South China University of Technology

About: South China University of Technology is a education organization based out in Guangzhou, China. It is known for research contribution in the topics: Catalysis & Adsorption. The organization has 62343 authors who have published 69468 publications receiving 1251592 citations. The organization is also known as: SCUT & Huánán Lǐgōng Dàxué.

Copper-Catalyzed Aerobic C(sp2)–H Functionalization for C–N Bond Formation: Synthesis of Pyrazoles and Indazoles

Abstract: A simple, practical, and highly efficient synthesis of pyrazoles and indazoles via copper-catalyzed direct aerobic oxidative C(sp2)–H amination has been reported herein. This process tolerated a variety of functional groups under mild conditions. Further diversification of pyrazoles was also investigated, which provided its potential for drug discovery.

High performance offline handwritten Chinese character recognition using GoogLeNet and directional feature maps

Abstract: Just like its great success in solving many computer vision problems, the convolutional neural networks (CNN) provided new end-to-end approach to handwritten Chinese character recognition (HCCR) with very promising results in recent years. However, previous CNNs so far proposed for HCCR were neither deep enough nor slim enough. We show in this paper that, a deeper architecture can benefit HCCR a lot to achieve higher performance, meanwhile can be designed with less parameters. We also show that the traditional feature extraction methods, such as Gabor or gradient feature maps, are still useful for enhancing the performance of CNN. We design a streamlined version of GoogLeNet [13], which was original proposed for image classification in recent years with very deep architecture, for HCCR (denoted as HCCR-GoogLeNet). The HCCR-GoogLeNet we used is 19 layers deep but involves with only 7.26 million parameters. Experiments were conducted using the ICDAR 2013 offline HCCR competition dataset. It has been shown that with the proper incorporation with traditional directional feature maps, the proposed single and ensemble HCCR-GoogLeNet models achieve new state of the art recognition accuracy of 96.35% and 96.74%, respectively, outperforming previous best result with significant gap.