Method and apparatus for analyzing images of a subject's body hair and skin

Abstract

In one aspect, a computer-implemented method for analyzing images of hair and skin of a subject's body is provided, the method including (i) receiving 101 one or more images of a body part of the subject, the one or more images including a skin region corresponding to skin of the body part and a hair region corresponding to hair of the body part, (ii) processing 103 the one or more images to determine a diffuse reflectance of the skin region at different wavelengths of light, (iii) processing 105 the one or more images to determine a diffuse reflectance of the hair region at different wavelengths of light, and (iv) determining 107 a first contrast measure for a first pair of values ​​of a first wavelength and a second wavelength, the first contrast measure being determined from the intensities of the diffuse reflectance of the skin region and the hair region in a first spectral band and a second spectral band of light. (v) repeating step (iv) in order to determine one or more further contrast measures for each of the further value pairs of the first and second wavelengths; and (vi) selecting first and second spectral bands corresponding to the value pairs of the first and second wavelengths that provide contrast measures that satisfy a criterion in which each of the first wavelength values ​​is in the range of 425 nm to 650 nm and each of the second wavelength values ​​is in the range of 600 nm to 850 nm, and the first wavelength value and the second wavelength value are different in the value pairs.

Description

[Technical Field]

[0001] The present disclosure relates to methods and apparatus for analyzing images of body hair and skin of a subject. [Background technology]

[0002] There are many different types of personal care operations that can evaluate images of hair and skin to determine characteristics of the skin, hair, or both and take these characteristics into account when performing the personal care operation or when providing guidance or recommendations to a user of a personal care device performing the personal care operation. Exemplary personal care operations include shaving, haircutting, hair removal, hair regrowth, photoepilation, skin massage, skin phototherapy, etc. In a haircutting or shaving device, it is useful to determine the thickness or density of the hair to be cut or shaved, for example, because this can be used to adjust the cutting or shaving parameters of the personal care device. In a photoepilation or skin phototherapy device, it is useful to determine the thickness or color of the hair or skin tone, for example, to use the appropriate light power and / or wavelength, and it is also useful to determine the number of hairs per area during treatment to enable monitoring the effectiveness of the photoepilation treatment.

[0003] While it is easy to capture images of hair and skin using cameras on smartphones, tablets, smart mirrors, and the like, analyzing the images to distinguish between hair and skin in the images or to determine the characteristics of hair and skin in the images is not always easy. It is known that the characteristics of skin and hair vary from person to person, and a general approach is known to "optimize" the recognition of hair on skin by defining skin types and hair types and deriving appropriate parameters from these general types. However, given the differences in skin types and hair types and their combinations, such optimization is not optimal in practice.

[0004] Figure 1 illustrates various skin type and hair color combinations. Six skin types, labeled I–VI from lightest (Type I) to darkest (Type VI), are defined, along with six hair color types: white, light blonde, dark blonde, light brown, dark brown, and black. Grid 2 illustrates the possible skin type and hair color combinations, although six of these combinations (marked with hatching) are not considered possible. Example images 4, 5, 6, and 7 were acquired by the device using common settings for different skin type / hair color combinations. Image 4 is of a dark-skinned, dark-haired subject; image 5 is of a light-skinned, dark-haired subject; image 6 is of a light-skinned, brown-haired subject; and image 7 is of a light-skinned, white-haired, and brown-haired subject. In images 4, 5, and 6, the hair is easily discernible; however, in image 7, the hair is difficult to see due to the low optical contrast between the hair and the skin.

[0005] It is noted that US Patent Application Publication No. 2012 / 0224042A1 discloses a method for detecting a skin area by having two LEDs that emit light at different wavelengths and a camera that receives reflected light from an object at different times and creates first and second or subsequent images, which include at least a skin detection area that is used to detect the skin area.

[0006] Furthermore, it should be noted that a paper entitled "Detection of Skin Region from Multiband Near-IR Spectral Characteristics" (Electronics and Communications in Japan, Vol. 92, No. 11, 2009) explains that the unique reflectance properties of materials can be used to detect the driver's facial region using a camera when driving at night.

[0007] In the case of shaving, the results of simulations and user tests indicate that taking into account user characteristics and differences (hair, skin, user handling, etc.) is more important to end-shaving performance results than more general improvements in the shaver's mechanical design. This is supported by objective and subjective test results, which show that there is much more variability between users than between shaver types.

[0008] Therefore, taking into account individual differences in hair and / or skin type can improve shaving performance, and similar benefits are expected for other types of personal care procedures. A user's hair and / or skin type, as well as hair and skin characteristics appropriate for a personal care procedure, can be determined using sensors or questionnaires. Hair characteristics appropriate for personal care procedure personalization include color, thickness, shape, density, and / or orientation. Skin characteristics appropriate for personal care procedure personalization include color, shade, and the presence or absence of scars, moles, freckles, acne, etc.

