Hand Biometric Verification with Hand Image-Based CAPTCHA

Two-stage human verification using HandCAPTCHA and anti-spoofed finger biometrics with feature selection

“…So, the HCA is less vulnerable, with 1.23% FARHC. Notably, γ-correction improves accuracy than only that of α-blending, compared to (Bera et al, 2018). Though, the probability is less likely in (Conti et al, 2015), yet, its accuracy and time are not satisfactory.…”

“…In our earlier conference paper (Bera et al, 2018), a simple two-stage human verification process has been proposed that leverages both the benefits of CAPTCHA and biometrics to enhance the security, and it showcases some elementary results. Now, in this proposed work, a novel spoofing attack detection method using quality metrics is newly introduced at Level-2 (which was not described in (Bera et al, 2018)) to strengthen the security of the system by allowing only legitimate users. We propose a new feature selection method that aims at optimizing the number of features required and minimizing the computation time for online-based verification.…”

“…Thus, the stripes are excluded from Algorithm-1. The experimental configuration of IHCAPTCHA verification described in our earlier work (Bera et al, 2018) has been briefed to maintain its consequences in the present experiments. During training, the volunteers are assigned to solve more than fifty IHCAPTCHA, and the accuracy is 94%.…”

Style matching CAPTCHA: match neural transferred styles to thwart intelligent attacks

“…FP-CAPTCHA [50] challenges the users to click on human facial points such as the eye, nose, and mouth, which are laid over a cluttered background and additional noises. HandCAPTCHA is implemented using a randomized combination of two real and five to seven fake hand-images [6]. In addition to hand biometric verification, liveness detection is added to the verification pipeline to improve security [7].…”

Development of authentication-based CAPTCHA mechanism on touch screen environment

“…This HandCAPTCHA is an approach for hand images that used the physiological biometric recognition based on hand images that was implemented over the traditional Text-based CAPTCHA. The ability of this method is it works against malicious threats with 98.34% accuracy [71]. In 2020, the keystroke features integrated with the IP address of the user are applied to classify a person under the use of Text-based CAPTCHA.…”