Weihua Liu

Institute of Crop Sciences

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Zhongxiong Lai

Fujian Agriculture and Forestry University

Chunzhen Cheng

Shanxi Agricultural University

Fanglan Chen

Fujian Agriculture and Forestry University

Shanshan Ni

Women's Hospital, School of Medicine, Zhejiang University

Yu‐Ling Lin

Zhujiang Hospital

Xinghu Yu

University of Nottingham Ningbo China

Xianqiang Yang

Harbin Institute of Technology

Hao Sun

University of Nottingham Ningbo China

Jinpeng Zhang

Institute of Crop Sciences

Yangyu Fan

Northwestern Polytechnical University

Cooperative Institutions

Chinese Academy of Sciences

Chinese Academy of Agricultural Sciences

Institute of Crop Sciences

University of Chinese Academy of Sciences

Sun Yat-sen University

Harbin Institute of Technology

Zhejiang University

Northwestern Polytechnical University

Shanghai Jiao Tong University

Peking University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Improvement of water pollution detection method based on convolution neural network

Desalination and Water Treatment (2023)

Weihua Liu Tingting Shi Xuetian Xu

In water resources pollution detection, because the collected data is not filtered, the rate of missed detection is high, and the accuracy of detection results is limited.Therefore, a method of water resources pollution detection based on convolution neural network is proposed.The data of water resource pollution is input into convolutional neural network to calculate its average fitness value, generate the optimal rule set, calculate its average fitness, train the water resource pollution detection data, and realize the accurate monitoring of water resource pollution.The experimental results show that the method has high detection accuracy, low leakage rate and high detection efficiency.

Convolution (computer science)

10.5004/dwt.2023.30047

Cite

Citations (0)

Genome-Wide Bisulphite Sequencing Uncovered the Contribution of DNA Methylation to Rice Short-Term Drought Memory Formation

Journal of Plant Growth Regulation (2021)

Shuyan Kou Gu QiongYao Liu Duan Gao-Jing Liu Pingrong Yuan

Abstract Rice is the important crop for more than half of the world population. However, drought can have a devastating impact on rice growth and reduce yield drastically. Understanding the response of rice to drought stress is especially important for crop breeding. Previously we found that rice enhanced its tolerance to drought stresses via stress memory mechanisms. Numerous memory genes were identified to play important roles in the process. DNA methylation was reported to mediate tolerance via regulating gene expression and enhances the survival rate of rice encountering drought stress. However, how DNA methylation involved in stress memory is still not clear. In this study, genome-wide bisulphite sequencing at a single base resolution methylome profiling level was performed and analyzed in rice cultivar under recurrent drought stresses and recovery treatments. We found that rice drought stress memory-related differentially methylated regions (DMRs) showed dynamic and distinct patterns. The drought-memory DMRs may regulate Transposable elements and gene expression to cope with short-term repeated drought stresses. Our findings of drought-memory DMRs can explain mechanisms of rice drought stress memory in a new perspective on global methylome details. Using epigenetic markers to breed drought-resistant rice would become a feasible way in the future research.

Differentially methylated regions

Bisulfite sequencing

10.1007/s00344-021-10483-3

Cite

Citations (36)

A new method for calibrating depth and color camera pair based on Kinect

International Conference on Audio, Language and Image Processing (2012)

Weihua Liu Yangyu Fan Zhong Zhang Tao Lei

In this paper, we present a method that can calibrates color camera and depth camera of Kinect simultaneously, and finally get the relative pose between them. The calibration process for color camera and depth camera is designed by taking advantage of two different method respectively. As we know, the depth information can be acquired readily from the depth image, however, under the consideration of the fuzzy boundary and the overlapping of object in depth image, we choose using one-dimensional object for calibrating depth camera and propose a nonlinear method to optimize its intrinsic parameter. As for color camera calibration, we design a cross shape object for calibration and validation. Both methods are strongly robust to noise and much easier to implement. The experiment result shows a better accuracy in comparison with the proprietary calibration procedure of the manufacturer.

10.1109/icalip.2012.6376614

Cite

Citations (18)

Soil physical and chemical properties in Shanghai's urban forests

Zhejiang Linxueyuan xuebao (2009)

Weihua Liu Guilian Zhang Fei Xu Yaping Wang Yu XueQin

To explore the soil physical and chemical properties in regards to the vegetation community of urban forest,and to inquire the relationship between soil and the vegetation from the urban ecology angle,soil was collected from park green spaces,road green belts,residential areas,and green space affiliated to units within the external circle highway of Shanghai City.Based on the community structure,a system of 6 forest types consists of evergreen broad-leaved forest,deciduous broadleaved forest,evergreen coniferous forest,deciduous needle-leaf forest,evergreen shrub forest and deciduous shrub forest was determined.And the polts were ascertained according to the fan-shaped grid method of landscape pattern.Physicochem-ical properties were measured at soil depths of 0-10,10-20,and 20-30 cm.Results showed that av-erage soil organic matter(SOM) above 30 cm was 15.52 g.kg-1,average total nitrogen(TN) was 3.78 g.kg-1,and average total phosphorus(TP) was 1.13 g.kg-1.SOM and TN in the surface layer were rich and gradual-ly decreased with depth,but TP did not show a gradual downward trend.Soil density was between 1.44-1.60 g.cm-3.Because average SOM,TN,and TP in the top 30 cm of urban green belt soil in Shanghai were low and compaction had increased soil density,measures are needed to improve fertility levels,espe-cially in the newly-established green belts,to enhance the nutrient content and increase the landscape effect.