[0009] Because the optical properties of skin and hair are similar, contrast between them is limited. This contrast depends not only on skin and hair color, but also on other properties of the skin and hair. For example, optical methods that provide high contrast with respect to skin and hair color mean that light, thin hair (unpulped blond hair) on light skin (i.e., skin types 1-2), as in Image 10, is difficult to detect. This can reduce the effectiveness of hair detection and potentially reduce the potential for personalization. Summary of the Invention [Problem to be solved by the invention]

[0010] Therefore, there is a need for improved techniques for analyzing images of a subject's body hair and skin, and in particular for selecting spectral bands for a subject that enable or enhance analysis of images of the subject's skin and hair. [Means for solving the problem]

[0011] In a first particular aspect, a computer-implemented method for analyzing images of hair and skin on a subject's body is provided, the method including the steps of: (i) receiving one or more images of a body part of the subject, the one or more images including a skin region corresponding to skin of the body part and a hair region corresponding to hair of the body part, (ii) processing the one or more images to determine diffuse reflectance of the skin region at different wavelengths of light, (iii) processing the one or more images to determine diffuse reflectance of the hair region at different wavelengths of light, and (iv) determining a first contrast measure for a first pair of values of first and second wavelengths, the first contrast measure being determined from the intensities of the diffuse reflectance of the skin region and the hair region in first and second spectral bands of light. (v) repeating step (iv) to determine one or more additional contrast measurements for each of the additional value pairs of the first and second wavelengths; and (vi) selecting first and second spectral bands corresponding to the first and second wavelength value pairs that provide contrast measurements that satisfy a criterion, wherein each first wavelength value is in the range of 425 nm to 650 nm and each second wavelength value is in the range of 600 nm to 850 nm, and the first wavelength value and the second wavelength value are different in the value pairs. Thus, in a first aspect, a technique is provided for selecting spectral bands for a subject that enables or enhances analysis of images of the subject's skin and hair.

[0012] Step (iv) includes determining a first contrast measure from (a) the difference between the intensity of the diffuse reflectance of the skin region in the first spectral band and the intensity of the diffuse reflectance of the skin region in the second spectral band, and (b) the difference between the intensity of the diffuse reflectance of the hair region in the first spectral band and the intensity of the diffuse reflectance of the hair region in the second spectral band. In these embodiments, step (iv) may include determining the first contrast measure as a ratio of the differences.

[0013] In some embodiments, the criterion is one of: the magnitude of the contrast measurement is above a threshold; the criterion is met when the magnitude of the contrast measurement is below a threshold; the criterion is met when the contrast measurement is the highest of the determined contrast measurements; the criterion is met when the contrast measurement is the lowest of the determined contrast measurements; and the criterion is met when the contrast measurement is an optimal value.

[0014] In some embodiments, the method further includes controlling one or more light sources to simultaneously or separately illuminate the subject with light comprising wavelengths corresponding to the selected first and second spectral bands, and receiving one or more additional images of the subject while the subject is illuminated with light comprising wavelengths corresponding to the selected first and second spectral bands. In these embodiments, the method may further include analyzing the received one or more additional images to determine one or more characteristics of the subject's skin and / or hair.

[0015] In an alternative embodiment, the method further includes analyzing the one or more images received in the selected first and second spectral bands to determine one or more characteristics of the subject's skin and / or hair.

[0016] In some embodiments, the method further includes receiving or determining an indication of the subject's skin tone and / or the subject's hair color in the received one or more images. In these embodiments, the method may further include repeating the method for a plurality of different subjects' hair and skin images having different skin tones and / or hair colors. In these embodiments, the method may further include storing, in a lookup table or database, the indications of the selected first and second spectral bands and / or the indications for pairs of first and second wavelength values corresponding to the selected first and second spectral bands, and the corresponding indications of skin tone and / or hair color. In these embodiments, the method may further include receiving an indication of the skin tone and / or hair color of a further subject to be imaged, and using the lookup table or database to determine the first and second spectral bands to be used to image the further subject according to the skin tone and / or hair color in the received indications. In these embodiments, the method may further include controlling one or more light sources to irradiate the additional subject with light including wavelengths corresponding to the determined first and second spectral bands, and acquiring images of the additional subject while the additional subject is irradiated with light including wavelengths corresponding to the determined first and second spectral bands. In these embodiments, the method may further include analyzing the acquired images of the additional subject in the determined first and second spectral bands to determine one or more characteristics of the additional subject's skin and / or hair.

[0017] In a second aspect, there is provided a computer program product comprising a computer readable medium having computer readable code embodied therein, the computer readable code being configured, when executed by a suitable computer or processor, to cause the computer or processor to perform a method according to the first aspect or any embodiment thereof.

[0018] In a third aspect, there is provided an apparatus for analyzing images of hair and skin on a subject's body, the apparatus including: (i) receiving one or more images of a body part of the subject, the images including a skin region corresponding to skin of the body part and a hair region corresponding to hair of the body part, (ii) processing the one or more images to determine diffuse reflectance of the skin region at different wavelengths of light, (iii) processing the one or more images to determine diffuse reflectance of the hair region at different wavelengths of light, and (iv) determining a first contrast measure for a first pair of values of first and second wavelengths, the first contrast measure being determined from the intensities of the diffuse reflectance of the skin region and the hair region in the first and second spectral bands of light, (v) repeating operation (iv) to determine one or more further contrast measurements for each of the further value pairs of the first and second wavelengths; and (vi) selecting the first and second spectral bands corresponding to the value pairs of the first and second wavelengths that provide contrast measurements that satisfy a criterion, wherein each first wavelength value is in the range of 425 nm to 650 nm and each second wavelength value is in the range of 600 nm to 850 nm, and the first wavelength value and the second wavelength value are different in the value pairs.

[0019] In some embodiments, the device receives one or more images from an image collection unit or a memory unit. In some embodiments, the image collection unit includes an illumination unit for generating light to illuminate a subject when collecting images. In some embodiments, the illumination unit selectively generates light in a particular spectral band.

[0020] In some embodiments, the device further comprises an image acquisition unit, hi other embodiments, the device is part of a system that includes an image acquisition unit. There is also provided a further embodiment of the third aspect, wherein the apparatus further performs a method according to any of the various embodiments of the first aspect.

