Kun Zhang

China Southern Power Grid (China)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Bingyu Liu

Jiangsu University of Science and Technology

Cuirong Wang

Xinjiang Institute of Ecology and Geography

Yanan Wang

National Supercomputing Center in Wuxi

Wei Hao

NOAA Center for Satellite Applications and Research

Wei Fan

University of Oxford

Haimin Wang

New Jersey Institute of Technology

Qinjun Zhao

University of Jinan

Zhiyong Li

Alibaba Group (China)

Yaotang Lv

China Southern Power Grid (China)

Kai Chen

University of Hong Kong

Cooperative Institutions

Chinese Academy of Sciences

University of Jinan

Hainan University

China Southern Power Grid (China)

Hebei University of Science and Technology

Shandong University of Science and Technology

Shandong University

Nanjing University of Science and Technology

Huazhong University of Science and Technology

Beihang University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Semi-Supervised Learning-Based Image Denoising for Big Data

IEEE Access (2020)

Kun Zhang Kai Chen

In this paper, the research of image noise reduction based on semi-supervised learning is carried out, and the neural network is used to reduce the noise of the image, so as to achieve more stable and good image display ability. Based on the convolutional neural network algorithm, the role of activation function optimization network is studied, combined with semi-supervised learning modes such as multi-feature extraction technology, to learn and extract the key features of the input image. Semi-supervised residual learning based on convolutional network is a good image denoising and denoising network model. Compared with other excellent denoising algorithms, it has very good results. At the same time, it greatly improves the image noise pollution and makes the image details clearer. At the same time, compared with other image denoising algorithms, this algorithm can show a good peak signal-to-noise ratio under various noise standard deviations. Through the research in this article, it is verified that the improved convolutional neural network denoising model and multi-feature extraction technology have strong advantages in image denoising.

Video denoising

Supervised Learning

Feature (linguistics)

Non-Local Means

10.1109/access.2020.3025324

Cite

Citations (0)

A Lightweight Convolutional Network for Visual Tracking

Frontiers in artificial intelligence and applications (2024)

Jian Wang Yaotang Lv Kun Zhang Yaotang Lv Xiaogang Wu

Regression trackers have been shown to perform superiorly in visual tracking. However, existing researches in regression trackers mainly explore deep models for feature extraction, and then use sophisticated architectures for online detection. Such systems should optimize a massive number of trainable parameters. In this paper, we present a simple yet effective visual tracking system, called LiteCNT. Our algorithm only consists of three convolutional layers for the whole tracking process. In addition, a multi-region convolutional operator is introduced for regression output. This idea is simple but powerful as it enables our tracker to capture more details of target object. We further derive an efficient and effective operator to approximate multi-region aggregation.

BitTorrent tracker

Operator (biology)

Tracking (education)

Feature (linguistics)

10.3233/faia240139

Cite

Citations (0)

Research on Localization and Path Planning of Indoor Robot Based on ROS

2021 International Conference on Computer Information Science and Artificial Intelligence (CISAI) (2021)

Xiaohong Wang XuePei Ma ShengLan Zhu Kun Zhang Yanan Sun -

To meet the needs of monitoring the indoor environment, a mobile robot that patrols indoors is designed. The mobile robot uses Raspberry Pi equipped with a ROS system as the control core, uses lidar to collect environmental information, and uses various function packages to develop algorithms under the ROS distributed framework. It is based on the Karto algorithm to build maps and uses the A* algorithm and DWA algorithm to realize autonomous navigation. The robot uses a two-wheel independent drive robot chassis to cope with the complex environment. After the indoor environment test, the goal of indoor map construction and path planning can be completed, which is of great significance to the research in the indoor environment in the future.

Chassis

Mobile Robot Navigation

10.1109/cisai54367.2021.00159

Cite

Citations (2)

Dynamic fault-tolerant routing based on link state in fat tree

WIT transactions on information and communication technologies (2013)

Bingyu Liu Cuirong Wang Kun Zhang

Link (geometry)

Tree (set theory)

10.2495/isme20131522

Cite

Citations (0)

The Verification Method of Low Voltage Station Topology Recognition Based on DTW Algorithm

2022 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA) (2021)

Weiting Zhu Ye Tian Haiping Yu Yitao Liu Yanan Wang

Aiming at the problems of high cost of manual investigation, low accuracy and difficult side-to-end information interaction in the verification of the relationship between distribution transformer terminal and household transformer in low-voltage substation area, a low-voltage substation area topology identification verification method based on DTW algorithm is proposed. According to the time series voltage data of smart meter and distribution transformer terminal, this paper uses DTW algorithm to calculate the similarity of time and voltage series, and checks the user by setting threshold. Then KNN algorithm is used to determine the station area, and the household transformer relationship is determined. Finally, according to the time series voltage data collected by the meter, the DTW algorithm is used to calculate the voltage sequence similarity between users, and the feeder position of users is located combined with the topology. Experiments show that the method proposed in this paper has the characteristics of high accuracy and high efficiency, and it can accurately verify the relationship between household and transformer of distribution transformer terminal equipment in low voltage substation area.

