2010 TRI/PRINCETON CONFERENCE 93 As expected by the theory, we observe a sinusoidal signal for both hair tresses. It is easy to observe the ±10° phase shift of the signal due to the orientation of the hair tress, on both hair types. The average intensity and its modulation for dark hair is about 3 times less than the blond hair. We can observe the measurement noise on the dark hair signal. From this data, we can easily measure the angle of orientation θ. ANGLE PRECISION In order to measure the angle of orientation θ and the precision of the measurement, we place a straightened blond hair bundle on a precision mechanical rotation stage (Figure 11). We measure the angle of orientation θ versus the theoretical angle θ in order to evaluate the precision of the angle measurement. Figure 12 shows the measured angle θ versus the theoretical angle θ. The experimental data are in very good agreement with the theory. Figure 13 shows the measured angle θ versus the theoretical angle θ for a small variation of θ. We estimate the angle precision to be +-0.1°. This is precise enough for the type of ex- periment and image we will carry on. EXPERIMENTAL RESULTS STUDY OF FLYAWAY HAIR Single hair fi ber resolution is achieved at a given stand-off distance by selecting the appro- priate resolution and objective lens for the camera. Fig 14 shows the image of a single hair Figure 11. Blond hair bundle on a precision mechanical rotation stage.

JOURNAL OF COSMETIC SCIENCE 94 Figure 12. Measured angle θ versus the theoretical angle θ. fi ber (blond) in the NIR (a) and its corresponding orientation image (b) In those images, the pixel resolution is about 80 μm × 80 μm. We can detect without any problem the orientation of the single hair fi ber. An application of single hair fi ber imaging is the detec- tion and identifi cation of fl yaway hair. Flyaway hair is caused when hair strands pick up Figure 13. Measured angle θ versus the theoretical angle θ for a small variation of θ.