Video Super-Resolution Based on Inter-Frame Information Utilization for Intelligent Transportation

Intelligent transportation infrastructure is essential to intelligent transportation system (ITS). With the continuous development of Internet of Things (IoT) technology, remote monitoring has become a critical part of ITS. However, due to the limitations of network transmission, production cost, and other factors, some video monitoring can obtain only low-resolution (LR) video. LR video features are seriously lost, thus affecting the performance of ITS. In this paper, based on the research of existing super-resolution algorithms, we focus on improving the reconstruction quality of video frame sequences by aiming at the insufficient utilization of inter-frame information and low reconstruction quality of existing video super-resolution algorithms. This paper proposes a video super-resolution algorithm based on inter-frame information utilization, which can effectively improve the performance of ITS. First, a novel U-shaped feature extractor is designed to fully extract the feature expression of video frame sequences. Second, a deformable inter-frame alignment module based on residual learning is constructed to make the inter-frame alignment more accurate and thus promote the mutual utilization of inter-frame information. Finally, an up and down sampling residual block is proposed to extract features that better match the upsampling reconstruction requirements. The experimental results show that the method has better reconstruction quality for monitoring video and is advanced and applicable compared to mainstream video oversampling methods.

I. Introduction

With the rapid development of artificial intelligence, the IoT, and other emerging industries, ITS construction has become a trend of strategic development. IoT technology is closely related to intelligent transportation infrastructure. The better the effect of remote monitoring of the IoT infrastructure, the better the management of the intelligent transportation infrastructure [1], [2]. ITS is a transportation-oriented service system based on modern electronic information technology. The outstanding feature of ITS is information collection, and thousands of video monitoring are pervasively deployed in the ITS. As one of the essential sources of video data, video capture monitoring can be seen from any corner of the road. In addition, the number of video monitoring is growing at a rate of 20% per year. In the context of pedestrian re-identification, crowd counting, recognizing activities of workers, and other upper-layer applications that continue to become more complex in terms of functionality, high-resolution (HR) video has become an essential basis for the development of ITS [3], [4], [5]. However, due to shooting equipment, network transmission, and hardware costs, people often have access only to LR playback videos. LR video contains a serious lack of detailed information, thereby making it difficult to fully extract important information contained in the video. This difficulty conflicts between the availability of LR video content and the demand for HR video in intelligent transportation infrastructure. Video super-resolution technology can reconstruct monitoring video with rich texture details and is thus crucial to providing high-quality public services, reducing costs, and achieving the continuous development of intelligent transportation infrastructure.