Tongji University

About: Tongji University is a education organization based out in Shanghai, China. It is known for research contribution in the topics: Computer science & Population. The organization has 76116 authors who have published 81176 publications receiving 1248911 citations. The organization is also known as: Tongji & Tóngjì Dàxué.

Efficient and Accurate Arbitrary-Shaped Text Detection With Pixel Aggregation Network

Abstract: Scene text detection, an important step of scene text reading systems, has witnessed rapid development with convolutional neural networks. Nonetheless, two main challenges still exist and hamper its deployment to real-world applications. The first problem is the trade-off between speed and accuracy. The second one is to model the arbitrary-shaped text instance. Recently, some methods have been proposed to tackle arbitrary-shaped text detection, but they rarely take the speed of the entire pipeline into consideration, which may fall short in practical applications. In this paper, we propose an efficient and accurate arbitrary-shaped text detector, termed Pixel Aggregation Network (PAN), which is equipped with a low computational-cost segmentation head and a learnable post-processing. More specifically, the segmentation head is made up of Feature Pyramid Enhancement Module (FPEM) and Feature Fusion Module (FFM). FPEM is a cascadable U-shaped module, which can introduce multi-level information to guide the better segmentation. FFM can gather the features given by the FPEMs of different depths into a final feature for segmentation. The learnable post-processing is implemented by Pixel Aggregation (PA), which can precisely aggregate text pixels by predicted similarity vectors. Experiments on several standard benchmarks validate the superiority of the proposed PAN. It is worth noting that our method can achieve a competitive F-measure of 79.9% at 84.2 FPS on CTW1500.

Constructing Ultrahigh-Capacity Zinc-Nickel-Cobalt Oxide@Ni(OH)2 Core-Shell Nanowire Arrays for High-Performance Coaxial Fiber-Shaped Asymmetric Supercapacitors.

Abstract: Increased efforts have recently been devoted to developing high-energy-density flexible supercapacitors for their practical applications in portable and wearable electronics. Although high operating voltages have been achieved in fiber-shaped asymmetric supercapacitors (FASCs), low specific capacitance still restricts the further enhancement of their energy density. This article specifies a facile and cost-effective method to directly grow three-dimensionally well-aligned zinc–nickel-cobalt oxide (ZNCO)@Ni(OH)2 nanowire arrays (NWAs) on a carbon nanotube fiber (CNTF) with an ultrahigh specific capacitance of 2847.5 F/cm3 (10.678 F/cm2) at a current density of 1 mA/cm2, These levels are approximately five times higher than those of ZNCO NWAs/CNTF electrodes (2.10 F/cm2) and four times higher than Ni(OH)2/CNTF electrodes (2.55 F/cm2). Benefiting from their unique features, we successfully fabricated a prototype coaxial FASC (CFASC) with a maximum operating voltage of 1.6 V, which was assembled by adopting Z...

The optimal cordon-based network congestion pricing problem

Abstract: This paper investigates the cordon-based second-best congestion-pricing problems on road networks, including optimal selection of both toll levels and toll locations. A road network is viewed as a directed graph and the cutset concept in graph theory is used to describe the mathematical properties of a toll cordon by examining the incidence matrix of the network. Maximization of social welfare is sought subject to the elastic-demand traffic equilibrium constraint. A mathematical programming model with mixed (integer and continuous) variables is formulated and solved by a combined use of a binary genetic algorithm and a grid search method for simultaneous determination of the toll levels and cordon locations on the networks. The model and algorithm are demonstrated with a numerical example.