[0021] These and other aspects will be apparent from and elucidated with reference to the embodiments described hereinafter. [Brief explanation of the drawings]

[0022] Exemplary embodiments are now described, by way of example only, with reference to the following drawings:

[0023] [Figure 1] Figure 1 shows some example images illustrating the combination of skin type and hair color. [Figure 2] FIG. 2 is a set of graphs showing the diffuse reflectance spectra of the skin and hair of several subjects with different skin types and hair colors. [Figure 3] FIG. 3 shows the dual spectral band derivative matrix for a particular subject with skin type II and light blonde hair. [Figure 4] FIG. 4 shows a graph of the contrast ratios of different spectral bands for the eight subjects in FIG. [Figure 5] FIG. 5 is a graph showing the improvement in contrast of hair and skin in images acquired using the described technique compared to images acquired using white light. [Figure 6] FIG. 6 illustrates an embodiment of an apparatus for analyzing images of a subject's body hair and skin in accordance with the techniques described herein. [Figure 7] FIG. 7 is a flow chart illustrating a method for analyzing images of body hair and skin of a subject. [Figure 8] FIG. 8 shows exemplary images and some graphs / plots illustrating the steps of the method of FIG. DETAILED DESCRIPTION OF THE INVENTION

[0024] As described above, analyzing images of a subject's skin and hair is useful, for example, to enable the parameters of a personal care operation to be adapted to the subject. As shown below, it has been found that capturing images from a subject using light of specific wavelengths can improve the contrast between hair and skin in the image and more easily derive characteristics of the hair and / or skin. Because different skin types and hair colors have different optical properties, there is no universal set of wavelengths that can be applied to all subjects. Therefore, the techniques described herein provide techniques for selecting spectral bands for a subject that are centered around specific wavelengths that enable or enhance analysis of images of the subject's skin and hair, particularly facial hair.

[0025] Before describing in more detail the techniques and apparatus for implementing this technique, the principles underlying this technique will be explained.

[0026] Below, we describe preliminary results obtained using a hyperspectral imaging system. This pilot experiment aimed to identify the spectral band that exhibits the greatest hair-skin contrast across all hair-skin types (i.e., all combinations of hair color and skin type). Using a hyperspectral imaging setup that utilizes polarized white light illumination and cross-polarized detection of light reflected from the skin in a bandwidth from 420 nanometers (nm) to 1100 nm in 5 nm steps, we performed hyperspectral imaging of the entire face of several subjects. Experiments were conducted on eight test subjects with different skin types, and the results are shown in Figure 2. Note that a hyperspectral imaging setup is not required to implement the techniques described herein, and therefore will not be described in detail.

[0027] FIG. 2 shows graphs illustrating the ratio of light intensity detected from skin (shown by line 10) at different wavelengths and the ratio of light intensity from hair (shown by line 12) at different wavelengths for each subject. The ratio of light intensity across a range of wavelengths detected from skin is referred to herein as the "skin diffuse reflectance spectrum," and the ratio of light intensity across a range of wavelengths detected from hair is referred to as the "hair diffuse reflectance spectrum." The eight subjects have different combinations of skin type and hair color. From left to right in the top row of graphs, subjects have skin type II and light blond hair, skin type II and dark blond hair, skin type II and light brown hair, and skin type III and dark blond hair. From left to right in the bottom row of graphs, subjects have skin type III and light brown hair, skin type III and dark brown hair, skin type IV and dark brown hair, and skin type V and black hair.

[0028] FIG. 2 also shows that hair-skin contrast is greatest in the spectral band from 550 nm to 850 nm, although the optimal range and location of the band will depend on the subject's particular hair-skin type.

[0029] In light of these results, we propose a differential imaging technique that evaluates light reflected from skin and hair in two or more specific spectral bands to enhance the contrast between the skin and hair. The center wavelength of each spectral band depends on the subject (particularly their skin type and / or hair color), and in some embodiments, the width of the spectral band also depends on the subject (particularly their skin type and / or hair color). While the technique is described below with reference to the use of light in two spectral bands, it will be understood that the technique can use light in three or more spectral bands. Table 1 below shows the broad wavelength range within which the center wavelengths of two specific spectral bands of a given subject lie. The spectral bands are defined by their respective center wavelengths, denoted λ1 and λ2, respectively. Table 1 shows that, regardless of skin type and hair color, the first spectral band has its center wavelength λ1 between 600 nm and 850 nm, and the second spectral band has its center wavelength λ2 between 425 nm and 650 nm. [Table 1]

[0030] Tables 2-9 below show the contrast obtained for different spectral band center wavelength pairs. The contrast was derived from the diffuse reflectance spectra shown in Figure 2. Thus, Table 2 shows the contrast obtained for different combinations of the first and second spectral bands for a subject with skin type II and light blond hair. [Table 2]

[0031] Table 3 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type II and dark blond hair. [Table 3]

[0032] Table 4 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type II and light brown hair. [Table 4]

[0033] Table 5 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type III and dark blond hair. [Table 5]

[0034] Table 6 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type III and light brown hair. [Table 6]

[0035] Table 7 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type III and dark brown hair. [Table 7]

[0036] Table 8 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type IV and dark brown hair. [Table 8]

[0037] Table 9 shows the contrast obtained with different combinations of the first and second spectral bands for a subject with skin type V and black hair. [Table 9]

[0038] Figure 3 shows the dual spectral band derivative matrix for a specific subject with skin type II and light blond hair. The graph in Figure 3(a) shows the calculated spectral integrated intensity difference between pairs of different spectral bands for skin. The graph in Figure 3(b) shows the calculated spectral integrated intensity difference between pairs of different spectral bands for hair. Additionally, the graph in Figure 3(c) shows the relative contrast in the form of a contour plot of the skin-hair dual spectral band derivative contrast ratio. Figure 3(c) allows visual determination of the optimal wavelength pair that results in high contrast between hair and skin. The relative contrast in Figure 3(c) is derived as the ratio between the graphs in Figure 3(a) and 3(b).

