Songbin Zhou

Guangdong Institute of Intelligent Manufacturing

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Yisen Liu

Guangdong Institute of Intelligent Manufacturing

Guixiong Liu

South China University of Technology

Chang Li

Huazhong University of Science and Technology

Zefan Qiu

Guangdong Institute of Intelligent Manufacturing

Chang Li

Hefei University of Technology

Kunkun Pang

Guangdong Institute of Intelligent Manufacturing

Wei Han

Qingdao University

Zhiyong Wan

Guangdong Institute of Intelligent Manufacturing

Hongmin Wu

Guangdong Institute of Intelligent Manufacturing

Hong Chen

Zhongda Hospital Southeast University

Cooperative Institutions

Guangdong Institute of Intelligent Manufacturing

Guangdong Academy of Sciences

Chinese Academy of Medical Sciences & Peking Union Medical College

Huazhong University of Science and Technology

Tsinghua University

Key Laboratory of Guangdong Province

South China University of Technology

Union Hospital

Chinese Academy of Sciences

Central South University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Blind Source Separation for a Robust Audio Recognition Scheme in Multiple Sound-Sources Environment

Wei Han Songbin Zhou Chang Li Yisen Liu Zhe Liu

In multiplesound-sources environment, robustness is a major challenge for audio recognition system based on audio fingerprinting, because mixed audio signals may make recognition rate has a significant decline.This paper proposes a novel audio fingerprinting method, which uses blind source separation to divide mixed audio signals into independent components and each is close to its original sound-source, then the classical Philips scheme can perform accurately identifying.Experimental results show that novel scheme is quite robust in noisy conditions where uncertain audio signals mixed by various numbers of sound-source, even though the feature of each original sound-source and their mixed model are unknown.

Source Separation

Separation (statistics)

10.2991/meic-15.2015.358

Cite

Citations (2)

A Self-Supervised Anomaly Detector of Fruits Based on Hyperspectral Imaging

Foods (2023)

Yisen Liu Songbin Zhou Zhiyong Wan Zefan Qiu Lulu Zhao

Hyperspectral imaging combined with chemometric approaches is proven to be a powerful tool for the quality evaluation and control of fruits. In fruit defect-detection scenarios, developing an unsupervised anomaly detection framework is vital, as defect sample preparation is labor-intensive and time-consuming, especially for exploring potential defects. In this paper, a spectral-spatial, information-based, self-supervised anomaly detection (SSAD) approach is proposed. During training, an auxiliary classifier is proposed to identify the projection axes of principal component (PC) images that were transformed from the hyperspectral data cubes. In test time, the fully connected layer of the learned classifier was used as a 'spectral-spatial' feature extractor, and the feature similarity metric was adopted as the score function for the downstream anomaly evaluation task. The proposed network was evaluated with two fruit data sets: a strawberry data set with bruised, infected, chilling-injured, and contaminated test samples and a blueberry data set with bruised, infected, chilling-injured, and wrinkled samples as anomalies. The results show that the SSAD yielded the best anomaly detection performance (AUC = 0.923 on average) over the baseline methods, and the visualization results further confirmed its advantage in extracting effective 'spectral-spatial' latent representation. Moreover, the robustness of SSAD is verified with the data pollution experiment; it performed significantly better than the baselines when a portion of anomalous samples was involved in the training process.

Robustness

Extractor

10.3390/foods12142669

Cite

Citations (4)

[Hyperspectral imaging technology distinguishes between Puerariae Lobatae Radix and Puerariae Lobatae Caulis].

PubMed (2023)

