Image/Video Processing

Color Image Reconstruction

After the acquisition of a digital image from the sensor, a lot of tasks are necessary to improve the quality of the image. In particular, the raw data captured by the sensor of a digital camera require a suitable interpolation (called “demosaicking” [2,3,4,5]) in order to provide a full color representation of the scene. Often also a “denoising” procedure is necessary to remove the noise introduced during the acquisition, and a “deblurring” algorithm is applied to sharpen the edges of the image. Moreover, in many applications the resolution of the image is limited, therefore it is necessary a procedure to enlarge the size, called “image magnification” (or “super-resolution” if several low-resolution versions of the scene are considered [1]).

In our group, we are analyzing several reconstruction approaches, exploiting different mathematical models and signal processing strategies, in order to provide efficient and competitive solutions.


[1] S.C. Park, M.K. Park, and M.G. Kang, “Super-Resolution Image Reconstruction: A Technical Overview”, IEEE Signal Processing Magazine, May 2003.
[2] B. K. Gunturk et al., “Demosaicking: Color Filter Array Interpolation”, IEEE Signal Processing Magazine, Jan. 2005.
[3] D. Menon, S. Andriani, G. Calvagno, “Demosaicing with directional filtering and a posteriori decision”, IEEE Trans. on Image Processing, Jan. 2007.
[4] D. Menon, G. Calvagno, “Regularization approaches to demosaicking”, IEEE Transactions on Image Processing, Oct. 2009.
[5] D. Menon, G. Calvagno, “Color image demosaicking: an overview”, Signal Processing: Image Communications, Oct. 2011.

People involved

Daniele Menon, Giancarlo Calvagno

Image Reconstruction via Sparse Signal Representation

The sparse representation of signals is an emerging paradigm with increasing interests for applications related to image compression, denoising and reconstruction.


[1] E. J. Candes, M. B. Wakin, “An Introduction To Compressive Sampling”, IEEE Signal Processing Magazine, vol. 25, no. 2, pp. 21-30.
[2] J. Romberg, “Imaging via Compressive Sampling”, IEEE Signal Processing Magazine, vol. 25, no. 2, pp. 14-20.
[3] M. Elad, “Sparse and redundant representations”, Springer, 2010.

People involved

Giancarlo Calvagno

Cross-layer optimization for video transmission

The advent of wireless multimedia communications has brought to evidence that the traditional configurations of network protocols are not adequate for video delivery over time-varying channels. The modularization of layered architectures permits great flexibility and interoperability among heterogeneous networks and devices, but at the same time, it leads to significant inefficiencies considering the issues raised by the transmission of video sequences over communication infrastructures that are affected by data losses (such as the wireless systems, as an example). During the last years, a considerable research effort has been made to investigate efficient cross-layer (CL) solutions that are able to grant a good Quality-of-Service (QoS) to the end user [1,2]. In our work, we have designed different CL optimization algorithms that vary the protection level of the video packets according to their relevance in the decoding process [3]. Moreover, we studied alternative source coding paradigms, such as Multiple Description Coding, which permits a both scalable and robust coding of the video sequence [4]. The current research is focused on the integration of these solutions and on their application to different types of networks (such as Peer-to-Peer).


[1] Q. Zhang and F. Yang and W. Zhu, “Cross-Layer QoS Support for Multimedia Delivery over Wireless Internet”, EURASIP Journal on Applied signal Processing,
vol. 2, no. 2, 2005, pp. 207-219.
[2] A. K. Katsaggelos, Y. Eisenberg, F. Zhai, R. Berry and T. N. Pappas,
“Advances in Efficient Resource Allocation for Packet-Based Real-Time Video Transmission”, Proc. of IEEE, vol. 93, no. 1, Jan. 2005, pp. 135-147.
[3] S. Milani, G. Calvagno, R. Bernardini, P. Zontone, “Cross-Layer Joint Optimization of FEC Channel Codes and Multiple Description Coding for Video Delivery over IEEE 802.11e Links”, Proc. of the 2008 IEEE International Conference on Future Multimedia Networks (FMN 2008), Sep. 17-18, 2008, Cardiff, Wales, GB, pp. 472-478.
[4] S. Milani, G. Calvagno, R. Bernardini, R. Rinaldo, “A Low-Complexity Packet Classification Algorithm for Multiple Description Video Streaming over IEEE802.11e Networks”, Proc. of the 2008 IEEE International Conference on Image Processing (ICIP 2008), San Diego, Ca, USA, Oct. 2008.

People involved

Simone Milani, Giancarlo Calvagno

Image and video post processing

The activity concern the investigation of advanced techniques for the improvement of picture quality for still images and video sequences coded using DCT based techniques. Moreover, emerging techniques related to the improvement of 3D TV (stereo video signals) fruition are investigated.


[1] F. Michielin, G. Calvagno, P. Sartor, O. Erdler, “A Wavelet Based Deblocking Technique for DCT Based Compressed materials”, Proceedings of the 2nd IEEE International Conference on Consumer Electronics – Berlin (ICCE-Berlin 2012), Berlin, Germany, September 3–5, 2012.

People involved

Giancarlo Calvagno, Francesco Michielin

Cellular Automata-Based Image Compression

Cellular automata are dynamical systems and models of massively parallel computation that share many properties of the physical world. If the pixels of an image are treated as cells of a cellular automata, and opportune reversible rules are used for configuration updating, such systems may work well as non-linear means for image compression. Recently, such possibilities have started to be investigated.

In the DISP lab, current research activities in this field include binary image and depth map compression (with spatially scalable features) using cellular automatas. Other issues that we are investigating are the utilization of cellular automatas for alpha channel coding in video applications and the development of improved reversible rules for enhanced compression performance.
Moreover, we are studying the application of this technique to message passing algorithms in order to reduce their computational complexity.


[1] C. Cruz-Reyes and J. Kari, “Non-linear Subband Coding with Cellular Automata”, Proc. of the XII International conference on Automata and Formal Languages (AFL 2008), Balatonfured, Hungary, May 27-30, 2008, pp. 146-157.
[2] L. Cappellari, S. Milani, C. Cruz-Reyes, G. Calvagno, “Resolution Scalable Image Coding with Reversible Cellular Automata”, IEEE Transactions on Image Processing, Vol. 20, No. 5, May 2011, pp. 1461–1468.

People involved

Carlos Cruz-Reyes, Simone Milani, Lorenzo Cappellari, Giancarlo Calvagno