Logo des Repositoriums

Auflistung nach Autor:in "Cremer, Sandra"

1 - 2 von 2
  • Konferenzbeitrag
    How a local quality measure can help improving iris recognition
    (BIOSIG 2012, 2012) Cremer, Sandra; Dorizzi, Bernadette; Garcia-Salicetti, Sonia; Lempérière, Nadège
    The most common iris recognition systems extract features from the iris after segmentation and normalization steps. In this paper, we propose a new strategy to select the regions of normalized iris images that will be used for feature extraction. It consists in sorting different sub-images of the normalized images according to a GMM-based local quality measure we have elaborated and selecting the N best sub-images for feature extraction. The proportion of the initial image that is kept for feature extraction has been set in order to compromise between minimizing the amount of noise taken into account for feature extraction and maximizing the amount of information available for matching. By proceeding this way, we privilege the regions for which our quality measure gives the highest values, namely regions of the iris that are highly textured and free from occlusion, and minimize the risks of extracting features in occluded regions to which our quality measure gives the lowest values. We also control the amount of information we use for matching by including, if necessary, regions that are given intermediate values by our quality measure and are free from occlusion but barely textured. Experiments were performed on three different databases: ND-IRIS- 0405, Casia-IrisV3-Interval and Casia-IrisV3-Twins, and show a significant improvement of recognition performance when using our strategy to select regions for feature extraction instead of using a binary segmentation mask and considering all unmasked regions equally.
  • Konferenzbeitrag
    Quality driven iris recognition improvement
    (BIOSIG 2013, 2013) Cremer, Sandra; Lemperiere, Nadege; Dorizzi, Bernadette; Garcia-Salicetti, Sonia
    The purpose of the work presented in this paper is to adapt the feature extraction and matching steps of iris recognition to the quality of the input images. To this end we define a GMM-based global quality metric associated to a pair of normalized iris images. It quantifies the amount of artifact in these images as well as the amount of texture in artifact-free regions. First we use this metric to adjust, for each pair of irises, the proportion of the normalized image selected on a local quality criteria for feature extraction. This approach is tested with two matching techniques: one performs a bit to bit comparison of binary feature vectors and the other one computes local cross-correlations between real valued vectors. We show that our approach is effective with both techniques. Then we use our metric to choose the matching technique that is best adapted to each image pair in order to make a good compromise between accuracy and speed.