Researchers have developed an olfactory sensor that can unlock the phone by blowing.

Researchers have developed an olfactory sensor that can unlock the phone by blowing.

Researchers have developed an olfactory sensor that can unlock a mobile phone by blowing. Japanese researchers have developed an olfactory sensor that can identify individuals by analyzing compounds in their breath. Researchers have developed an olfactory sensor, which can also unlock the phone by blowing.

Researchers have developed an olfactory sensor, and blowing can also unlock the phone. What are the ways to unlock 1 on my smartphone?

People unlock mobile phones, from the initial key and graphic unlocking to the widely used fingerprint unlocking, voice unlocking and face recognition. Recently, a scientific research in Japan has provided another brand-new solution for unlocking smart phones-you can unlock them with just one breath.

The effect diagram of olfactory sensor shows that in the near future, it may be possible to use artificial olfactory sensor for personal identity authentication based on breath odor (image source: Kyushu University).

The research team of Kyushu University and Tokyo University cooperated to analyze the compounds in human breath through the artificial olfactory sensor system for the first time, so as to achieve the effect of verifying personal identity. They used 16 channel chemical resistance sensor array and combined with machine learning technology to create an "artificial nose", and successfully achieved an average accuracy of more than 97%. In addition, the study also shows the influence of the number of sensors on accuracy and reproducibility.

Recently, a related paper was published in Chemical Communications, titled "Individual Authentication of Breathers Based on Artificial Olfactory Sensor System and Machine Learning" [1].

Related papers (source: chemical communication)

The key to unlocking a smart phone is to confirm the uniqueness of a person's identity through biometric technology. In the past identification methods, the conventional methods are fingerprint, palm print, voice and face recognition. In addition, it also includes unusual identification methods such as acoustics and finger vein.

"These technologies depend on everyone's physical uniqueness, but they are not foolproof. Physical features may be copied or even damaged by injury, "Chaiyanut Jirayupat, the first author of the paper and a doctoral student in the Department of Materials Science and Engineering, Faculty of Engineering, University of Tokyo, told the media. "Recently, human smell has emerged as a new biometric identity, mainly using your unique chemical composition to confirm who you are."

Fig. 2 Artificial olfactory sensor for biometric authentication based on respiration (source: Kyushu University)

In order to achieve this effect, the team first thought of compounds produced from human skin to identify identity information. However, this method is not safe. Paradoxically, the identity information detected by the machine must meet a certain concentration to realize "recognition", but the skin cannot produce high concentration of volatile compounds.

Based on this, the researchers put forward another idea: can we use human breath for biometric identification?

"The concentration of volatile compounds in the skin may be as low as billions or trillions, while the compounds exhaled from the breath may be as high as millions," Gilaupat continued. "In fact, human breath has been used to identify whether a person has cancer, diabetes or even coronavirus pneumonia-19."

Image system and olfactory sensor. The subject first exhales into the collection bag. Then, the bag is connected to an olfactory sensor, which can analyze the compounds in an individual's breath. Based on the concentration of compounds, the machine learning system identifies individuals (source: Kyushu University).

After determining this new biometric method, the team first conducted a comprehensive analysis of the subjects' breath samples to determine which compounds can be used for biometric authentication. According to the related results, the sensor can identify human respiratory components including 28 compounds used for biological identification.

Specifically, the team made an "artificial nose" olfactory sensing system, and each independent channel in the olfactory sensor array 16 can identify a specific range of compounds. Then, the sensor data is transmitted to the machine learning system, and the chemical composition in each breath sample is analyzed in this way.

Figure 2 Graphic workflow of personal authentication based on breath odor perception (source: chemical communication)

In related experiments, the research team tested the breath samples of six subjects through this system. The results show that the system can not only identify individuals, but also achieve an average accuracy rate of 97.8%. The main group consists of people of different ages, genders and nationalities. When the researchers further expanded the sample size to 20 subjects, the accuracy remained at a high level and was not affected by the sample size expansion.

Will the respiratory components of the same person be disturbed by dietary factors, thus affecting the test results? It is with this in mind that the subjects began to keep fasting 6 hours before the relevant test to ensure the accuracy of the test.

Of course, if the system wants to be used on a large scale in the future, it needs to solve more technical problems, such as making the relevant results no longer affected by factors such as diet, but accurately identifying the user's identity information through breathing in any state.

