美瞳检测方法

• Format: Patent
• Language: Chinese
• Published: 02.11.2018
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

本发明公开了种美瞳检测方法，其包括以下步骤：a.数据采集模块对已知美瞳和活体虹膜进行采集，并将采集到的已知美瞳作为人工美瞳存储到数据库中；b.美瞳学习模块将数据库中的人工美瞳和活体虹膜作为训练样本，通过机器学习的方法得到美瞳分类模型，并将美瞳分类模型发布到第检测模块的美瞳分类器中；c.第检测模块对未知样本进行美瞳检测；d.第检测模块认定未知样本的检测结果是非美瞳后，第二检测模块对所述未知样本进行二次检测；e.美瞳学习模块根据更新的数据库，通过机器学习的方法优化美瞳分类模型，并重新发布到第检测模块的美瞳分类器中；f.优化后的美瞳分类模型对数据库中的机器美瞳再次进行判断，剔除机器美瞳中的非美瞳数据。 The present invention discloses a method for detecting beautifying contact lenses. The method comprises the following steps that: a. a data collection module collects the known beautifying contact lenses and living irises, and stores the collected known beautifying contact lenses as the artificial beautifying contact lenses in the database; b. a beautifying contact lenses learning module takes theartificial beautifying contact lenses in the database and the living irises as training samples, obtains the beautifying contact lenses classification model through the machine learning method, and publishes the beautifying contact lenses classification model to a beautifying contact lenses classifier of a first detectio