Topcon’s AI development strategy: “Sensing AI”

With the progress and widespread use of Internet of Things (IoT), huge amounts of sensing data are being collected from the real world. Artificial intelligence (AI) services that instantaneously offer solutions generated by computing are rapidly growing around the world.

Yet there are intrinsic noises within the sensing data, which may adversely affect the quality of AI services.

Topcon has been by far providing high-performance sensors to the fields of healthcare, agriculture, and infrastructure.

By combining our high-performance sensor technology with AI, we are further developing the technology to reduce and/or remove noises and fluctuations from the sensing data, as part of our missions to provide the ever improved data quality to our users.

“Sensor to AI” is the key of Topcon’s R&D in fulfilling this mission.

It is expected that the number of IoT devices will reach 20 billion by 2025, and with the increasing connectivity, data from the real world can be instantaneously acquired and connected.

With the ever growing amount of data, the rapid progress of deep learning, and the increasing computational power, data analysis methods based on AI are being widely developed.

While there are many AI companies developing the application AI (AI app) that offers decisions made by AI, noise that remains in the input data may limit the accuracy in the decision making.

The “Sensor to AI” under development at Topcon makes a step forward to reduce or remove the noises that are within the raw data acquired from the sensors, and thus lower the uncertainly and raise the accuracy for the decisions to be made.

The concept of Topcon “Sensor to AI” is to develop both sensor and AI technology and implement them into one unit. With our strong foundations of sensor technology and our new development efforts in AI, we are striving to bring the innovative and practically high value products to the customers.