Panasonic releases cloud-based facial recognition API for Japanese market
Panasonic has introduced to the Japanese market a facial recognition API that leverages deep learning technology.
With no initial fee and a usage-based pricing model, the API uses a cloud-based service that can be integrated with smartphone apps, websites and access management systems which turns it into a viable solution for identity verification, identity theft prevention, and entry/exit management needs in today’s market, according to the company announcement.
The API micro-service is part of Panasonic’s μSockets B2B IoT service currently available in Japan. μSockets can link to other APIs for services such as out-of-stock detection, demographic (age and sex) estimation, dwell time detection, package sorting assistance, certification issuance, and cyberattack detection.
Available for on-premise and cloud-based solutions, Panasonic’s API can be integrated with software for multiple industries such as marketing analysis, accounting, and entry/exit management.
Top features include a feature quantity generation method that not only recognizes the entire face, but applies deep learning technologies to identify key facial features, and a proprietary algorithm that calculates the similarity degree in face matching and adjusts photographic errors due to environment challenges. These specifications enable the technology to be accurate in any condition, whether it is bad lighting or the person is wearing sunglasses, Panasonic says.
The company started its research and development into facial recognition in the 1990s and has explored the features by integrating them with consumer projects or camera-based security solutions
Panasonic’s facial recognition technology has been successfully implemented for passport control at airports across Japan, for security and access control in a school in Missouri, and by a Danish football stadium to improve fan safety.