No abstract

As the upcoming video coding standard, Versatile Video Coding (i.e., VVC) achieves up to 30% Bjøntegaard delta bit-rate (BD-rate) reduction compared with High Efficiency Video Coding (H.265/HEVC). To eliminate or alleviate different kinds of compression artifacts like blocking, ringing, blurring and contouring effects, three in-loop filters, i.e. de-blocking filter (DBF), sample adaptive offset (SAO) and adaptive loop filter (ALF), have been involved in VVC. Recently, Convolutional Neural Network (CNN) has attracted tremendous attention and shows great potential in many tasks in image processing. In this work, we design a CNN-based in-loop filter as an integrated single-model solution which is adaptive to almost any scenarios in video coding. An architecture named as ADCNN (i.e., Attention based Dual-scale CNN) with an attention based processing block is proposed to reduce artifacts of I frames and B frames, which take advantage of informative priors such as the quantization parameter (QP) and partitioning information. Different from existing CNN-based filtering methods, which are mainly designed for the luma component and may need to train different models for different QPs, the proposed filter is adapted to different QPs and different frame types, and all the components (i.e., both luma and chroma) are processed simultaneously with feature exchange and fusion between components for information supplementary. Experimental results show that the proposed ADCNN filter can achieve 6.54%, 13.27%, 15.72% BD-rate savings for Y, U, V respectively under the all intra configuration and 2.81%, 7.86%, 8.60% BD-rate savings under the random access configuration. It can be used to replace all the conventional in-loop filters and also outperforms them without increase in encoding time. INDEX TERMSCoding artifacts, convolutional neural network (CNN), in-loop filter, video coding, versatile video coding (VVC).

No abstract

This paper presents an experimental study on dynamic performance of China Railway Track System (CRTS) series track systems using a full-scale test rig. The test rig has been constructed based on 55.17 m long full-scale nonballasted tracks composed of four typical CRTS track elements in high-speed railways. First, the dynamic characteristics of different nonballasted tracks are investigated by conducting wheel-drop tests, where a wheel-drop testing vehicle with a dropping wheelset is devised to provide the wheel-drop load. The vibration levels of different track systems are assessed by the root-mean-square acceleration per one-third octave band, and the vibration transmission characteristics of the CRTS series tracks are evaluated by transfer functions. Further, a mathematical track model is used to extract the dynamic stiffness and damping coefficient of the four types of nonballasted track systems based on the wheel-rail impact response. The vibration characteristics, the dynamic stiffness, and damping coefficient of different nonballasted track systems under various wheel-drop heights are compared and discussed in detail.

Due to the fact that ballastless tracks in high-speed railways are not only subjected to repeated train–track dynamic interaction loads, but also suffer from complex environmental loads, the fundamental understanding of mechanical performance of ballastless tracks under sophisticated service conditions is an increasingly demanding and challenging issue in high-speed railway networks. This work aims to reveal the effect of train–track interaction and environment loads on the mechanical characteristic variation of ballastless tracks in high-speed railways, particularly focusing on the typical interface damage evolution between track layers. To this end, a finite element model of a double-block ballastless track involving the cohesive zone model for the track interface is first established to analyze the mechanical properties of the track interface under the loading–unloading processes of the negative temperature gradient load (TGL) followed by the same cycle of the positive TGL. Subsequently, the effect of wheel–rail longitudinal interactions on the nonlinear dynamic characteristics of the track interface is investigated by using a vehicle-slab track vertical-longitudinal coupled dynamics model. Finally, the influence of dynamic water pressure induced by vehicle dynamic load on the mechanical characteristics and damage evolution of the track interface is elucidated using a fluid–solid coupling method. Results show that the loading history of the positive and negative TGLs has a great impact on the nonlinear development and distribution of the track interface stress and damage; the interface damage could be induced by the wheel–rail longitudinal vibrations at a high vehicle running speed owing to the dynamic amplification effect caused by short wave irregularities; the vehicle dynamic load could produce considerable water pressure that presents nonlinear spatial–temporal characteristics at the track interface, which would lead to the interface failure under a certain condition due to the coupled dynamic effect of vehicle load and water pressure.