In the face of the growing influence of facial recognition, the second half of fingerprint recognition is gradually becoming less prominent. Many smartphone manufacturers are now exploring new technologies for facial recognition, despite its current limitations in user experience and technical maturity. However, the pursuit of innovation continues, and the development of fingerprint recognition is now encountering two major challenges.
Since the introduction of Face ID on the iPhone X, facial recognition has become a hot trend. Whether it's vivo, OPPO, or Huawei, all have either adopted or are preparing to adopt facial recognition technology. As a result, fingerprint recognition, which was once widely used, is now experiencing a slowdown in development.
[Image: "Face Recognition and Fingerprint Recognition Develop Together – Why Is Fingerprint Recognition One Step Behind?"]
Although fingerprint recognition remains the best option for full-screen devices, the development of optical or ultrasonic under-screen fingerprint technology has not yet reached the commercial stage. Therefore, the iPhone X opted for 3D facial recognition instead. Companies like vivo, OPPO, and others have adopted a coexistence model, combining facial recognition with capacitive fingerprint sensors. From this perspective, the competitiveness of fingerprint recognition is slowly fading.
However, user feedback on the iPhone X suggests that Face ID is not as fast or convenient as traditional fingerprint scanning. Additionally, many domestic brands still treat facial recognition as a novelty rather than a primary security method. At one press conference, Luo Yonghao from HuaWei pointed out that facial recognition is less secure than fingerprint recognition and recommended users continue to rely on fingerprint authentication.
So, how can capacitive fingerprint recognition remain competitive when facial recognition hasn't fully met user expectations and under-screen fingerprint technology is still in development?
According to Guo Zhenhua from Tsinghua University’s Shenzhen Graduate School, there are two main research directions. First, applying deep learning to enhance the accuracy of fingerprint recognition. Second, collecting physiological features of living bodies to improve anti-spoofing capabilities.
Currently, fingerprint recognition faces two key challenges. On one hand, its performance under large-scale population conditions is limited. Some users have unclear or difficult-to-read fingerprints, and while one-to-one matching is accurate, the accuracy drops significantly as the number of users increases. On the other hand, the anti-spoofing capability of fingerprint systems is still weak. There have been numerous reports of fake fingerprints being cracked using films, photos, or even broken fingers.
To address these issues, Guo Zhenhua proposed solutions. For improving accuracy, he suggested integrating deep learning into fingerprint recognition. Unlike traditional methods, which may hit a performance ceiling with large data sets, deep learning allows for continuous improvement as more data becomes available, leading to better accuracy.
Regarding anti-spoofing, Guo introduced Optical Coherence Tomography (OCT), a medical imaging technique originally used for diagnosing eye and cardiovascular diseases. This technology uses near-infrared light and optical interference to create detailed images of biological tissues. It can capture both internal and external fingerprints, as well as detect sweat glands between layers. Fake fingerprints lack these internal structures, making them easier to identify.
Additionally, Optical Angular Gradient (OAG) can be used to process OCT signals, separating static and dynamic particles in the sample. Guo explained this by comparing it to milk flow in a pipeline—higher flow rates increase the dynamic signal strength. In fingerprint terms, this helps distinguish real fingerprints with blood flow from fake ones without it.
While fingerprint recognition may appear less dominant compared to facial recognition, it has been a standard feature in smartphones for years. It will likely continue to be popular in the low-end market, where growth is steady. With the integration of deep learning and live detection technologies, fingerprint recognition is becoming more accurate and secure, increasing the difficulty and cost of attacks. As such, its future remains uncertain, but it still holds significant potential in the evolving landscape of biometric security.
Stage Lights,Beam Lights,Stage Concert Lights,Moving Head Lights
Guangzhou Chengwen Photoelectric Technology co.,ltd , https://www.cwleddisplay.com