Lei Zhang Yuping Zhao Kunkun Pang Songbin Zhou Yisen Liu

Puerariae Lobatae Radix, the dried root of Pueraria lobata, is a traditional Chinese medicine with a long history. Puerariae Lobatae Caulis as an adulterant is always mixed into Puerariae Lobatae Radix for sales in the market. This study employed hyperspectral imaging(HSI) to distinguish between the two products. VNIR lens(spectral scope of 410-990 nm) and SWIR lens(spectral scope of 950-2 500 nm) were used for image acquiring. Multi-layer perceptron(MLP), partial least squares discriminant analysis(PLS-DA), and support vector machine(SVM) were employed to establish the full-waveband models and select the effective wavelengths for the distinguishing between Puerariae Lobatae Caulis and Puerariae Lobatae Radix, which provided technical and data support for the development of quick inspection equipment based on HSI. The results showed that MLP model outperformed PLS-DA and SVM models in the accuracy of discrimination with full wavebands in VNIR, SWIR, and VNIR+SWIR lens, which were 95.26%, 99.11%, and 99.05%, respectively. The discriminative band selection(DBS) algorithm was employed to select the effective wavelengths, and the discrimination accuracy was 93.05%, 98.05%, and 98.74% in the three different spectral scopes, respectively. On this basis, the MLP model combined with the effective wavelengths within the range of 2 100-2 400 nm can achieve the accuracy of 97.74%, which was close to that obtained with the full waveband. This waveband can be used to develop quick inspection devices based on HSI for the rapid and non-destructive distinguishing between Puerariae Lobatae Radix and Puerariae Lobatae Caulis.

Radix (gastropod)

VNIR

10.19540/j.cnki.cjcmm.20230515.103

Cite

Citations (0)

Regional Country Information Service Platform Based on Hybrid Network

Springer eBooks (2008)

Songbin Zhou Guixiong Liu Cheng Tao-bo

Informatization

Lagging

10.1007/978-0-387-77251-6_3

Cite

Citations (2)

Hyperspectral classification using deep fusion spectral–spatial features

Journal of Applied Remote Sensing (2019)

Yisen Liu Songbin Zhou Wei Han Chang Li Weixin Liu

The deep convolution neural network (CNN) has been of great interest in hyperspectral image classification recently. Current CNN-based approaches adopt a two- or three-dimensional convolution network to extract spectral–spatial features from local pixel neighborhoods. High computation cost and overfitting problems should be handled carefully for these large-scale networks. A compact one-dimensional (1-D) CNN-based method is proposed to explore the joint spectral–spatial features by employing the pixel coordinates as the spatial information. Furthermore, two kinds of CNN architecture for spectral–spatial feature fusion are compared. To be specific, the shallow fusion model (SF-CNN) concatenates the spatial information to the spectral features before they are fed into the final fully connected layer, whereas the deep fusion model (DF-CNN) combines the spectral and the spatial information in an early stage and extracts the high-level spectral-spatial features by the 1-D convolution layers. The experimental results demonstrate that the DF-CNN method provides competitive classification results to state-of-the-art methods. Moreover, attributed to the concise spatial information and the effective feature fusion structure, the proposed method is economical in terms of computation cost when compared with the current CNN-based spectral-spatial methods.

Overfitting

Convolution (computer science)

Feature (linguistics)

10.1117/1.jrs.13.038505

Cite

Citations (2)

Target localization based on LSSVR in wireless sensor networks

Optics and Precision Engineering (2010)

Songbin Zhou

In consideration of the direct influence of Received Signal Strength Indicator(RSSI)fluctuation on the target localization accuracy in wireless sensor networks(WSN),the basic principle of target localization using Least Square Support Vector Regression(LSSVR)is discussed.Then,the characteristics of LSSVR modeling are analyzed for given and variable detection sensors,respectively.Furthermore,a method for Target Localization based on Adaptive LSSVR Modeling(TL-AML)in WSN is proposed.By considering localization accuracy and real-time performance comprehensively,LSSVR models are built for locating target at initial time,and at follow-up time it is used to decide whether new models need to be built or not according to the inclusion relation between detection nodes and sensor nodes.The performance of TL-AML is verified based on CC2430WSN experiment platform.Results show that the Root Mean Square Error(RMSE)of target localization based on TLAML has reduced by 34%～37%and 60%～65%as compared with those of MLE and LSE,respectively.When modeling parameters are taken in reasonable value ranges,the localization accuracy of TL-AML is improved evidently compared with MLE and LSE.Moreover,the consuming time of TLAML is 0.2～0.4s,If LSSVR modeling is needed.Otherwise,the consuming time is only about 0.04 s.The results indicate that TL-AML method can weaken the influence of RSSI fluctuation on the accuracy of target localization and has good real-time target localization accuracy.

