Prof. Mingyi He Profile

Prof. Mingyi He

at Northwestern Polytechnical Univ

SPIE Involvement:

Author

SPIE Journal Paper | 1 December 2010

Band regrouping-based lossless compression of hyperspectral images

Mingyi He, Lin Bai, Yuchao Dai, Jing Zhang

JARS, Vol. 4, Issue 01, 041757, (December 2010) https://doi.org/10.1117/12.10.1117/1.3530875

KEYWORDS: Hyperspectral imaging, Image compression, 3D image processing, Data modeling, Remote sensing, Data processing, Environmental sensing, Quantization, 3D modeling, Environmental monitoring

Read Abstract +

Proceedings Article | 24 August 2010 Paper

Hyperspectral image compression algorithm using wavelet transform and independent component analysis

Mingyi He, Lin Bai, Fatima Syeda Narjis

Proceedings Volume 7810, 781005 (2010) https://doi.org/10.1117/12.863149

KEYWORDS: Independent component analysis, Image compression, Hyperspectral imaging, Discrete wavelet transforms, Principal component analysis, Quantization, Signal to noise ratio, Image segmentation, Data analysis, Image classification

Read Abstract +

Proceedings Article | 24 August 2010 Paper

Remote sensing image restoration based on compressive sensing and two-step iteration shrinkage algorithm

Mingyi He, Weihua Liu, Lin Bai

Proceedings Volume 7810, 78100E (2010) https://doi.org/10.1117/12.863151

KEYWORDS: Image restoration, Remote sensing, Compressed sensing, Wavelets, Inverse problems, Image compression, Algorithms, Image visualization, Visualization, Reconstruction algorithms

Read Abstract +

Proceedings Article | 31 August 2009 Paper

Hyperspectral image lossless compression algorithm based on adaptive band regrouping

Mingyi He, Lin Bai, Yuchao Dai, Jing Zhang

Proceedings Volume 7455, 745504 (2009) https://doi.org/10.1117/12.827450

KEYWORDS: Hyperspectral imaging, Image compression, 3D modeling, Spectrometers, Spectral resolution, 3D image processing, Data storage, Data compression, Visible radiation, Image processing

Read Abstract +

Proceedings Article | 31 August 2009 Paper

A novel adaptive noise filtering method for SAR images

Weibin Li, Mingyi He

Proceedings Volume 7455, 745510 (2009) https://doi.org/10.1117/12.827428

KEYWORDS: Speckle, Synthetic aperture radar, Image filtering, Digital filtering, Remote sensing, Image processing, Optical filters, Nonlinear filtering, Filtering (signal processing), Earth sciences

Read Abstract +

Proceedings Article | 5 September 2008 Paper

Hopfield neural network based mixed pixel unmixing for multispectral data

Shaohui Mei, David Feng, Mingyi He

Proceedings Volume 7084, 70840C (2008) https://doi.org/10.1117/12.796341

KEYWORDS: Neurons, Signal to noise ratio, Neural networks, Spatial resolution, Quantitative analysis, Remote sensing, Earth observing sensors, Multispectral imaging, Evolutionary algorithms, Landsat

Read Abstract +

Proceedings Article | 11 June 2003 Paper

High-dimensional multispectral image fusion: classification by neural network

Mingyi He, Jiantao Xia

Proceedings Volume 4898, (2003) https://doi.org/10.1117/12.468120

KEYWORDS: Multispectral imaging, Neural networks, Image classification, Image fusion, Statistical analysis, Neurons, Algorithms, Sensor technology, Signals intelligence, Signal processing

Read Abstract +

Advances in sensor technology for Earth observation make it possible to collect multispectral data in much higher dimensionality. Such high dimensional data will it possible to classify more classes. However, it will also have several impacts on processing technology. First, because of its huge data, more processing power will be needed to process such high dimensional data. Second, because of its high dimensionality and the limited training samples, it is very difficult for Bayes method to estimate the parameters accurately. So the classification accuracy cannot be high enough. Neural Network is an intelligent signal processing method. MLFNN (Multi-Layer Feedforward Neural Network) directly learn from training samples and the probability model needs not to be estimated, the classification may be conducted through neural network fusion of multispectral images. The latent information about different classes can be extracted from training samples by MLFNN. However, because of the huge data and high dimensionality, MLFNN will face some serious difficulties: (1) There are many local minimal points in the error surface of MLFNN; (2) Over-fitting phenomena. These two difficulties depress the classification accuracy and generalization performance of MLFNN. In order to overcome these difficulties, the author proposed DPFNN (Double Parallel Feedforward Neural Networks) used to classify the high dimensional multispectral images. The model and learning algorithm of DPFNN with strong generalization performance are proposed, with emphases on the regularization of output weights and improvement of the generalization performance of DPFNN. As DPFNN is composed of MLFNN and SLFNN (Single-Layer Feedforward Neural Network), it has the advantages of MLFNN and SLFNN: (1) Good nonlinear mapping capability; (2) High learning speed for linear-like problem. Experimental results with generated data, 64-band practical multispectral images and 220-band multispectral images show that the new algorithm can overcome the over-fitting phenomena effectively and improve the generalization performance of DPFNN greatly. The classification accuracy of DPFNN with the new learning algorithm is much better than the traditional one.

Showing 5 of 7 publications

Conference Committee Involvement (4)

High-Performance Computing in Remote Sensing

19 September 2011 | Prague, Czech Republic

Satellite Data Compression, Communications, and Processing VI

3 August 2010 | San Diego, California, United States

Satellite Data Compression, Communication, and Processing V

4 August 2009 | San Diego, California, United States

Satellite Data Compression, Communication, and Processing IV

10 August 2008 | San Diego, California, United States

Keywords/Phrases

Search In:

Publication Years