首页>
外国专利>
METHOD OF COMPARING IMAGES, NOTABLY FOR IRIS RECOGNITION, IMPLEMENTING AT LEAST ONE QUALITY MEASUREMENT DETERMINED BY APPLYING A STATISTICAL LEARNING MODEL
METHOD OF COMPARING IMAGES, NOTABLY FOR IRIS RECOGNITION, IMPLEMENTING AT LEAST ONE QUALITY MEASUREMENT DETERMINED BY APPLYING A STATISTICAL LEARNING MODEL
展开▼
机译:比较图像的方法,尤其是用于虹膜识别的方法,是采用统计学习模型确定的至少一项质量度量的实施方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
The present invention relates to a method of comparing at least one test image with at least one reference image, comprising the steps consisting in: a) generating from the test image (I) a plurality of thumbnail images, b) calculating for each thumbnail image: - at least one quality measurement (Qj(b)) associated with this thumbnail image, - at least one quantity representative of the correlation (Ci) with the reference image or a subimage of the latter, c) calculating for the test image (I) a comparison score S(I,J) with the reference image (J), this score being a function of the quantity or quantities representative of the correlation (Ci) and of the quality measurement or measurements (Qj(b)).
展开▼