability guess, allowing the programmer to configure the response correctly for each individual application.

Cloud-based solutions

IoT can be used to measure and monitor the direct and indirect environmental influences on a factory using cloud-based solutions, set thresholds and automated response processes. As far as IoT customer service systems are concerned, the degree of user customization can be increased through connectivity. For example, a scenario is conceivable in which the seat of a machine is automatically adjusted to the size and stature of the user after facial recognition has identified the user. Or future integrated room systems will react to gestures alone in order to adjust the brightness and light color of the lighting as well as the air conditioning and heating as desired.

Author:
Gabriele Fulco is Manager Product Marketing Europe at Omron Electronics Components Europe in Hoofddorp, Netherlands.