[0039] Figure 4 shows a graph of the contrast ratios derived as shown in Figure 3 for the eight subjects of Figure 2. Figure 4 shows that different optimal wavelength pairs yield maximum contrast values for different skin and hair color combinations, and therefore, no wavelength pair is applicable to all skin and hair color combinations. Therefore, as provided by the techniques described herein, the wavelengths of a wavelength pair can be "tuned" to improve or maximize the contrast for a particular hair-skin combination.

[0040] The graph in Figure 5 shows the improvement in hair-to-skin contrast in images acquired using the described technique in the appropriate dual-spectral band for the eight subjects in Figures 2 and 4 compared to images acquired using white light.

[0041] The block diagram of Figure 6 illustrates an embodiment of an apparatus 60 for analyzing images of body hair and skin of a subject in accordance with the techniques described herein. In this figure, apparatus 60 is shown as part of a system 62 that also includes a separate image collection unit 64 that acquires one or more images of the subject. In alternative embodiments, apparatus 60 may include image collection unit 64 or may include functionality to perform the functions of image collection unit 64. In other embodiments, apparatus 60 may be implemented independently of the device or unit that acquires images of the subject.

[0042] Device 60 can be any type of electronic or computing device. For example, device 60 is or is part of a smartphone, tablet, smartwatch, smart mirror, laptop, computer, or server (e.g., a server in a data center (also called "in the cloud"). In some embodiments, device 60 is or is part of a personal care device, such as a shaver, epilator, or skin treatment device. Image collection unit 64, if present, is in the form of or is part of a smartphone, tablet, smartwatch, smart mirror, laptop, or other device capable of obtaining an image of a subject. In some embodiments, image collection unit 64 is in the form of or is part of a personal care device, such as a shaver, epilator, or skin treatment device.

[0043] Device 60 includes a processing unit 66 that controls the operation of device 60 and may be configured to perform or implement the methods described herein. Processing unit 66 may be implemented in various ways using software or hardware to perform the various functions described herein. Processing unit 66 may include one or more microprocessors or digital signal processors (DSPs), which may be programmed using software or computer program code to perform the required functions and / or to control the components of processing unit 66 to cause the required functions. Processing unit 66 may be implemented as a combination of dedicated hardware (e.g., amplifiers, preamplifiers, analog-to-digital converters (ADCs) and / or digital-to-analog converters (DACs)) to perform some functions and processors (e.g., one or more programmed microprocessors, controllers, DSPs, and associated circuitry) to perform other functions. Examples of components that may be employed in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, DSPs, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), hardware for implementing neural networks, and / or so-called artificial intelligence (AI) hardware accelerators (i.e., processors or other hardware designed specifically for AI applications that can be used in conjunction with a main processor).

[0044] Processing unit 66 is connected to memory unit 68, which may store data, information, and / or signals used by processing unit 66 in controlling the operation of device 60 or in performing or carrying out the methods described herein. In some implementations, memory unit 68 stores computer-readable code executable by processing unit 66 to cause processing unit 66 to perform one or more functions, including the methods described herein. The memory unit 68 may include any type of non-transitory machine-readable medium, such as a cache or system memory, including volatile and non-volatile computer memory such as random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), read only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), and electrically erasable PROM (EEPROM), and the memory unit 68 may be implemented in the form of a memory chip, an optical disk (such as a compact disk (CD), digital versatile disk (DVD), or Blu-Ray disk), a hard disk, a tape storage solution, or a solid-state device such as a memory stick, solid-state drive (SSD), or memory card.

[0045] In some embodiments, the device 60 may also include interface circuitry 70 to enable data connection and / or exchange with other devices, such as the image collection unit 64, and optionally with other devices, such as a server, database, user device, etc. The connection may be direct or indirect (e.g., via the Internet); thus, the interface circuitry 70 may connect the device 60 and the image collection unit 64 over a network (e.g., the Internet) or directly connect the device 60 and the image collection unit 64 via any desired wired or wireless communication protocol. For example, the interface circuitry 70 may operate using WiFi, Bluetooth®, ZigBee®, or any cellular communication protocol. For wireless connections, the interface circuitry 70 (and thus the device 60) may include one or more suitable antennas for transmitting and receiving over a transmission medium (e.g., air). Alternatively, for wireless connections, the interface circuitry 70 may include means (e.g., connectors, plugs, etc.) for connecting the interface circuitry 70 to one or more suitable antennas external to the device 60 for transmitting and receiving over a transmission medium (e.g., air). The interface circuit 70 is connected to the processing unit 66 so that information or data received by the interface circuit 70 can be provided to the processing unit 66 and information or data from the processing unit 66 can be transmitted by the interface circuit 70.

[0046] In some embodiments, device 60 includes a user interface 72 that includes one or more components that allow a user of device 60 (e.g., a subject) to input information, data, and / or commands into device 60 and / or that allow device 60 to output information or data to the user of device 60. For example, user interface 72 may include a display screen for displaying one or more images of the subject and / or analysis results of the subject's skin and hair images. User interface 72 may include any suitable input component, including, but not limited to, a keyboard, a keypad, one or more buttons, switches, or dials, a mouse, a trackpad, a touchscreen, a stylus pen, a camera, a microphone, etc. And / or user interface 72 may include any suitable output component, including, but not limited to, a display screen, one or more lights or light elements, one or more speakers, a vibration element, etc.

