j. Cosmet. Sci., 52, 237-250 (July/August 2001) Development of a device to measure human hair luster YUTAKA TANGO and KOICHI SHIMMOTO, KOSt• Corporation, Research & Development Division, Fundamental Research Laboratory, 1-18-4 Azusawa, Itabashi-ku, Tokyo 174-0051, Japan. Accepted for publication April 15, 2001. Synopsis Evaluating human hair luster is important in developing effbctive hair care products. Many methods of measuring human hair luster have been proposed, but most have major disadvantages: some require cutting the subject's hair, some utilize bulky equipment, and some take much time. A device that does not impose excessive burden on the subject and hair, but can easily and conveniently measure human hair luster, has not yet been developed. Overcoming the disadvantages of the traditional method, our new device can measure luster accurately without cutting the hair. Neither the subject's hairstyle nor its color influences the measurements. The device is small, and the time required for measurement is only 0.2 sec. The hair of 84 subjects was evaluated using this device, and there was a high correlation between the sensory score and the measurements obtained. INTRODUCTION Luster is an extremely important factor in evaluating the health of human hair and the effectiveness of its care. Moreover, hair luster is one of the esthetic standards for judging hair, and is important in evaluating hair-finishing cosmetics. Many methods of mea- suring human hair luster have been proposed some utilize goniophotometric meter measurements and calculation of the ratio of specular reflection and diffuse reflection (1). Others analyze hair luster by image processing, using a TV camera to capture an image of a hairstyle and dividing it into three planes of red, green, and blue (2). However, these methods have disadvantages: The goniophotometric method requires cutting of the subject's hair to use as a sample, and the reflective characteristics obtained from a single hair sample do not necessarily reflect the state of luster when human hair is gathered. The TV-camera method is influenced by the subject's hairstyle or head shape, and measurement takes a considerable amount of time. The increasing prevalence of dyed or bleached hair has made measurement of hair luster even more difficult. To overcome these disadvantages, we developed a new measurement device. In this study, the optical reflective process in human hair was assumed to be consistent with the dichromatic reflection model (DRM) of Shafer (3), and so verification was performed and a measurement theory was established. The device based on our theory was then actually developed, and human hair luster was evaluated. 237

238 JOURNAL OF COSMETIC SCIENCE THEORY Human hair consists of the cuticle, cortex, and medulla. The cortex and medulla contain melanin, which provides the hair a particular color tone (4). Therefore, human hair can be considered as a heterogeneous substance (e.g., cortex, medulla) in which a coloring substance (e.g., melanin) is embedded. The reflective process of light on such a hetero- geneous substance is commonly expressed using a DRM (dichromatic reflection model) (3,5). DRM corresponds to the Phong's model in computer graphics, and is also called SRM (standard reflective model) or NIR (neutral interface reflection model). The method assumes that the reflective light from a heterogeneous substance can be described as a linear combination of two different reflective compositions, as shown in Figure 1. The first composition is called a specular reflection or an interface reflection, and is recognized as the color of the illumination light--the color of the composition reflected from the surface of the object, regardless of its color. The second composition, which corresponds to optical dispersion by coloring particles of matter, is called a diffuse reflection or an internal reflection, and is recognized as the object's color. In DRM, when a composition of specular reflection is Is and a composition of diffuse reflection is Id, the measuring value of a reflectance Ir is given as: Ir(0,)•) -- Is(0) + Id(0,)•) (1) A composition of specular reflection is expressed as: Is(0): (2) Generally, gloss of an object (luster) is expressed by the strength of compositions of specular reflection (6), and luster L is the same as the composition of specular reflection Is in equation 2, as follows: L(0) = Ir(0,}t) - Id(0,}t) (3) EXPERIMENTAL VERIFICATION OF THE DICHROMATIC REFLECTION MODEL IN HUMAN HAIR DRM is not effectuated in all objects. For example, a copper plate will have unique reflective characteristics, and will not apply to DRM. Therefore, it was necessary to confirm whether DRM would be effectuated in human hair. An important assumption of DRM is that spectrum compositions of a reflection are constant, even if there is any change in the measurement angle, since separation of wavelength and a geometric parameter is possible. By performing the examination on this point, it is possible to verify whether human hair can be described by DRM. The schematic representation of an optical system is shown in Figure 2. In our experiment, the reflectance of human hair was measured at the incident angle 0 of 45 degrees, with the receiving angle 0 changing from -20 degrees to 70 degrees. INFLUENCE OF ARRANGEMENT ANGLE OF HUMAN HAIR ON REFLECTANCE The influences on reflectance by an arrangement angle, which is made by the plane containing both light sources and a receiver, and the direction in which human hair