Recently researchers have started working in the field of sensory decoding for amputees. They have used either the visual feedback or the response from the subject as their feedback. Many researchers, however, believe that sensory decoding can become more precise if the neuronal changes that occur in the process of stimulations can be isolated and measured. In this study, we have used lab-developed multichannel functional near-infrared spectroscopy (fNIRS) to discriminate between hemodynamic responses (HRs) of two different touch stimulations (handshake and ball grasp) applied to the right hand of five healthy male subjects. For all the stimulations, we measured HRs from the left somatosensory cortex. For each stimulation, the activated areas in the brain are analyzed by the t-value comparison between the measured data and the desired hemodynamic response function. The results showed that the corresponding primary somatosensory cortex area is activated for each type of stimulation. Linear discriminant analysis is utilized to classify the acquired data into two classes based on the features (mean and peak values) attained from the HbO signals. Our results indicate the potential of fNIRS to be used as an objective measure to discriminate between different tactile stimulations in the somatosensory cortex of brain.