Those who love criminal investigation anti-terrorism dramas must have seen the American drama "Person of Interest." In the play, Harold Finch, a genius of AI technology, developed an artificial intelligence system called "The Machine" that can predict terrorist attacks.

The system uses image and video recognition technology to identify the characters and vehicles captured by the lens, cleanse, classify, and link the information, and then determine whether there are abnormal behaviors based on daily activity data analysis. If a suspicious person is detected, the person's information is compared with the information in the library, allowing the AI ​​to assist the police in identifying the crime information and even predicting the possible criminal behavior. The efficiency of the criminal investigation work has thus been greatly improved.

However, if you think that real criminal investigation relies only on face recognition technology, then you are already out. Whether in criminal investigation, public security or intelligent transportation, your casual "Freestyle" (such as gait movements) can become an important factor in AI recognition.

Gait recognition, as an uncontrolled recognition method, can greatly improve the work efficiency in a specific industry. As an emerging technology, “gait recognition” will gradually become a new milestone in the recognition of humans after the recognition of human faces and irises. As the "national team" of identifying Chinese gaits, the Yinhe water droplets have long been widely used in such fields as security, criminal investigation, rail transportation, and industrial identification, and the technological dividend of "gait recognition" is gradually being highlighted.

Create a pan-identity recognition platform and identify gait as "optimal solution"

As the current emerging technology, gait recognition has become a powerful factor in the field of biometric identification with the characteristics of long distance recognition, wide application range, and non-controlled characteristics, which promotes the continuous development of identity recognition technology.

1. General physiological characteristics can only be identified

What we call BIOMETRICS technology refers to personal identification through the use of the physical characteristics (fingerprints, irises, facial, DNA, etc.) or behavioral characteristics (gait, keystroke habits, etc.) inherent in the human body. Technology.

Judging from the market performance of current power feature recognition, ordinary physiological feature recognition has certain fragmentation in its application process due to its own unique defects. For example, fingerprints and iris recognition all have their own specific technology and it is difficult to Many elements are perfectly integrated.


Specifically, technologies such as fingerprint recognition, face recognition, and iris recognition are difficult to achieve both in terms of ease of use, safety, and accuracy. High-accuracy iris recognition technology requires high accuracy, and relatively simple face recognition and fingerprint recognition are not accurate.

However, in the current “society of efficiency”, “single fight alone” with a single common physiological characteristic can hardly satisfy people’s fast-paced and high-efficiency work and life demands. Therefore, the development of a more convenient and efficient identification method will be likely to become a biological recognition. The "new blue ocean" in the field.

2. Gait recognition + face recognition, build a new model of pan-identity recognition

In view of this situation, can we combine individual behavior characteristics with physiological characteristics to improve the convenience and accuracy of recognition?

The generic identification model built by Yinhe Water Drops provides a solution to this problem. The Galaxy Water Drainage Identity Recognition Model, which fully utilizes individual behavioral characteristics and physiological characteristics, builds a new model of pan-identity recognition through the method of “full-view gait recognition + full-type face recognition” to assist the development of biometric recognition.

For gait recognition, it avoids the disadvantages of identifying other physiological features. On the one hand, gait recognition has a wide range of applications and can be identified beyond 50 meters. On the other hand, "non-controlled" guarantees that gait recognition does not require object cooperation and can achieve 360-degree dead-end recognition; finally, different The characteristics of body shape, head shape, muscle strength, sensitivity of motor nerves, walking posture and other characteristics determine the gait has better distinguishing ability.

As a gait recognition “national team” of the Galaxy Drop Enterprise, based on the realization of the basic gait recognition technology, it also applies full-scale face recognition and crowd computing to the construction of pan-identity identification to achieve more accurate identification. result. Taking the full-scale face recognition technology as an example, the Galaxy Drop is one of the few companies in the industry that can provide all types of face recognition technology. The face recognition scheme based on ARM embedded platform achieves 99% recognition rate on mainstream databases and can maintain real-time performance. In long-distance face recognition, real-time face recognition can be performed on 25-meter crowds in ordinary 2K cameras. .


3. Biometric "New Blue Ocean"

In CCTV1’s artificial intelligence program “Wistic Man”, Yinhui Dripping CEO Huang Yongsheng led his project to successfully defeat the “Most Powerful Brain” memory master, and identified the target “suspected” from 10 people of similar size and height, from 21 The golden retriever dog with the similar coat color and the silhouette identified the target golden retriever and was praised by the Turing Award Winner Yao Chih-chih for "wisdom."

Galactic Dripping Success Recognizes Target Golden Retriever Through Gait

In fact, gait recognition technology has been widely used. As the “new blue ocean” for the development of the identification field, Yinhe Drop applies its technology to multiple fields such as smart security, smart home, robotics, and smart medical care.

Before the technology lands, it is still necessary to solve these problems

At present, with the continuous improvement of China’s scientific and technological strength, especially advanced technologies such as gait recognition have shifted from “running runners” to “leaders”. While we need technological breakthroughs, we can actively promote the technology to land, and then better benefit the society to give back to the community. However, there are still several problems that need to be resolved in the technology of gait recognition.

From the current point of view, the perfect commercialization of new technologies will require the “few levels” to be faced:

First, the commercial nature of AI is to increase production efficiency. Only when the efficiency of AI application is much higher than that of humans, it will have a certain value of landing. Take the security field as an example. At present, the public security system still lacks effective video analysis and retrieval tools. On average, every current case still requires manual review of hundreds or even thousands of hours of video data. The workload is voluminous and the police resources are greatly occupied. , affect the efficiency of detection. Galactic droplets use gait recognition technology to detect criminal suspects, can quickly target the target population, greatly increase the level of public security system information, speed up the detection of various criminal cases.

Second, whether the technology has a market. The market is the terminal for technology applications. In terms of market development and commercialization promotion, Galaxy Drop collaborates with the United States, public security agencies, and other institutions to jointly develop smart home appliances based on the “human-computer new generation” strategy of the United States; and an integrated search system for face gait recognition created jointly with public security agencies. To further expand the artificial intelligence market.

Finally, is there sufficient adequacy of technological innovation? With the continuous development of AI technology, gait recognition as an emerging technology still needs innovation or progress. Galactic dripping gait recognition technology, as an important project of the STS (Science and Technology Service Network Program) of the Chinese Academy of Sciences, is technically directed by Academician Tan Tiediu, and has more than 100 R&D teams to assist in the subsequent innovation breakthrough.

In general, the development of gait recognition technology undoubtedly provides a “booster” for smart home, transportation, criminal investigation, security and other fields. The technology backs up the society and commercialization provides powerful support for technology research and development at the same time. The gait recognition technology, the "new blue ocean," will eventually be further and further away on the road to help businesses.

DTF Film

We are a China-based factory offering PET printing film for DTF printer printing factory, we are looking for buyers and global agents. DTF film roll, DTF printing film sheets, matte dtf film, glossy dtf film, hot & Cold Peeling Film, and various others for you to choose from!

From 2012 years beginning to export to overseas markets, covering multiple countries and regions around the world, including Southeast Asia, the Middle East, Europe, and North America.

Dtf Film,Pet Printing Film,Hot Peeling Dtf Film,Dtf Printing Film

Guangzhou YESUN Digital Technology Limited , https://www.dtffilmcn.com