Lu Qin

A Fingerprint Identification Algorithm Based on Local Minutiae Topological Property

Minutiae is widely used in fingerprint recognition, however, single minutiae feature is still unable to cover the influences from the acquisition process. Interferes such as dirty fingers, uncertain pressing position and so on can easily affect the accuracy. In order to improve this situation, we propose a new feature model called local minutiae topological. Unlike other methods, the proposed model is based on the relative location of minutiae and core point, in which a core point of the fingerprint is firstly confirmed, and then minutiae around the core point is extracted by an improved FVS algorithm. The topological relationship is built on the extracted minutiae and the core point. Finally we adopt Neural Network to verify the proposed feature model. The experiments are based on FVC2000, the comparison results to several similar excellent algorithms show that the proposed model has high computational efficiency and a significant improvement on robustness.

Visual object tracking challenges revisited: VOT vs. OTB

Numerous benchmark datasets and evaluation toolkits have been designed to facilitate visual object tracking evaluation. However, it is not clear which evaluation protocols are preferred for different tracking objectives. Even worse, different evaluation protocols sometimes yield contradictory conclusions, further hampering reliable evaluation. Therefore, we 1) introduce the new concept of mirror tracking to measure the robustness of a tracker and identify its over-fitting scenarios; 2) measure the robustness of the evaluation ranks produced by different evaluation protocols; and 3) report a detailed analysis of milestone tracking challenges, indicating their application scenarios. Our experiments are based on two state-of-the-art challenges, namely, OTB and VOT, using the same trackers and datasets. Based on the experiments, we conclude that 1) the proposed mirror tracking metrics can identify the over-fitting scenarios of a tracker, 2) the ranks produced by OTB are more robust than those produced by VOT, and 3) the joint ranks produced by OTB and VOT can be used to measure failure recovery.