组合滤波EOH-SIFT特征和模糊推理实现的掌纹识别

Palmprint recognition based on integrated filtering EOH-SIFT features and fuzzy reasoning

为了提高掌纹图像对于位置、方向和亮度变化的鲁棒性,提出一种结合滤波技术实现掌纹识别的算法。首先,按特定顺序过滤过滤器,然后提取掌纹图像的EOH-SIFT特征并创建类别数据库。将要识别的掌纹图像的关键点与类别数据库中的所有关键点之间的模糊匹配度的最大值用作该点的模糊隶属度,由模糊推理规则完成掌纹识别。实验结果表明,所提算法提高了EOH-SIFT特征识别掌纹的有效性,并且对掌纹图像的位置、方向、亮度变化和噪声具有较高的鲁棒性。

An algorithm combined with the filtering technique to realize the palmprint recognition is proposed to improve the robustness of palmprint image to the change of position,direction and brightness.The filters are filtered in a specific order,and then the EOH⁃SIFT features of palmprint images are extracted to build the classification database.The maximum value of the fuzzy matching degree between the key points of the palmprint image under recognition and all key points in the classification database is used as the fuzzy membership of the point,and the palmprint recognition is completed with the fuzzy inference rules.The experimental results demonstrate that the proposed algorithm can improve the effectiveness of the palmprint recognition according to the EOH⁃SIFT features,and has the robustness of palmprint image to the change of position,direction and brightness as well as noise.