Chinese People's Public Security University

About: Chinese People's Public Security University is a education organization based out in Beijing, China. It is known for research contribution in the topics: Feature extraction & Deep learning. The organization has 1006 authors who have published 757 publications receiving 4740 citations.

Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression

Abstract: Bounding box regression is the crucial step in object detection. In existing methods, while ln-norm loss is widely adopted for bounding box regression, it is not tailored to the evaluation metric, i.e., Intersection over Union (IoU). Recently, IoU loss and generalized IoU (GIoU) loss have been proposed to benefit the IoU metric, but still suffer from the problems of slow convergence and inaccurate regression. In this paper, we propose a Distance-IoU (DIoU) loss by incorporating the normalized distance between the predicted box and the target box, which converges much faster in training than IoU and GIoU losses. Furthermore, this paper summarizes three geometric factors in bounding box regression, i.e., overlap area, central point distance and aspect ratio, based on which a Complete IoU (CIoU) loss is proposed, thereby leading to faster convergence and better performance. By incorporating DIoU and CIoU losses into state-of-the-art object detection algorithms, e.g., YOLO v3, SSD and Faster R-CNN, we achieve notable performance gains in terms of not only IoU metric but also GIoU metric. Moreover, DIoU can be easily adopted into non-maximum suppression (NMS) to act as the criterion, further boosting performance improvement. The source code and trained models are available at https://github.com/Zzh-tju/DIoU.

Constructing road safety performance indicators using Fuzzy Delphi Method and Grey Delphi Method

Abstract: The main goal of this paper is to construct three sets of road safety performance indicators, which are regional road safety performance indicators, urban road safety performance indicators and highway safety performance indicators, respectively. Fuzzy Delphi Method and Grey Delphi Method are applied to quantify experts' attitudes to regional road safety, urban road safety and highway safety. Comparing the results of two methods, the different results of two methods are analyzed, and then the final safety performance indicators are obtained by taking the intersection of results of two methods. Finally, three sets of performance indicators are constructed, which can be described and evaluated the safety level of region, urban road and highway, respectively. The research findings show that the method used in this paper is feasible and practical and can be provided as a reference for the administrative authority of road safety.

Size Effects on Properties of NiO Nanoparticles Grown in Alkalisalts

Abstract: NiO nanoparticles with sizes of 3.5–12.4 nm were grown by thermal decomposing of nickel acetate at different temperatures in NaCl and Li2CO3 alkalisalts. The properties of the nanoparticles were characterized by X-ray diffraction spectrometer, transmission electron microscope, absorption spectrometer, micro-Raman microscope, and superconducting quantum interference device. The effects of the nanoparticle sizes on the crystal structure, exciton ground state energy, vibration modes, and magnetic properties were studied. Lattice parameter of NiO increases with a decrease in nanoparticle sizes. The band gap of NiO nanoparticles increases with a decrease in the nanoparticle size. LO modes of NiO nanoparticles shift red, and the intensity increases with a decrease in the nanoparticle sizes. Surface phonon modes are observed. Bifurcation temperature and blocking temperature of NiO nanoparticles shift to lower temperature with a decrease in nanoparticle sizes. Two peaks are present in all nanoparticles’ zero-fiel...

Oil accumulation mechanisms of the oleaginous microalga Chlorella protothecoides revealed through its genome, transcriptomes, and proteomes.

Abstract: Microalgae-derived biodiesel is a promising substitute for conventional fossil fuels. In particular, the green alga Chlorella protothecoides sp. 0710 is regarded as one of the best candidates for commercial manufacture of microalgae-derived biofuel. This is due not only to its ability to live autotrophically through photosynthesis, but also to its capacity to produce a large amount of biomass and lipid through fermentation of glucose. However, until the present study, neither its genome sequence nor the platform required for molecular manipulations were available. We generated a draft genome for C. protothecoides, and compared its genome size and gene content with that of Chlorella variabilis NC64A and Coccomyxa subellipsoidea C-169. This comparison revealed that C. protothecoides has a reduced genome size of 22.9 Mbp, about half that of its close relatives. The C. protothecoides genome encodes a smaller number of genes, fewer multi-copy genes, fewer unique genes, and fewer genome rearrangements compared with its close relatives. In addition, three Chlorella-specific hexose-proton symporter (HUP)-like genes were identified that enable the consumption of glucose and, consequently, heterotrophic growth. Furthermore, through comparative transcriptomic and proteomic studies, we generated a global perspective regarding the changes in metabolic pathways under autotrophic and heterotrophic growth conditions. Under heterotrophic conditions, enzymes involved in photosynthesis and CO2 fixation were almost completely degraded, either as mRNAs or as proteins. Meanwhile, the cells were not only capable of quickly assimilating glucose but also showed accelerated glucose catabolism through the upregulation of glycolysis and the tricarboxylic acid (TCA) cycle. Moreover, the rapid synthesis of pyruvate, upregulation of most enzymes involved in fatty acid synthesis, and downregulation of enzymes involved in fatty acid degradation favor the synthesis of fatty acids within the cell. Despite similarities to other Chlorella, C. protothecoides has a smaller genome than its close relatives. Genes involved in glucose utilization were identified, and these genes explained its ability to grow heterotrophically. Transcriptomic and proteomic results provided insight into its extraordinary ability to accumulate large amounts of lipid. The C. protothecoides draft genome will promote the use of this species as a research model.

Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression.

Abstract: Bounding box regression is the crucial step in object detection. In existing methods, while $\ell_n$-norm loss is widely adopted for bounding box regression, it is not tailored to the evaluation metric, i.e., Intersection over Union (IoU). Recently, IoU loss and generalized IoU (GIoU) loss have been proposed to benefit the IoU metric, but still suffer from the problems of slow convergence and inaccurate regression. In this paper, we propose a Distance-IoU (DIoU) loss by incorporating the normalized distance between the predicted box and the target box, which converges much faster in training than IoU and GIoU losses. Furthermore, this paper summarizes three geometric factors in bounding box regression, \ie, overlap area, central point distance and aspect ratio, based on which a Complete IoU (CIoU) loss is proposed, thereby leading to faster convergence and better performance. By incorporating DIoU and CIoU losses into state-of-the-art object detection algorithms, e.g., YOLO v3, SSD and Faster RCNN, we achieve notable performance gains in terms of not only IoU metric but also GIoU metric. Moreover, DIoU can be easily adopted into non-maximum suppression (NMS) to act as the criterion, further boosting performance improvement. The source code and trained models are available at this https URL.