Frequency histogram

From VibraImagePedia
Jump to: navigation, search

Frequency histogram (FH) shows head movement frequency distribution for every pixel and for accumulated time period (default time period is equal 20 seconds). FH is one of base vibraimage characteristics. Vibraimages (AV, FV, Aura-VI) display spatio-temporal distribution of vibration, but FH displays temporal vibration distribution, irrelative to movement place. FH informative shows the difference between vibraimage and local Facial Acting Coding System (FACS) [1] technologies, because FACS is only internal facial coordinates sensitive and is not sensitive to head movements. Contrary to FACS, FH is the base for emotional parameters calculation and allows using principals and terms of information theory [2], mathematics, statistics, probability theory and thermodynamics as energy, information, probability density function and entropy for emotion theory [3] and practice. Tests and experiments confirm that character of FH depends on human emotional state and character of person. Some real samples histograms and explanations coordinated with emotions and energy levels are setting on this article.

Frequency Histogram Samples

histogram1 histogram2

Normal distribution also called Gaussian distribution is applicable in many fields. In vibraimaging and FH normal distribution usually corresponds to normal emotional and physiological states. The position of maximum probability density function (PDF) depends on energy level of person [4]. Calm person in normal state has maximum PDF moved to the left side of frequency scale. High active person in normal state has maximum PDF moved to the right side of frequency scale. Standard deviation and variancy of PDF also depends of emotion normality level and typically lower standard deviation and lower variancy corresponds to better emotion state or positive emotions, like happiness level.

histogram8 histogram7

Extreme left or right frequency position of normal distribution characterized extreme levels of emotion tiredness or activity, it could be corrected without pathology, during rest or relaxation.


Several modes and visible PDF asymmetry display emotional or physiology pathology as aggression and stress in person state.


This histograms show frequency spectrums of high speed vibraimage signals as two frames difference (without accumulation). This signals analogical to other physiological sygnals as EEG, ECG and GSR. It is well known that relation between high/low frequency in spectrum depend on human state and firstly from nervousness and anxiety. More anxiety and nervousness corresponds to more level of high frequency density in spectrum.


1. P. Ekman and W. V. Friesen. Facial Action Coding System. Consulting Psychologists Press Inc., 577 College Avenue, Palo Alto, California 94306, 1978.

2.Shannon, C.E. A Mathematical Theory of Communication. Bell System Technical Journal, 27, pp. 379–423 & 623–656, July & October, 1948. PDF.

3. Viktor Minkin, Libb Thims. Vibraimage and Human Thermodynamics. 2007

4. Emotion Recognition System VibraImage Version 7.0 Manual, Elsys Corp., publications, 2008