Robust Face Recognition Based on a New Supervised Kernel Subspace Learning Method.

Face recognition is one of the most popular techniques to achieve the goal of figuring out the identity of a person. This study has been conducted to develop a new non-linear subspace learning method named "supervised kernel locality-based discriminant neighborhood embedding," which performs data classification by learning an optimum embedded subspace from a principal high dimensional space. In this approach, not only nonlinear and complex variation of face images is effectively represented using nonlinear kernel mapping, but local structure information of data from the same class and discriminant information from distinct classes are also simultaneously preserved to further improve final classification performance. Moreover, in order to evaluate the robustness of the proposed method, it was compared with several well-known pattern recognition methods through comprehensive experiments with six publicly accessible datasets. Experiment results reveal that our method consistently outperforms its competitors, which demonstrates strong potential to be implemented in many real-world systems.

Optimization of an Optical Test Bench for Tire Properties Measurement and Tread Defects Characterization.

Tire characteristics and behavior are of great importance in vehicle dynamics since the forces transmitted in the tire-road contact are the main contributors to global vehicle performance. Several research groups have focused on the study and modeling of tires. Some of the most important factors that need to be known are tread characteristics and pressure distribution in the tire-ground contact patch. In this work, a test bench has been used to adequately determine the aforementioned factors. The measurement principle of the test bench is the frustration of total internal reflection (FTIR) of light. It makes use of a laterally illuminated glass on which the tire leans. An interposed plastic interface between them causes the reflection of light. Finally, a video camera captures the bright image formed through the glass. The brightness level in each pixel of the image is related to existing normal pressure. A study of the parameters that affect the test bench calibration such as type of interface material used, diffuse light, hysteresis, creep and transverse light absorption is performed. Experimental tests are conducted to relate tire inflation pressure and camber angle to the pressure distribution. Furthermore, the test bench is used to detect and evaluate the influence of defects in the tire on the contact pressures.

Robust Road Condition Detection System Using In-Vehicle Standard Sensors.

The appearance of active safety systems, such as Anti-lock Braking System, Traction Control System, Stability Control System, etc., represents a major evolution in road safety. In the automotive sector, the term vehicle active safety systems refers to those whose goal is to help avoid a crash or to reduce the risk of having an accident. These systems safeguard us, being in continuous evolution and incorporating new capabilities continuously. In order for these systems and vehicles to work adequately, they need to know some fundamental information: the road condition on which the vehicle is circulating. This early road detection is intended to allow vehicle control systems to act faster and more suitably, thus obtaining a substantial advantage. In this work, we try to detect the road condition the vehicle is being driven on, using the standard sensors installed in commercial vehicles. Vehicle models were programmed in on-board systems to perform real-time estimations of the forces of contact between the wheel and road and the speed of the vehicle. Subsequently, a fuzzy logic block is used to obtain an index representing the road condition. Finally, an artificial neural network was used to provide the optimal slip for each surface. Simulations and experiments verified the proposed method.