However, the team is optimistic about solving this problem. Takeshi Yanagida, the author of the paper and a professor at the Graduate School of Engineering of the University of Tokyo, told the media: "Fortunately, our current research shows that adding more sensors and collecting more data can overcome this obstacle."

Researchers have developed an olfactory sensor, which can also unlock the phone by blowing. Biometric authentication such as fingerprint and iris scanning once appeared in movies. However, the use of these technologies has expanded, and fingerprint verification and facial recognition have become commonplace on smartphones. Now, a new option has been added to the biometric security toolkit: breathing. According to a report published in Chemical Newsletter on 22nd, Japanese researchers have developed an olfactory sensor, which can identify individuals by analyzing the compounds in their breath.

Combined with machine learning, this "artificial nose" constructed by 16 channel sensor array can authenticate up to 20 people, with an average accuracy rate of over 97%.

In this information and technology era, biometric authentication is an important way to protect valuable assets. From common fingerprint, palm print, voice and facial recognition to less common ear and finger veins, machines can use various biometric methods to identify a person.

The researchers explained that these technologies depend on each person's physical uniqueness, but they are not foolproof. Physical features may be copied or even destroyed by injury. Human body odor has become a new biometric technology, which uses personal unique chemical composition to confirm "who are you".

One of the targets is transdermal gas, a compound produced by the skin. However, these methods have their limitations, because the concentration of volatile compounds produced by skin is not enough for the machine to detect. Therefore, the research team turned to study whether it can be replaced by human breath. The concentration of volatile compounds in skin may be as low as a few parts per trillion, while the concentration of exhaled compounds may be as high as a few parts per million.

The research team began by analyzing the breath of the subjects and observed which compounds could be used for biological authentication. The researchers found that 28 compounds were feasible choices. On this basis, they developed an olfactory sensor array with 16 channels, each of which can identify a specific range of compounds. Then, the sensor data is transmitted to the machine learning system to analyze each person's respiratory composition and develop characteristics that distinguish individuals.

The researchers tested the system with six people's breath samples and found that it can identify individuals with an average accuracy rate of 97.8%. Even if the sample size is increased to 20 people, this high level of accuracy remains unchanged.

However, researchers say that there is still a lot of work to be done before this technology really enters the homes of ordinary people. In this study, they asked subjects to fast six hours before the test. Next, they will improve the technology so that it is not affected by diet. Current research shows that adding more sensors and collecting more data can overcome this obstacle.

Researchers have developed an olfactory sensor, which can unlock mobile phone fingerprints by blowing. Biometric authentication such as iris scanning once appeared in spy movies, but the use of this technology has expanded, and fingerprint verification and facial recognition have become commonplace on people's mobile phones. Now, a new option has been added to the biometric security toolkit: breathing. According to a study published in Chemical Newsletter on 22nd, Japanese researchers have developed an olfactory sensor, which can identify individuals by analyzing the compounds in their breath.

In this information and technology era, biometric authentication is an important way to protect valuable assets. From the most common fingerprints, palm prints, voice and facial recognition to the less common ear and finger veins, machines can use various biometric methods to identify a person.

The researchers explained that human smell is a new biometric technology, which uses a person's unique chemical composition to confirm "who you are".

One of them is transdermal gas, a compound produced by skin, but the concentration of volatile compounds produced by skin is difficult to be detected by machine. Therefore, the research team turned to study whether it can be replaced by human breath. In fact, human breath has been used to identify whether a person has cancer, diabetes or even COVID-19's disease.

The research team started with analyzing the breath of the subjects, and screened out the compounds that can be used for biological authentication, and * * * found 28 feasible compounds. On this basis, they developed an olfactory sensor array with 16 channels, each of which can identify a specific range of compounds. Then the sensor data is transmitted to the machine learning system, and the respiratory composition of each person is analyzed to distinguish individual characteristics.

The researchers tested the system with breath samples of six people and found that the average accuracy rate of identifying individuals was 97.8%. Even if the sample size is increased to 20 people, this high level of accuracy remains unchanged.

However, researchers say that there is still a lot of work to be done before this technology can really enter the application. In this study, they asked subjects to fast six hours before the test. Next, they will improve the technology so that it is not affected by diet. Current research shows that adding more sensors and collecting more data can overcome this obstacle.