In this blog post, we’ll explore the principles behind capacitive fingerprint recognition technology, its applications in mobile devices, and the challenges that still need to be addressed.

The Evolution of Fingerprint Recognition Technology and the Emergence of Capacitive Fingerprint Recognition

When Apple’s “iPhone 5s” was released in 2013, one of the technologies that garnered significant global attention was fingerprint recognition. While various changes—such as the high-resolution camera and 64-bit processor—drew attention, what captured the most interest was the full-scale popularization of fingerprint recognition technology, which anyone could easily use on a mobile device. In the 19th century, fingerprint recognition was used as a specialized police technique for criminal investigations, but beginning in the 1960s, it began to be utilized in various security fields due to its high recognition rate and fast verification speed. Subsequently, with the advent of capacitive fingerprint recognition technology suitable for mobile devices, individuals were able to easily use biometric authentication, and fingerprint recognition established itself as one of the most widely used technologies in the biometric market. So, how does capacitive fingerprint recognition work, and what challenges remain to be addressed in the future?

The Structure of Fingerprints and the Principles of Capacitive Fingerprint Recognition

First, to recognize a fingerprint, one must identify its distinctive features. Fingerprints consist of ridges and valleys; the raised parts on the finger are called ridges, and the indented parts between them are called valleys. These ridges and valleys intertwine in a complex pattern to form a fingerprint; representative feature points include the ending point (where a ridge ends), the bifurcation (where a ridge splits), the upper core (the highest point of a ridge’s curve), the lower core (the lowest point of a ridge’s curve), and the delta (where a ridge converges at three points). When a fingerprint is scanned, multiple coordinates are generated based on these features; if the registered coordinates match the scanned coordinates, the fingerprint is recognized as the same one.
To generate these coordinates, a fingerprint image must first be acquired. Various technologies are used for fingerprint recognition, including optical methods that utilize prisms and light, thermal methods that rely on temperature differences, CCD sensor methods, and ultrasonic methods. Currently, the most widely used method in mobile devices is capacitive fingerprint recognition. This method recognizes the shape of a fingerprint by utilizing the subtle differences in capacitance that occur when the ridges of the fingerprint come into contact with the sensing electrodes, while the valleys maintain a certain distance from the electrodes.
So, what happens if a foreign object is trapped between the ridges and valleys? Capacitive fingerprint recognition is not significantly affected by general dirt or surface debris. This is because the epidermis is an insulator that does not conduct electricity well, whereas the dermis beneath it is relatively conductive. Since the sensing electrodes measure the characteristics of the conductive areas, small surface debris or shallow scratches do not significantly affect recognition.

Fingerprint Templates and the Fingerprint Recognition Process

After capturing a fingerprint image, the image must be processed to establish coordinates. This involves mathematical algorithms and generally consists of three stages: smoothing, binarization, and thinning.
The smoothing process removes noise generated by the hardware and enhances the contrast between light and dark areas, making the fingerprint lines appear more natural and distinct. This process is further divided into histogram processing and median filtering. Histogram processing increases the contrast of pixels with unclear light and dark tones to make the lines more distinct, while median filtering uses the values of surrounding pixels to reduce noise and smoothly connect the lines.
Binarization is the process of classifying the smoothed image into 0s and 1s based on light and dark levels.
After this process, the fingerprint can be stored as digital data. In the final step, line thinning, the thickness of the ridges is reduced to approximately 1 pixel to enable more accurate extraction of distinctive features. The resulting matrix of 0s and 1s is called a coordinate or fingerprint template.
Mobile devices store only the fingerprint templates generated through this process; the original fingerprint images are generally not stored. When a user attempts authentication, the system compares the newly generated fingerprint template with the stored template to determine if they match. Furthermore, since it is extremely difficult to reconstruct the original fingerprint image from the template alone, this helps enhance security. Recently, most smartphones have further strengthened security by managing fingerprint data in a dedicated secure area rather than in general storage, where it is encrypted.

Challenges of Capacitive Fingerprint Recognition Technology

Capacitive fingerprint recognition technology has established itself as a leading biometric technology for mobile devices, offering both convenience and a high level of security. However, there are still challenges that need to be addressed. People whose fingerprints have been damaged due to prolonged manual labor or accidents, or those who have difficulty with fingerprint recognition—whether due to congenital or acquired conditions—may experience difficulties during the authentication process. Furthermore, recognition rates may vary depending on the condition of the finger or the surrounding environment.
Today, various biometric technologies—such as facial recognition, iris recognition, and vein recognition—are used in conjunction to compensate for these limitations, and some devices combine multiple authentication methods to enhance both security and convenience. It is expected that fingerprint recognition technology will continue to address these limitations and evolve into an even safer and more convenient biometric technology in the future.