Urban forest

Topsoil

Soil Compaction

Evergreen forest

Source

Cite

Citations (0)

A Novel Enhancement Algorithm for Non-Uniform Illumination Particle Image

Lecture notes in electrical engineering (2013)

Weihua Liu

Particle (ecology)

10.1007/978-94-007-7618-0_26

Cite

Citations (0)

Subpixel segmentation for ceramic defects

IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society (2021)

Weihua Liu Xianqiang Yang Yimin Wang Xuebo Yang Weiyang Lin

Because of various problems in the production of ceramics, such as material and transportation, there exists many defects in ceramics products. Traditional methods use pixel-level segmentation to locate defects which only achieve pixel-level accuracy. In order to realize higher-precision defect localization, we propose a novel subpixel interpolation method based on the blurred edge model. Based on the analysis to the gray histogram of the defect image, the threshold segmentation method is first used to roughly locate the defect. The formula for calculation of subpixel position using interpolation method is then derived, which is faster than the traditional fitting method. Finally, the 2-D subpixel edge localization problem is disassembled 1-D calculation problems in the X-direction and Y-direction according to the direction of the defect edge gradient. The subpixel location calculation method is further given. Experiments on different defects show that the proposed method can precisely segment the defects.

Subpixel rendering

Interpolation

10.1109/iecon48115.2021.9589391

Cite

Citations (0)

Subpixel Vision Measurement Method for Rectangular-Pin SMDs Based on Asymmetric Gaussian Gradient Edge Profile

IEEE Transactions on Industrial Informatics (2024)

Weihua Liu Yimin Wang Juan J. Rodríguez-Andina Xinghu Yu

Subpixel rendering

10.1109/tii.2024.3488789

Cite

Citations (0)

The analysis on the sitution and development of the resources construction in the new local university libraries

Journal of Tongren University (2010)

Weihua Liu

This paper analyses the situation on the resources construction of the collection in libraries,it starts from the situation of the resources construction of the collection in the new local university libraries,and proposes some problems on the existence of a number of solutions,meanwhile,I talked about my own views on the ideas of the special resources construction of libraries in the furture time.

Collection development

Source

Cite

Citations (0)

The effect of whitening transformation on pooling operations in convolutional autoencoders

EURASIP Journal on Advances in Signal Processing (2015)

Zuhe Li Yangyu Fan Weihua Liu

Convolutional autoencoders (CAEs) are unsupervised feature extractors for high-resolution images. In the pre-processing step, whitening transformation has widely been adopted to remove redundancy by making adjacent pixels less correlated. Pooling is a biologically inspired operation to reduce the resolution of feature maps and achieve spatial invariance in convolutional neural networks. Conventionally, pooling methods are mainly determined empirically in most previous work. Therefore, our main purpose is to study the relationship between whitening processing and pooling operations in convolutional autoencoders for image classification. We propose an adaptive pooling approach based on the concepts of information entropy to test the effect of whitening on pooling in different conditions. Experimental results on benchmark datasets indicate that the performance of pooling strategies is associated with the distribution of feature activations, which can be affected by whitening processing. This provides guidance for the selection of pooling methods in convolutional autoencoders and other convolutional neural networks.

Pooling

Benchmark (surveying)

Feature (linguistics)

10.1186/s13634-015-0222-1

Cite

Citations (26)

Fast intra mode decision algorithm in H.264/AVC using characteristics of transformed coefficients

2008 5th International Conference on Visual Information Engineering (VIE 2008) (2008)

Wang Fengqin Yangyu Fan Lan Yajing Weihua Liu

In H.264/AVC standard, intra prediction is allowed to perform with 9 or 4 candidate modes according to 4×4 or 16×16 block to achieve low bitrate and high quality. The exhaustive mode decision algorithm to choose the best mode is computationally expensive. To reduce the amount of calculations in intra prediction, a novel fast mode decision algorithm for H.264/AVC intra prediction is proposed based on the analysis of the characteristics of transform domain. By using characteristics of transformed coefficients of blocks with edge of directional distribution, only a small number of most possible intra prediction modes are chosen for RDO calculation and periodic insertion of exhaustive mode decision algorithm and adaptive threshold to prevent error diffusion. In that way, the fast mode decision algorithm helps to speed up intra coding significantly. Experimental results show that the proposed scheme performs nearly 5 times faster than the exhaustive search strategy adopted in current H.264/AVC encoder with little degradation in the coding efficiency.

Mode (computer interface)

Algorithmic efficiency

Context-adaptive variable-length coding

10.1049/cp:20080316

Cite

Citations (10)