Q345 is a kind of ferrite and pearlite dual phase steel that is widely used on bearing force components in architectural structures and mechanical systems,the long term dynamic loading during their service life induces the fatigue fracture under the stress amplitude that far below the tensile strength of the material,which require the study of the fatigue failure of the material.The conventional fatigue test was carried out with the help of electromagnetic resonance fatigue test machine (140 Hz) to study the fatigue failure of Q345,the stress-life (S-N) curve of Q345 was collected in high cycle fatigue regime.The initiation and propagation of the cracks of Q345 under cyclic loading was studied by the scanning electron microscope (SEM),and the intrinsic dissipation energy during the fatigue failure of the material was investigated with the help of infrared camera additionally.The fatigue failure of low carbon steel Q345 under high frequency cyclic loading was induced by the micro cracks initiated from ferrites grains.The propagation of the micro cracks was influenced by the micro structure of the material,and apt to propagate along the ferrite grains and the grain boundaries,but could be easily hampered by the presents of pearlite grains in the crack tip.The presence of pearlite grains helped retard crack propagation,which made the fatigue crack tortuous.The variation of the temperature field was not distinct until the stress amplitude was higher than the fatigue limit in the high cycle fatigue regime,therefor,the fatigue limit can be quickly determined based on the temperature variation of the specimen surface under the cyclic loading.Furthermore,a model was established in thermodynamics framework to characterize the intrinsic dissipation energy of the material under high frequency cyclic loading, and the result showed that the relation between intrinsic dissipation energy of unit volume material and limited fatigue life loading presented to be nonlinearly.

In order to solve the difficult problems of caused by mutual occlusion of face and face orientation, a PNMS algorithm based on penalty factors was proposed to improve the accuracy of face detection and alignment. Firstly, according to the overlap degree between face candidate windows and the corresponding detection scores of candidate windows, a non-continuous linear function and a continuous function based on Gaussian distribution were proposed and used as penalty factors for non-maximum suppression. Then the traditional non-maximum suppression algorithm was improved and replaced, and the detection score of the candidate window was redistributed. On this basis, combining the characteristics of the first two kinds of penalty factors, the continuous nonlinear function was further proposed as the penalty factor of the non-maximum suppression algorithm. Consequently, the greater the overlap value between windows, the more severe the penalty is, and the function is continuous throughout the overlapping value range. The proposed algorithm performed detailed face detection experiment verification on two face detection data sets of FDDB and WIDER FACE. The face alignment experiments were verified on the AFLW data set. The results showed that the proposed PNMS algorithms compared with other algorithms not only effectively improves the accuracy and reliability of face detection and alignment, but also solves a certain degree of face occlusion, and reduces the rate of detection failure of occluded faces.