Hyperspectral Noise Reduction
The invention comprises proprietary techniques to improve the signal-to-noise ratio (SNR) of multi-dimensional data (e.g., hyperspectral datacubes) by removing noise in the data. This technology was developed to improve the SNR of hyperspectral data received from a satellite sensor, thus increasing the SNR from approximately 600:1 to more than 1000:1. The technology has been validated in several studies.
The CSA technology, referred to as Hybrid Spatial-Spectral Noise Reduction (HSSNR), uses a numerical processing approach to increase the SNR of acquired satellite images or sensor data. This processing maintains the signal unharmed, while removing the noise. The techniques are implemented in software, using conventional programming tools.
The technology uses wavelet-shrinkage, denoising-based methods to reduce noise in the observed signal: the wavelet transform operation decomposes the observed signal into signal-associated coefficients and noise-associated coefficients that are separable. The CSA's approach tackles the problem of the variable noise level of the multidimensional data. The approach is a hybrid of spatial and spectral wavelet shrinkage that benefits from the dissimilarity of the signal nature in the spatial dimensions and the spectral dimension and works in the spectral derivative domain.
This invention has potential to be used in other areas where increasing the SNR of multi-dimensional data is beneficial, such as three-dimensional medical imagery (MRI, positron emission tomography, enhanced ultrasound and computed tomography modalities), resulting in sharper images. In principle, the technology may find application in any field in which the data have three or more dimensions. Such applications include optical, radar and other forms of images.
The Business Opportunity
The invention has direct application to the processing of hyperspectral datacubes from spaceborne or airborne remote sensing. The invention can improve the SNR and data usefulness for end-users in government, university and industry (e.g., in agriculture, mining and forestry). The technology will interest companies that provide hyperspectral image analysis software or customized data processing services to end-users.
The technology will be of interest to companies and research organizations that are developing medical diagnostic imaging systems for hyperspectral or multi-dimensional data, such as MRI and advanced CT image analysis.
Technology Transfer Details
The technology is available for licensing.
The business opportunity may be referred to by its CSA case ID: 50769
- Qian, S.-E., "Enhancing space-based signal-to-noise ratios without redesigning the satellite," SPIE Newsroom, Article-3421, pp. 1-2, January 2011 (doi: 10.1117/2.1201012.03421)
- Qian, S.-E., J. Lévesque and R. Rashidi Far, "Assessment of Noise Reduction of Hyperspectral Imagery using a Target Detection Application," International Journal of Remote Sensing, accepted (in press).
- Qian, S.-E. and J. Lévesque, "Target Detection from Noise-Reduced Hyperspectral Imagery using Spectral Unmixing Approach," Journal of Optical Engineering, Vol. 48, No. 2, pp. 1-11 (026401), February 2009.
- Othman, H. and S.-E. Qian, "Evaluation of Wavelet Deniosed Hyperspectral Data for Remote Sensing," Canadian Journal of Remote Sensing, Vol. 34, Supplement 1, pp. 59-67, April 2008.
- Qian, S.-E., H. Othman and J. Lévesque, "Spectral Angle Mapper based Assessment of Detectability of Man-made Targets from Hyperspectral Imagery After SNR Enhancement," Proceedings SPIE, Vol. 6361, pp. 1-8, September 2006.
- Othman, H. and S.-E. Qian, "Noise Reduction of Hyperspectral Imagery Using Hybrid Spatial-Spectral Derivative-Domain Wavelet Shrinkage," Institute of Electrical and Electronics Engineers (IEEE), Transactions on Geoscience and Remote Sensing, Vol. 42, No. 2, pp. 397-408, February 2006.
- Number: 2,640,683
- Status: Pending patent application
- Source: Canadian Intellectual Property Office (CIPO)
- Number: 12/223,283
- Status: Pending patent application
- Source: United States Patent and Trademark Office (USPTO)
- Number: EP07719355
- Status: Issued patent
- Source: European Patent Office (EPO)
- Date modified: