Although face recognition ushered in the long-awaited outbreak in market share, this does not mean that all the factors that hinder the development of face recognition have disappeared. Some technically recognized flaws still bring a lot of hidden dangers to the application of face recognition.
As a social networking site with hundreds of millions of users around the world, the introduction of face recognition technology that automatically recognizes people in photos has triggered a new round of privacy leak concerns. In addition, photo software and online services such as Google Picasa and Apple iPhoto also use face recognition technology. Facebook, Google Picasa and Apple have hundreds of millions of users, and the use of face recognition technology is bound to cause more difficult privacy issues.
As another example, with the rollout of the next generation Internet in the United States, the FBI of the United States will also launch its NGI facial recognition service, and IAFIS (American Biodatabase) will add criminal and civil fingerprint databases, including multimodality Biometric identifiers, such as iris scans, palm prints, photos, voice data, etc. The service will also allow law enforcement agencies to access photos of public and private surveillance cameras, which means that anyone who is not involved in a crime can appear in the database. The Electronic Outpost Foundation slammed the list of technologies that would make personal privacy protection worse.
In addition to the personal privacy protection that ordinary people know, what are the problems with face recognition?
First of all, technically, the longevity of the recognition template is still criticized for face recognition. Although the facial image is indeed the most intuitive and easy to judge biological feature, it is also the human body area that is most prone to large-scale changes. Due to changes in lighting, posture, and age, it is not easy to make the template always work with these changes.
Especially the issue of age change. As the age changes, the appearance of the face also changes, especially for teenagers, this change is more obvious. For different ages, the recognition rate of face recognition algorithms is also different. The impact of age changes on face recognition algorithms must also be addressed.
Secondly, face recognition is a relatively new technology in the entire biometrics, but this new wave is not only an advantage. It is undeniable that the rapid popularization of face recognition has some people's "early taste" psychology-people are full of curiosity about new technologies and eager to try. However, it is precisely because of "new" that its accuracy has not been sufficiently proven, so users are often prone to mistrust, so when face recognition with uncertain factors face, if you lose the freshness After that, I am afraid that many users will be more inclined to choose the well-known biometric technology.
Third, the occlusion problem is very serious for face image collection under non-cooperative situations. Especially in the monitoring environment, the monitored object may wear glasses, hats and other accessories, making the collected face image may be incomplete, which affects the subsequent feature extraction and recognition, and even leads to the face detection algorithm. Failure. How to effectively remove the influence of occlusion is a very urgent research topic.
Finally, the recognition of active faces requires a higher recognition speed of the system. These include the speed of face detection and the speed of face recognition. Face detection can be achieved within 40ms. In the database of 100,000 people, the entire process from face detection to face recognition can be completed within 2 seconds, which should meet the requirements of most practical applications.
With the widespread application of face recognition technology, some legal issues involving personal privacy and other infringements have also become prominent. Face recognition technology is used to maintain national security and public safety. It is used for public security, airports, border ports, security, etc., as well as smart access, door locks, attendance and other civilian markets. Is a great thing. However, due to the advanced nature and convenience of face recognition software, its development space is getting larger and larger, and ordinary home computers can be installed and run. Therefore, some people questioned: Can such a powerful identification software be used by anyone? Does collecting data involve infringement?
Therefore, although face recognition has achieved very brilliant achievements in the market, "salted fish turned over" to achieve its own take-off. However, if you want to completely realize half of the land that captures fingerprint products, and realize the absolute mainstream of applications, I am afraid there are still many roads to be pursued.
The Barrier Terminal Blocks can realize safe, reliable and effective connection, especially in the environment of high current and high voltage.
Barrier Terminal block with easy connection, These connectors are available in all types of wires with 6.35-60mm pitch, with cover and without cover available.
Barrier Terminal Blocks
Barrier Terminal,Barrier Terminal Blocks,Barrier Terminal Block Connector,Panel Mount Barrier Terminal Block
Suzhou WeBest Electronics Technology Co.Ltd , https://www.webestet.com