*Related Literature*

As development in science and technology progresses, smarter ways of operating household and office equipment are becoming more popular. For instance, Reference [20] described a gesture-based technology that makes use of some specific grammatical terms in its operation. This is possible due to these technologies and continuous issues of fake identity behind finger print detection [22] and walking patterns (also known as "gait") [23].

According to Reference [24], traditional singular biometry identifiers are flawed due to varying interclass sensitivity, data that bears a lot of noise, or very high error margins. For instance, in today's world of mobile phones, Reference [25] tried to analyze using a mobile phone, hand segmentation, and fingertip readings of about a hundred subjects. It was observed that the sensitivity of the technique yielded around 52%, a figure which may have been more precise if the method was bi-biometric-based or multi-biometric based. As a result of the sensitivity shortcoming common to uni-biometric systems, Reference [24] proposed a multi-model technique by adopting fused faces and fingerprints, i.e., making use of a collaboration-based classifier to identify the faces of a different person. A selective neural network was jointly used with "Viola–Jones method-based PCA" by Reference [26] for a gridding system. Here, over 100 faces from a face-based database were distinguishable, which proves the success

of the method. In comparison to the work of References [25,27], the merged vascular recognition with hand geometry in a multi-modal design derives an equal error rate of 6%. While singular biometric techniques are currently less preferred, the unique and greater sensitivity of multi-biometric technologies seem to be revolutionizing the field of biometry. In a fascinating technological twist, biometry has stepped forward beyond mere gesture identification to commence blood-based verification using physiological features, as reported by Reference [28]. Table 1 summarizes a few multi-modal biometric studies (within the last 10 years) with their corresponding error rates. From the table, HG denotes hand geometry, FV denotes finger vein (vascular structure), PP is the palm print, MFV is the multi-finger vein, FK is the finger knuckle, KS is the knuckle shape, and FG is the finger geometry.


**Table 1.** Some multi-modal biometric studies from 2009 until today.

Except otherwise stated, the following performance ratings (abbreviations) have been used throughout this paper and in similar biometric literature.


The rest of this paper is organized as follows. Section 2 further describes the problems upon which the study lies, and for which more multi-modal techniques are required. Section 3 describes in detail the newly proposed multi-biometric method, while experimental testing of the system and discussion of the generated results are reflected in Section 4. The final section discusses the limitations of the new method and suggests ways to improve the limitation in future research studies.
