Conference Details

May 30, 2016

 

Keynote: “Smart surveillance systems: trends and approaches”, Robert LaganiŹre  (9h0-10h00)
 

In the past decades, video surveillance systems have evolved from simple local digital video recorder to large scale cloud-based video monitoring solutions. The large amount of visual data captured by these visual systems calls for the design of more intelligent methods that can extract higher-level understanding of the observed scenes. If the pixel was the fundamental element that gave to computers the sense of sight, object detectors and classifiers are becoming the new fundamental bricks that will give computer the faculty to see and interpret the world. This talk discusses some of the challenges related to the development of the next generation of video surveillance systems.

 

 

Session 1: 3D&Depth acquisition, processing and applications (10h30-12h10)

Chair: G. Mączkowski

 

Š       “No-reference 3D mesh quality assessment based on dihedral angles model and support vector regression”. Ilyass Abouelaziz, Mohammed El Hassouni and Hocine Cherifi

Š       Kinect depth holes filling by similarity and position constrained sparse representation”. Jinhui Hu, Zhongyuan Wang and Ruolin Ruan

Š       “Color correction in 3D digital documentation. Case study”. Krzysztof Lech, Grzegorz Maczkowski and Eryk Bunsch

Š       “The Traveling Optical Scanner - Case Study on 3D Shape Models of Ancient Brazilian Skulls”. Camilla Trinderup, Vedrana Dahl, Kristian Gregersen, Ludovic Or- lando and Anders Dahl

Š       “Privacy Preserving Dynamic Room Layout Mapping”. Xinyu Li, Yanyi Zhang, Ivan Marsic and Randall Burd

 

Session 2: Segmentation, filtering and tracking (14h00-16h00)

Chair: R. LaganiŹre

 

Š       “On the benefit of state separation for tracking in image space with an Interacting Multiple Model Filter”. Stefan Becker, Hilke Kieritz, Wolfgang Hu¨bner and Michael Arens

Š       “Feature Asymmetry of the Conformal Monogenic Signal”. Ahror Belaid

Š       “Edge detection based on Riesz Transform”. Ahror Belaid, Soraya Aloui and Djamal  Boukerroui

Š       “Defect detection on patterned fabrics using entropy cues”. Maricela Martinez-Leon, Rocio Lizarraga-Morales, Carlos Rodriguez- Donate, Eduardo Cabal-Yepez and Ruth I. Mata-Chavez

Š       “Curve extraction by geodesics fusion: Application to polymer reptation analysis”. Somia Rahmoun, Fabrice Mairesse, Hiroshi Uji-I, Johan Hofkens and Tadeusz Sliwa

Š       “Single Image Super-Resolution using Sparse Representation on a K-NN Dictionary”. Ning Liu and Shuang Liang

 

 

 

Posters (16h00-17h00)

 

Š       “A Fragile Watermarking Scheme for Image Authentication Using Wavelet Transform”. Assma Azeroual and Afdel Karim

Š       “Unsupervised Classification of Synthetic aperture radar Imagery using a bootstrap version of the generalized mixture expectation maximization algorithm”. Ahlem Bougarradh

Š       “Palm trees detection from high spatial resolution satellite imagery using a new Contextual classification method with constraints”. Soufiane Idbraim and Driss Mammass

Š       “Fast Autonomous Crater Detection–for unmanned landing on unknown terrain”. Payel Sadhukhan and Sarbani Palit

Š       “Automatic Detection and Classification of Oil Tanks in Optical Satellite Images Based on Convolutional Neural Network”. Qingquan Wang, Jinfang Zhang, Xiaohui Hu and Yang Wang

Š       “A Chaotic Cryptosystem for Color Image with Dynamic Look-Up Table”. Med Karim Abdmouleh, Ali Khalfallah and Med Salim  Bouhlel

Š       “A New Method for Arabic Text Detection in Natural Scene Image based on the color homogeneity”. Houda Gaddour, Slim Kanoun and Nicole Vincent

 

May 31, 2016

 

Keynote: “Hierarchical Analysis of Multimodal Data”, Jocelyn Chanussot  (9h0-10h00)

 