[0047] It will be appreciated that a practical implementation of device 60 may include additional components in addition to those shown in Figure 6. For example, device 60 may also include a power source, such as a battery, and components to enable device 60 to be connected to a mains power source.

[0048] The image collection unit 64 is provided to collect one or more images of the subject's skin and hair. These images are analyzed by the device 60 according to the techniques described herein. Accordingly, the image collection unit 64 includes an imaging unit 74 for collecting images. The imaging unit 74 includes any components suitable for capturing one or more images or a video sequence composed of multiple images, such as a charge-coupled device (CCD) and one or more lenses and / or mirrors. In some embodiments, the imaging unit 74 is a camera, such as a digital camera. In some embodiments, the imaging unit 74 can be configured or configurable to acquire images in one or more specific spectral bands, for example, at least two spectral bands each having a center wavelength between 425 nm and 850 nm. For example, the imaging unit 74 includes one or more filters to allow light in the desired spectral bands to be used to form the image. Alternatively, a sensor within the imaging unit 74 can be configured to sense only light in one or more specific spectral bands. In other embodiments, imaging unit 74 is sensitive to light between 425 nm and 850 nm, and subsequent image data processing can extract or select image data corresponding to the desired spectral band.

[0049] In some embodiments, image acquisition unit 64 may include interface circuitry 76 to enable data connection and / or exchange with other devices such as apparatus 60 (via interface circuitry 70), and optionally any other devices such as servers, databases, user devices, etc. Interface circuitry 76 may be implemented in a manner similar to interface circuitry 70 described above.

[0050] In some embodiments, the imaging unit 74 acquires an image of the subject using ambient light in the subject's environment. However, in other embodiments, the image collection unit 64 includes an illumination unit 78 for generating light to illuminate the subject as the imaging unit 74 collects the image. The illumination unit 78 can generate white light or other light in a broad wavelength spectrum, for example, at least from 425 nm to 850 nm. In other embodiments, the illumination unit 78 can selectively generate light in a specific spectral band, for example, at least two spectral bands each having a center wavelength in the range of 425 nm to 850 nm. The illumination unit 78 can generate light in the specific spectral bands simultaneously or at different times (i.e., the illumination unit 78 can generate light in a first spectral band during a first period and light in a second spectral band during a second period). In some embodiments, the illumination unit 78 generates light in the required spectral bands. In particular, the illumination unit 78 can be controlled to generate light in two spectral bands that are particularly suited to the subject being imaged, as identified in accordance with the techniques described herein. In embodiments in which the lighting unit 78 produces light in multiple spectral bands, the lighting unit 78 may be provided with an appropriate arrangement of light sources, filters, etc. to enable it to produce light of the required wavelengths.

[0051] The flowchart in Figure 7 illustrates a method for analyzing an image of body hair and skin of a subject. The method may be implemented by device 60, for example, by processing unit 66 executing suitable computer code stored in memory unit 68. The method is also described with reference to Figure 8, which shows an exemplary image 80 and several graphs / plots 82-96 illustrating the steps of the method.

[0052] Thus, in step 101, one or more images of a body part of a subject are received by device 60. The images may have been acquired by image collection unit 64. The images may be received directly from image collection unit 64, for example, in real time as the images are collected, or the images may be received by retrieving the images from a memory, such as memory unit 68. An exemplary image 80 is shown in FIG. 8. The body part of the subject in image 80 includes one or more "skin regions" 81, which are areas of skin on the body part, and one or more "hair regions" 82, which are areas of hair on the body part. Note that the hair regions 82 correspond to the hairs themselves, while the skin regions 81 include the areas of skin between the individual hairs.

[0053] In some embodiments, multiple images are received in step 101, with each image 80 being acquired for a different wavelength or narrow wavelength band of light. Each image 80 may be acquired by image acquisition unit 64 while the body part is continuously or sequentially illuminated with light of a different wavelength or narrow wavelength band. Alternatively, multiple images may be acquired by image acquisition unit 64 while the body part is illuminated with white light, with imaging unit 74 using filters or other means to capture a set of images from each wavelength or each different narrow wavelength band of light.

[0054] In step 103, the image 80 is processed to determine the intensity of diffuse reflectance from the skin area 81 at different wavelengths of light. Figure 8 shows an exemplary graph 84 illustrating the intensity of diffuse reflectance from the skin area 81 for different wavelengths, where R skin (λ). The diffuse reflectance of skin area 81 for different wavelengths is also referred to as the "skin area diffuse reflectance spectrum." Those skilled in the art will understand that the diffuse reflectance spectrum is a measure of the diffuse reflectance of light from the sample in image 80, and excludes the specular reflectance of light from the sample.

[0055] In step 105, image 80 is processed to determine the intensity of diffuse reflectance from hair region 82 at different wavelengths of light. Step 105 may occur before, after, or simultaneously with step 103. Figure 8 shows an exemplary graph 86 illustrating the intensity of diffuse reflectance from hair region 82 for different wavelengths, R hair The diffuse reflectance of the hair region 82 for different wavelengths is also referred to as the "hair region diffuse reflectance spectrum."