10.1109/aeeca52519.2021.9574242

Cite

Citations (4)

Construction of Intangible Cultural Heritage Spot Based on AR Technology—Taking the Intangible Cultural Heritage of the Li Nationality in the Areca Valley as An Example

IOP Conference Series Earth and Environmental Science (2019)

Yue Lang Xi Deng Kun Zhang Yiwen Wang

The tourist area is a display area mainly for tourism and cultural visit. With the continuous advancement of the digitization process of scenic spots in China, a number of information software products showing the digital achievements of science and technology have been emerging continuously, and the technology of virtual and real fusion has become the main technical means at present. AR technology is adopted to display the intangible cultural heritage of the scenic spot. This paper analyses the advantages and disadvantages of existing scenic spot construction and analyzes the advantages of AR technology in the construction of scenic spot. Taking penang valley li intangible cultural heritage scenic spot as an example, this paper expounds the display mode of AR technology in the scenic spot. The visit and tour of the scenic area based on AR technology brings real visual stereo effect, mobile convenience and interest to hands, and also provides beneficial reference for the wide application and rapid improvement of AR technology, as well as digital innovation and rapid popularization of the historical background of intangible cultural heritage.

Intangible cultural heritage

Digitization

10.1088/1755-1315/234/1/012119

Cite

Citations (10)

Penetration State Recognition for GMAW Process Using Stereo Vision Based on Deep Learning

2022 IEEE 18th International Conference on Automation Science and Engineering (CASE) (2023)

Kun Zhang Chuangqi Yue Zhimin Liang

As a major arc welding process, pulsed gas metal arc welding (GMAW-P) is extensively applied in the production of various critical components due to its benefits such as high weld strength and ease of automation. The geometry of the weld pool contains a lot of welding quality characteristics in the process of welding, which is an important basis for judging the penetration states of the weld pool. For such key features, we develop a single camera stereo vision system based on a biprism and propose a deep learning-based penetration states recognition strategy. Using a stereo matching network based on attention concatenation volume, filter redundant information in concatenation volumes by generating attention weights to predict the accurate disparity map, and compare experimental results to evaluate its validity in weld pool disparity estimation. Secondly, to train the penetration states recognition model, a depth-wise separable convolutions neural network is built. This model directly maps the weld pool penetration relationship to disparity without any preprocessing and extracts deeper features such as pool contours and depth of fusion, then accurately identify the penetration states. According to the results of the experiments, based on disparity characteristics, penetration states classification accuracy is 99.83%.

Weld pool

10.1109/case56687.2023.10260610

Cite

Citations (1)

Design and exploration of neural network microsystem based on SiP

SN Applied Sciences (2021)

Hao Lv Shengbing Zhang Bao Deng Wang Jia Desheng Jing

Abstract In recent years, microelectronics technology has entered the era of nanoelectronics/integrated microsystems. System in package (SiP) and system on chip (SoC) are two important technical approaches for the realization of microsystems. Deep learning technology based on neural networks is used in graphics and images. Computer vision and target recognition are widely used. The deep learning technology of convolutional neural network is an important research field in the miniaturization and miniaturization of embedded platforms. How to combine the lightweight neural network with the microsystem to achieve the optimal balance of performance, size, and power consumption is a difficult point. This article introduces a micro-system implementation scheme that combines SiP technology and FPGA-based convolutional neural network. It uses Zynq SoC and FLASH and DDR3 memory as the main components, and uses SiP high-density system packaging technology to integrate. PL end (FPGA) design Convolutional Neural Network, convolutional neural network accelerator, adopt the method of convolution multi-dimensional division and cyclic block to design the accelerator structure, design multiple multiplication and addition parallel computing units to provide the computing power of the system. Improving and accelerating perform on the YOLOv2_Tiny model. The test uses the COCO data set as the training and test samples. The microsystem can accurately identify the target. The volume is only 30 × 30 × 1.2 mm. The performance reaches 22.09GOPs and the power consumption is only 0.81 W under the working frequency of 150 MHz. Multi-objective balance (performance, size and power consumption) of lightweight neural network Microsystems has realized.

Microsystem

10.1007/s42452-021-04766-3

Cite

Citations (3)

Bagging

Encyclopedia of Database Systems (2016)

Wei Fan Kun Zhang

10.1007/978-1-4899-7993-3_567-2

Cite

Citations (0)

Research on the Verification Method of Capillary Viscometer Based on Connected Domain

Rongyao Jing Kun Zhang Qinjun Zhao Tao Shen Kehua Du

In order to solve the problem that the dust in the insulation cabinet is mistakenly identified as the liquid level in the process of automatic verification of the capillary viscometer, this paper studied the verification method of the capillary viscometer based on connected domain. On the basis of the common automatic verification system of capillary viscometer based on computer vision, the dust recognition method based on connected domain is added. Industrial cameras are used to acquire viscometer video images in real-time. In order to make the images clearer and improve the processing speed of the images, the following preprocessing is carried out on the acquired images first, including the ROI region selection, frame difference method to capture moving targets, binarization, corrosion, and expansion. Then, the preprocessed image is marked with connected domain. The parameter difference of the connected domain combining liquid level and dust, the problem of misidentified dust as liquid level is solved. The experimental results show that the time repeatability and constant reproducibility of the verification results of the proposed method are better than those of the ordinary verification method based on computer vision, which reduces the probability of misidentified dust as the liquid level and improves the accuracy and efficiency of the verification of capillary viscometer.

Repeatability

10.1007/978-3-031-50580-5_24

Cite

Citations (0)