Skin Detection Technique Based on HSV Color Model and SLIC Segmentation Method ⋆ Kseniia Nikolskaia 1 , Nadezhda Ezhova 1 , Anton Sinkov 1 , and Maksim Medvedev 1 South Ural State University, Chelyabinsk, Russia nikolskaya174@gmail.com,eris909@mail.ru, sinkov 96@mail.ru,own77d@gmail.com Abstract. The paper is devoted to new skin detection technique based on the HSV color model and SLIC segmentation method. The algorithm of skin detection is described. Experiments results are presented. The influence of training images on the skin detection is shown. New skin detection algorithm implemented in Python language using OpenCV li- brary is described. Keywords: Skin detection · HSV · SLIC Superpixels · Computer vision · Pattern recognition 1 Introduction Skin detection is the process of finding skin-colored pixels and regions in an image or a video. Skin detection applications are used for personality recognition, body-parts tracking, gesture analysis and adult content filtering and etc [1]. When the standard RGB color space is used, the skin detection can be very difficult under conditions of variable lighting and contrast. Therefore, the input image must be converted to another color space [2–4] that is invariant or at least insensitive to lighting changes, such as HSV [5]. The implemented skin detector converts the image into required color space and then uses the image histogram to mark each pixel: whether it belongs to skin. Image pixels are grouped in superpixels using the SLIC clustering method [6]. Thus, advantages of our skin detector are high processing speed and invariance under rotation and lighting changes. The main steps of skin detection in the image are: 1. download the input image; 2. convert image to HSV color space; 3. generate the image histogram; 4. apply classifier to determine the probability of a given pixel being skin- colored; ⋆ The work was partially supported by the Government of the Russian Federation according to Act 211 (contract №02.A03.21.0011.) and the Council on grants of the President of the Russian Federation (contract SP - 5430.2018.5).