Signal strength

Root mean square

Source

Cite

Citations (4)

WVSN video coding method based on low complexity self-adaptive frame

China Measurement & Test (2013)

Songbin Zhou

For resolving the problem of constrained resources of node in WVSN and high complexity video coding,a video coding method based on low complexity self-adaptive frame was proposed.The difference value is calculated based on interlaced-diagonal method,and the type mechanism of the self-adaptive frame is established to avoid judging each frame to decrease the computation complexity.And intra-frame image coding is based on polygonal DCT pruning JPEG to decrease the complexity of 2D DCT.Simulation results indicate that on the premise of video quality,the proposed method can effectively decrease the complexity of video coding.Compared to JPEG without DCT pruning,the computation amount of DCT in current-frame and difference frame are decreased by 9.4% and 56.3% respectively.Compared to M-JPEG,the average image size decreased by 55.0% while the average PSNR decreased by 3.1%.

JPEG

Lossless JPEG

Residual frame

Source

Cite

Citations (0)

Comparison of Performance of Electrodeposited Lead-Silver Anode and Cast Lead-Silver Alloy Anode

Cailiao baohu (2013)

Songbin Zhou

Pb-0. 8% Ag anode was prepared on the surface of Al substrate by electrodeposition. In the meantime,conventional Pb0. 8% Ag alloy anode for a comparative study was also prepared by casting. The electrochemical properties,microstructure and morphology of as-prepared Pb-Ag anodes were comparatively investigated based on measurements of galvanostatic polarization curves,cyclic voltammetry curves and electrochemical impedance spectra as well as X-ray diffraction analysis and scanning electron microscopic observation. Results showed that,as compared with conventional cast Pb-0. 8% Ag alloy anode,electrodeposited Pb 0. 8% Ag alloy anode was more compact and uniform,and contained a higher content of PbO 2,while it possessed better corrosion resistance and electric conductivity as well.

Source

Cite

Citations (0)

Least-squares support vector machine and successive projection algorithm for quantitative analysis of cotton-polyester textile by near infrared spectroscopy

Journal of Near Infrared Spectroscopy (2018)

Yisen Liu Songbin Zhou Weixin Liu Xinhui Yang Jun Luo

The application of near infrared spectroscopy for quantitative analysis of cotton-polyester textile was investigated in the present work. A total of 214 cotton-polyester fabric samples, covering the range from 0% to 100% cotton were measured and analyzed. Partial least squares and least-squares support vector machine models with all variables as input data were established. Furthermore, successive projection algorithm was used to select effective wavelengths and establish the successive projection algorithm-least-squares support vector machine models, with the comparison of two other effective wavelength selection methods: loading weights analysis and regression coefficient analysis. The calibration and validation results show that the successive projection algorithm-least-squares support vector machine model outperformed not only the partial least squares and least-squares support vector machine models with all variables as inputs, but also the least-squares support vector machine models with loading weights analysis and regression coefficient analysis effective wavelength selection. The root mean squared error of calibration and root mean squared error of prediction values of the successive projection algorithm-least-squares support vector machine regression model with the optimal performance were 0.77% and 1.17%, respectively. The overall results demonstrated that near infrared spectroscopy combined with least-squares support vector machine and successive projection algorithm could provide a simple, rapid, economical and non-destructive method for determining the composition of cotton-polyester textiles.

Least-squares function approximation

10.1177/0967033518757069

Cite

Citations (25)

Detection of Adulteration in Infant Formula Based on Ensemble Convolutional Neural Network and Near-Infrared Spectroscopy

Foods (2021)

Yisen Liu Songbin Zhou Wei Han Chang Li Weixin Liu

Adulteration in dairy products has received world-wide attention, and at the same time, near infrared (NIR) spectroscopy has proven to be a promising tool for adulteration detection given its advantages of real-time response and non-destructive analysis. Regardless, the accurate and robust NIR model for adulteration detection is hard to achieve in practice. Convolutional neural network (CNN), as a promising deep learning architecture, is difficult to apply to such chemometrics tasks due to the high risk of overfitting, despite the breakthroughs it has made in other fields. In this paper, the ensemble learning method based on CNN estimators was developed to address the overfitting and random initialization problems of CNN and applied to the determination of two infant formula adulterants, namely hydrolyzed leather protein (HLP) and melamine. Moreover, a probabilistic wavelength selection method based on the attention mechanism was proposed for the purpose of finding the best trade-off between the accuracy and the diversity of the sub-models in ensemble learning. The overall results demonstrate that the proposed method yielded superiority regression performance over the comparison methods for both studied data sets, and determination coefficients (R2) of 0.961 and 0.995 were obtained for the HLP and the melamine data sets, respectively.

Overfitting

Ensemble Learning

Initialization

10.3390/foods10040785

Cite

Citations (22)