A Color Spot Extraction Method Based on the Multifused Enhancement Algorithm

The color spot can reflect the skin status and the physiological change. It has been used in the evaluation of skin quality and medical diagnosis. However, the gray level of the color spot is very similar to that of normal skin. Because the shape, size, and position of the color spots are irregular. It is difficult to extract the accurate region of the color spot with the traditional methods. In order to extract the region of the color spot, we propose a multifused enhancement algorithm to enhance the feature of the color spot and extract the accurate region. The multifused enhancement algorithm mainly includes three components: the polarized image acquirement, the color model transformation, and the wavelet transform and enhancement. The color spot, which is in the intraepidermal basal cell, is captured by the polarized camera with the polarized light. The acquired method can remove the reflective light, covering the color spot. To further enhance the feature of the color spot, we use saturation instead of the gray level to characterize the color spot. The color spot is more conspicuous in the saturation image. Then, the wavelet transform and enhancement are used to enhance the feature of the color spot and reduce the effect of uneven illumination. The color spot is extracted by the proposed self-adaptive segmentation method. In addition, we discuss the performance with the different wavelets to choose the optimal wavelet. The actual reason for removing uneven illumination in the proposed algorithm is also analyzed in this work. The experimental results indicate that the proposed algorithm achieves the best accuracy among all the algorithms compared.