Aim and Scope:

We live in a world with multiple signals from different sensor modalities, such as thermal sensors for temperature measurement, RGB sensor for texture and color capturing, depth sensor for 3D geometric shape estimation, hyperspectral sensor for property diagnostics and so on. The information in one single modality is always limited, thus ways to synergistically derive individual and correlated knowledge from multiple complementary modalities favors a lot of emerging real applications based on multiple sensors. For example, image segmentation by using both thermal cameras and hyperspectral imaging techniques can achieve better performance than using them independently. Due to the increasing volume and modeling complexity of the multimodality date, how to process the available big data is challenging because of their heterogeneous nature, disparate origins, data quality of completeness. In addition, large volume multimodality data are usually noisy and prone to outliers. Therefore, opportunities of effective signal processing approaches for multimodality fusion arise. More specifically, problems under consideration of this special issue include: how to estimate the correlations among signals from different modalities, how to mine information from massive data, how to extract features in the presence of noise and interference and how to better exploit the available data at training stage when some modalities are disabled at the testing stage. In addition, signals from different sensors are of different domains, and how to bridge them is an emerging topic as well.

This special issue seeks to present and highlight ongoing research and real applications of signal processing for multiple modalities fusion. Real-world implementations or algorithms development to handle challenges are encouraged.

Topics of interest include, but are not limited to,
• Multimodal signal processing techniques: data acquisition, reconstruction.
• Multimodal signal preprocessing, denoising, artifact removal, data compression and interference reduction.
• Multimodal data analysis, multi-subject analysis, group analysis.
• Feature extraction and representation from different modalities by using compressive sensing, dictionary learning and sparse coding, low-rank models and dimension reduction algorithms
• Multi-modal data fusion, collaborative extraction and information fusion.
• Mobile phone based signal processing, energy management, living pattern recognition.
• Transfer learning and domain adaptive methods for multi-sensors
• Computer vision and color image & hyperspectral image processing
• Signal based event detection and localization in sensor networks
• Distributed multi-modal scene analysis and event interpretation
• Multi-modal human-computer interfaces

Authors can submit their manuscripts via the Manuscript Tracking System at http://mts.hindawi.com/submit/journals/jece/signal.processing/spmm/.

Manuscript Due Friday, 19 February 2016
First Round of Reviews Friday, 13 May 2016
Publication Date Friday, 8 July 2016