[0056] Step 103 and / or step 105 can be performed in a number of different ways. In either case, the intensity of diffuse reflectance reflected from the skin region 81 and from the hair region 82 at a series of different wavelengths or a series of narrow wavelength bands is determined. In embodiments where each image 80 is associated with a particular wavelength or a particular narrow wavelength band, each image 80 can be analyzed to determine the intensity of diffuse reflectance from the skin region 81 and the hair region 82 in the image 80. For an image 80 at a particular wavelength (or narrow wavelength band), the intensity of diffuse reflectance from the skin can be determined by averaging the intensity of several pixels in the image 80 that correspond to the skin region 81. This is repeated for the images 80 at different wavelengths to form a skin region diffuse reflectance spectrum. In some embodiments, the intensity of diffuse reflectance is determined using only a subset of pixels that correspond to the skin region 81. For example, only a subset of pixels in a portion of the image 80 that does not include the hair region 82 and / or is not near the hair region 82 (e.g., to avoid errors or artifacts due to hair shadows) is used. In a similar manner, the hair region diffuse reflectance spectrum can be determined from image 80 .

[0057] It will be appreciated that for some of the above embodiments of step 103 and / or step 105, prior to determining the diffuse reflectance intensity, the received image 80 can be processed to identify one or more portions of the image 80 that correspond to skin and one or more portions that correspond to hair. Techniques for identifying skin and / or hair in an image are known in the art and will not be described further herein.

[0058] In step 107, a first contrast measurement is determined for a pair of first values of the first and second wavelengths. Specifically, the first contrast measurement is determined from the diffuse reflectance intensity of the skin region 81 and the hair region 82 in the first and second spectral bands of light. The first and second spectral bands are centered around a first wavelength value (λ1) and a second wavelength value (λ2), respectively. The first wavelength is a value between 425 nm and 650 nm, and the second wavelength is a value between 600 nm and 850 nm. The spectral bands are also referred to herein as "detection bands." Plot 87 in FIG. 8 illustrates an exemplary first spectral band centered at λ1, and plot 88 in FIG. 8 illustrates an exemplary second spectral band centered at λ2. The first and second wavelengths have different values. The first spectral band is defined as a function F1 of wavelength λ, peaking at λ1, and represents a sensitivity curve or filter to be applied to the diffuse reflectance spectra determined in steps 103 and 105. Similarly, a second spectral band is defined as a function F2 of wavelength λ with a peak at λ2, and represents a sensitivity curve or filter applied to the diffuse reflectance spectra determined in steps 103 and 105.

[0059] In some embodiments, the detectable skin diffuse reflectance intensity S in the first spectral band skin1 (λ1) is calculated as follows: S skin1 (λ1)=Σ λ R skin (λ)F1(λ) (1) That is, it is the sum of the products of the first spectral band and the skin region diffuse reflectance spectrum for all wavelengths (more specifically, 425 nm to 850 nm). skin1 (λ1) is shown in plot 90 of FIG.

[0060] In these embodiments, the detectable skin diffuse reflectance intensity S in the second spectral band skin2 (λ2) is also calculated as follows: S skin2 (λ2)=Σ λ Rskin (λ)F2(λ) (2) That is, it is the sum of the products of the second spectral band and the skin region diffuse reflectance spectrum for all wavelengths (more specifically, 425 nm to 850 nm). skin2 (λ2) is shown in plot 92 of FIG.

[0061] In these embodiments, the detectable hair diffuse reflectance intensity S in the first spectral band hair1 (λ1) is also calculated as follows: S hair1 (λ1)=Σ λ R hair (λ)F1(λ) (3) That is, it is the sum of the products of the first spectral band and the hair region diffuse reflectance spectrum for all wavelengths (more specifically, 425 nm to 850 nm). hairn1 (λ1) is shown in plot 94 of FIG.

[0062] Finally, in these embodiments, a detectable hair diffuse reflectance intensity S in the second spectral band is hair2 (λ2) is also calculated as follows: S hair2 (λ2)=Σ λ R hair (λ)F2(λ) (4) That is, it is the sum of the products of the second spectral band and the hair region diffuse reflectance spectrum for all wavelengths (more specifically, 425 nm to 850 nm). hairn2 (λ2) is shown in plot 96 of FIG.

[0063] In some embodiments, a contrast measure can be determined from the difference between the diffuse reflectance intensity detected for the skin region in a first spectral band (e.g., plot 90) and the diffuse reflectance intensity detected for the skin region in a second spectral band (plot 92), and the difference between the diffuse reflectance detected for the hair region in the first spectral band (plot 94) and the diffuse reflectance detected for the hair region in the second spectral band (plot 96). In some embodiments, the contrast measure is determined from the ratio of these differences. In particular, the contrast measure for a first pair of values for the first and second wavelengths is determined as follows:

number

[0064] Thus, equation (5) gives the contrast of skin regions relative to hair regions. It will be appreciated that (less preferably) the contrast measure may also be related to the contrast of hair regions relative to skin regions by inverting the ratio in equation (5).

[0065] In a more computationally efficient approach than that outlined above with respect to equations (1)-(5) used to determine the skin region diffuse reflectance spectrum and the hair region diffuse reflectance spectrum in step 103 and step 105, the first contrast measure can be calculated directly for a first pair of values for the first and second wavelengths as follows:

number

[0066] Similar to equation (5), the contrast measure in equation (6) gives the contrast of skin regions relative to hair regions.

[0067] Next, in step 109, step 107 is repeated for additional pairs of values for each of the first and second wavelengths to determine one or more additional contrast measurements. That is, one or both of the first and second wavelength values are changed and additional contrast measurements are determined as described above. While step 107 can be repeated any number of times, it is preferred that contrast measurements be determined for a range of first wavelength values from 425 nm to 650 nm and a range of second wavelength values from 600 nm to 850 nm.

[0068] In some embodiments, a skin spectral difference map can be determined for different values of the first and second wavelengths, and a hair spectral difference map can be determined for different values of the first and second wavelengths. The skin spectral difference map is given as: D skin (λ1,λ2)=|S skin1 (λ1)-S skin2 (λ2)| (7) The hair spectral difference map is given as: D hair (λ1,λ2)=|S hair1 (λ1)-S hair2 (λ2)| (8)

[0069] Finally, in step 111, the plurality of contrast measurements are evaluated to select a particular pair of first and second spectral bands. In particular, a pair of first and second spectral bands is selected in which a pair of corresponding values of the first and second wavelengths provides a contrast measurement that satisfies a criterion. For example, this criterion may be the highest contrast measurement, and thus the first and second spectral bands (and corresponding first and second wavelengths) that provide the highest contrast measurement are selected. Mathematically, step 111 can be expressed as follows:

number

[0070] The criteria evaluated in step 111 can take any suitable form. For example, the criteria can be or include a threshold value, and the criteria can be met by a contrast measurement value that exceeds the threshold. Alternatively, the criteria can be or include a threshold value, and the criteria can be met by a contrast measurement value that is below the threshold. Alternatively, the criteria can be met by the highest of the determined contrast measurements. Alternatively, the criteria can be met by the lowest of the determined contrast measurements. In another alternative, the criteria is met when the contrast measurement value is an optimal value, which can be a maximum or minimum value.

[0071] In some embodiments, once the first and second spectral bands are selected in step 111, one or more light sources (e.g., in illumination unit 78) may be used to simultaneously or separately illuminate the subject with light including wavelengths corresponding to the selected first and second spectral bands. Imaging unit 74 may be used to acquire one or more further images of the subject illuminated with this light, which are received by processing unit 66. These further images may be analyzed to determine one or more characteristics of the subject's skin and / or hair.

[0072] Alternatively, after selecting the first and second spectral bands in step 111, the images received in step 101 may be analyzed to determine one or more characteristics of the subject's skin and / or hair. In particular, if multiple images 80 are received in step 101, each image 80 being acquired for a respective wavelength or narrow range of wavelengths of light, the analysis to determine one or more characteristics of the skin and / or hair may be performed on the images 80 corresponding to the selected first and second spectral bands.

[0073] In any of the above embodiments, the one or more characteristics of the hair include, for example, color, thickness, shape, density, and / or orientation. The one or more characteristics of the skin include, for example, color, shade, presence or absence of scars, moles, freckles, acne, etc. Techniques for deriving these characteristics from images are known in the art and will not be further described herein.

[0074] In some embodiments, the analysis results can be used in a personal care operation, such as shaving, to improve the performance of the personal care operation or to provide feedback to the subject regarding the personal care operation. For example, the analysis can determine that there are (still) hairs to be shaved on a body part and therefore that the body part needs to be shaved (again) for a cleaner shave or to achieve a particular facial hair style. Alternatively, the analysis can determine that all hairs have been sufficiently shaved, informing the subject that the body part does not need to be shaved again, thereby reducing skin irritation from unnecessary shaving strokes. In the case of a personal care operation that involves applying light to a body part, such as photoepilation, the analysis results of the skin and / or hair can be used to select an appropriate power level of light output from the photoepilator and / or an appropriate wavelength of light output from the photoepilator. Feedback to the subject determined from the analysis of the skin and / or hair can be provided to the subject via the user interface 72. From the above, it will be appreciated that images are acquired and analyzed during the performance of a personal care operation, such as shaving or a photoepilation treatment, to provide guidance and feedback regarding the personal care operation.

[0075] In some embodiments, the method further includes receiving or determining an indication of the subject's skin tone and / or hair color in the image received in step 101. Such an indication may be input by the subject themselves, for example via user interface 72. Alternatively, as described above, the skin tone and / or hair color may be determined from an analysis of the image itself. The skin tone and / or hair color indication may be stored in a look-up table or database, along with pairs of first and second wavelength values corresponding to the first and second spectral bands selected in step 111.

[0076] To expand the lookup table or database, the method of FIG. 7 can be repeated for multiple subjects with different skin types and / or hair colors, and these skin types and / or hair colors can be stored in the table or database along with an index to pairs of first and second wavelength values corresponding to the first and second spectral bands selected for each subject.

[0077] Thus, when imaging another subject, the skin tone and / or hair color can be entered into device 60 (e.g., using user interface 72), and a lookup table or database can be consulted to determine appropriate first and second wavelengths to use for that skin tone and / or hair color. Illumination unit 78 can be controlled to illuminate the subject with light in the appropriate spectral band, and one or more images of the subject can be acquired for analysis to determine other characteristics of the skin and / or hair. Alternatively, the subject can be illuminated with white light (e.g., by illumination unit 78), and images can be acquired from light in the appropriate spectral band using appropriate filters in imaging unit 74. Again, these images can be analyzed to determine other characteristics of the skin and / or hair.

[0078] Variations of the disclosed embodiments can be understood and effected by those skilled in the art in practicing the principles and techniques described herein, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the singular form of an element does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain means are recited in mutually different dependent claims does not indicate that a combination of these means cannot be used to advantage. A computer program can be stored or distributed on any suitable medium, such as an optical storage medium or a solid-state medium, supplied together with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless communication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. 1. A computer-implemented method for analyzing an image of body hair and skin of a subject, the computer-implemented method comprising: (i) receiving one or more images of a body part of the subject, the one or more images including a skin region corresponding to skin of the body part and a hair region corresponding to hair of the body part; (ii) processing the one or more images to determine the diffuse reflectance of the skin area at different wavelengths of light; (iii) processing the one or more images to determine the diffuse reflectance of the hair region at different wavelengths of light; (iv) determining a first contrast measure for a first value pair of a first wavelength and a second wavelength, the first contrast measure being determined from (a) a difference between the diffuse reflectance intensity of the skin region in a first spectral band and the diffuse reflectance intensity of the skin region in a second spectral band, and (b) a difference between the diffuse reflectance intensity of the hair region in the first spectral band and the diffuse reflectance intensity of the hair region in the second spectral band, the first spectral band and the second spectral band being centered around the value of the first wavelength and the value of the second wavelength in the first value pair, respectively; (v) repeating step (iv) to determine one or more additional contrast measurements for additional pairs of values of each of the first wavelength and the second wavelength; (vi) selecting a first spectral band and a second spectral band corresponding to the pair of values of the first wavelength and the second wavelength that provides a contrast measurement that meets a criteria; Including, 10. A computer-implemented method, wherein each value of the first wavelength is in a range of 425 nm to 650 nm and each value of the second wavelength is in a range of 600 nm to 850 nm, and in each pair of values, the value of the first wavelength and the value of the second wavelength are different.

2. The computer-implemented method of claim 1 , wherein step (iv) comprises determining the first contrast measure as a ratio of the differences.

3. the criterion is met if the magnitude of the contrast measure is above a threshold; the criterion is met if the magnitude of the contrast measure is below a threshold; the criterion is met if the contrast measure is the highest of the determined contrast measures; The criterion is met if the contrast measure is the lowest of the determined contrast measures; and the criterion is met when the contrast measure is at an optimum value; 3. The computer-implemented method of claim 1, wherein the optimum value is one of a maximum value or a minimum value.

4. controlling one or more light sources to simultaneously or separately illuminate the subject with light comprising wavelengths corresponding to the selected first and second spectral bands; receiving one or more additional images of the subject while the subject is illuminated with the light including wavelengths corresponding to the selected first and second spectral bands; The computer-implemented method of claim 1 or 2, further comprising:

5. 5. The computer-implemented method of claim 4, further comprising analyzing the received one or more additional images to determine one or more characteristics of the skin and / or hair of the subject.

6. 3. The computer-implemented method of claim 1, further comprising analyzing the one or more images received in the selected first and second spectral bands to determine one or more characteristics of the skin and / or hair of the subject.

7. 3. The computer-implemented method of claim 1 or 2, further comprising receiving or determining an indication of the skin tone of the subject and / or the hair color of the subject in the received one or more images.

8. The computer-implemented method of claim 7 , further comprising repeating the computer-implemented method for hair and skin images of a plurality of different subjects having different skin tones and / or hair colors.

9. 8. The computer-implemented method of claim 7, further comprising storing in a lookup table or database indices of the selected first and second spectral bands and / or indices for the pairs of values of the first and second wavelengths corresponding to the selected first and second spectral bands, and corresponding indices of the skin tone and / or the hair color.

10. receiving an indication of skin tone and / or hair color of a further subject to be imaged; using the look-up table or database to determine a first spectral band and a second spectral band to be used for imaging the further subject according to the skin tone and / or hair color in the received indicia; The computer-implemented method of claim 9 further comprising:

11. controlling one or more light sources to illuminate the further subject with light comprising wavelengths corresponding to the determined first and second spectral bands; acquiring an image of the additional subject while the additional subject is illuminated with light including wavelengths corresponding to the determined first and second spectral bands; The computer-implemented method of claim 10 further comprising:

12. 12. The computer-implemented method of claim 11, further comprising analyzing the images of the additional subject acquired in the determined first and second spectral bands to determine one or more characteristics of the skin and / or hair of the additional subject.

13. A computer readable medium having computer readable code embodied thereon that, when executed by a suitable computer or processor, causes the computer or processor to perform the computer-implemented method of claim 1 or 2.

14. 1. An apparatus for analyzing an image of body hair and skin of a subject, the apparatus comprising: (i) receiving one or more images of the subject's body part, the images including a skin region corresponding to skin of the body part and a hair region corresponding to hair of the body part; (ii) processing the one or more images to determine the diffuse reflectance of the skin area at different wavelengths of light; (iii) processing the one or more images to determine the diffuse reflectance of the hair region at different wavelengths of light; (iv) determining a first contrast measure for a first value pair of a first wavelength and a second wavelength, the first contrast measure being determined from (a) a difference between the diffuse reflectance intensity of the skin region in a first spectral band and the diffuse reflectance intensity of the skin region in a second spectral band, and (b) a difference between the diffuse reflectance intensity of the hair region in the first spectral band and the diffuse reflectance intensity of the hair region in the second spectral band, the first spectral band and the second spectral band being centered around the value of the first wavelength and the value of the second wavelength in the first value pair, respectively; (v) repeating operation (iv) to determine one or more additional contrast measurements for additional pairs of values of each of the first wavelength and the second wavelength; (vi) selecting a first spectral band and a second spectral band corresponding to the pair of values of the first wavelength and the second wavelength that provides a contrast measurement that meets a criteria; Run wherein each value of the first wavelength is in a range of 425 nm to 650 nm and each value of the second wavelength is in a range of 600 nm to 850 nm, and in any pair of values, the value of the first wavelength and the value of the second wavelength are different.

Images