News

OMRON launches image-sensing unit for IoT

August 30, 2016

OMRON Corp. of Kyoto City, Japan, a provider of automation based on its core sensing and control technology, has debuted its HVC-P2 B5T-007001 Series, a built-in human condition recognition unit dubbed a "Human Vision Components" system. It has a maximum recognition speed 10 times greater than that of previous models, according to a company press release.

The HVC is an image-sensing unit incorporating proprietary "OKAO (R) Vision" technology for recognizing human face expression, gender, age, gaze and blink into a camera module. The HVC-P2 has a maximum recognition speed 10 times that of a previous model introduced in March 2014, making it possible to detect a human body four times per second, keeping track of a person within a detection area, the release said.

Customers can also choose from two camera heads: A long-distance detection type and a wide-angle detection type, depending on their specific application purposes. A piece of equipment embedded with the HVC-P2 can detect and presume attributes and conditions of a user coming in its vicinity, without the user knowing the presence of a camera, according to the release.

Examples of application:

• digitalizing people's attention to advertisements, including digital signage;
• the long-distance version of the HVC-P2 can detect and presume attributes of people, including gender and age, as well as their sight line and facial expression from a maximum distance of 3 meters, measuring, for example, the degree of attention people walking through a railway station pay station to a digital signage system installed there;
• contributing to optimization and development of products for sale from vending machines; and
• a wide-angle version of the machine is capable of covering an area 100 cm by 75 cm from a distance of 50 cm, keeping track of people buying naturally from a vending machine, collecting data to be used later for product refill, new product development and marketing activities.

Additionally, the HVC-P2 can be embedded in a variety of equipment and machines to do various jobs involving humans and machines, including safeguarding people in manufacturing workplaces, keeping track of congestion in elevators and monitoring people receiving care at nursing homes, the release said.

Main features:

• Ten types of image-sensing functions are available for recognizing human conditions in various perspectives: face detection, human body detection, hand detection, face direction estimation, gaze estimation, blink estimation, age estimation, gender estimation, expression estimation, (five facial expressions: neutral, happiness, surprise, anger and sadness) and face recognition. Also, recognition and presumption come in the form of digital data, including the number of detections, angles and age as well as text data including facial expressions and gender.
• The HVC-P2 consists of a camera and a separate main board connected via a flexible flat cable, allowing it to be installed on the edge of a flat display unit.
• Output image can be chosen from three types: No image output, 160 x 120 pixels and 320 x 240 pixels.