Facial wearable items recognition refers to the judgment of whether the face in a face image wears a facial item, such as an eyeglass or a mask, which belongs to the category of face attribute analysis. In recent years, face attribute analysis mainly focuses on the study of gender, age, expression and other aspects, ignoring the study of facial wearable items recognition. However, this technology has great potential for application. It can be applied to large-scale image retrieval, and according to whether the face is wearing facial items to classify the face images. When searching for a face in the face database, searching directly in the category of images can accelerate the search to the target face. In this paper, an effective method of facial wearable items recognition is proposed. The color space conversion, image segmentation and feature extraction are used to analyze the multi-perspective and multi-angle face images effectively to judge whether the face wears eyeglass and mask. The related methods are discussed in detail in this paper, and the experiment proves the effectiveness of the algorithm.