There is a growing interest in the development of adapted processing tools for multimodal images (several images acquired over the same scene with different characteristics). Allowing a more complete description of the scene, multimodal images are of interest in various image-processing fields, but their optimal handling and exploitation raise several issues. In parallel, over the past decade, hierarchical representations have proven themselves extremely useful and powerful tools for data analysis in a variety of contexts. In this talk, we will investigate how hierarchical representations, a powerful tool for classical image analysis and processing, can be extended to multimodal images in order to better exploit the additional information brought by the multimodality and improve classical image processing techniques. The talk focuses on three different multimodalities frequently encountered in the remote sensing field. We first investigate the spectral-spatial information of hyperspectral images. Based on an adapted construction and processing of the hierarchical representation, we derive segmentation, which is optimal with respect to the spectral unmixing operation. We then focus on the temporal multimodality and sequences of hyperspectral images. Using the hierarchical representation of the frames in the sequence, we propose a new method to achieve object tracking and apply it to chemical gas plume tracking in thermal infrared hyperspectral video sequences. Finally, we study the sensorial multimodality, in which images are acquired with different sensors. Relying on the recently introduced concept of braids of partitions, we propose a novel methodology of image segmentation, based on an energetic minimization framework.

 

Session 3: Multispectral and Color Imaging (10h30-12h30)

Chair: N. Tsumura

 

Š       “Nonlinear Estimation of Chromphore Concentrations and Shading from Hyperspectral Images”. Rina Akaho, Misa Hirose and Norimichi Tsumura.

Š       “A color image database for haze model and methods evaluation”. Jessica El Khoury, Jean-Baptiste Thomas and Alamin Mansouri

Š       “Computer Vision Color Constancy from Maximal Projections Mean Assumption”. El khamssa Lakehal and Djemel Ziou

Š       Demosaicking Method for Multispectral Images Based on Spatial Gradient and Inter-channel Correlation”. Shu Ogawa, Kazuma Shinoda, Madoka Hasegawa, Shigeo Kato, Masahiro Ishikawa, Hideki Komagata and Naoki Kobayashi

Š       “Collaborative Unmixng Hyperspectral Imagery via Nonnegative Matrix Factorization”. Yaser Esmaeili Salehani and Saeed Gazor

Š       “Measuring Spectral Reflectance and 3D Shape using Multi-primary Image Projector”. Keita Hirai, Ryosuke Nakahata and Takahiko Horiuchi

 

 

Session 4: Signal Processing (14h00-16h00)

Chair: H. Douzi

 

Š       “Speaker Classification via Supervised Hierarchical Clustering using ICA Mixture Model”. Muhammad Azam and Nizar Bouguila

Š       “Speaker discrimination using several classifiers and a relativistic speaker characterization”. Siham Ouamour, Zohra Hamadache and Halim Sayoud

Š       “Speaker discrimination based on a fusion between neural and statistical classifiers”. Siham Ouamour and Halim Sayoud

Š       “Multiple-Instance Multiple-Label Learning for the Classification of Frog Calls With Acoustic Event Detection”. Jie Xie, Michael Towsey, Liang Zhang, Kiyomi Yasumiba, Lin Schwarzkopf, Jinglan Zhang and Paul Roe

Š       “Feature Extraction Based on Bandpass Filtering for Frog Call Classification”. Jie Xie, Michael Towsey, Liang Zhang, Jinglan Zhang and Paul Roe

Š       “Detection of Activities during Newborn Resuscitation Based on Short-Time Energy of  Acceleration Signal”. Huyen Vu, Trygve EftestŅl, Kjersti Engan, Joar EilevstjŅnn, Ladislaus Yarrot, Jrgen Linde and Hege Ersdal

 

Session 5: Watermarking, authentication and coding (16h30-17h50)

Chair: A. Elmoataz

 

Š       “Digital Watermarking Scheme Based on Arnold and Anti-Arnold Transforms”. M. Omair Ahmad and M. Abdallah Elayan

Š       “A JND model using a texture-Edge selector based on Faber-Schauder Wavelet lifting scheme”. Meina Amar, Rachid Harba, Hassan Douzi, Frédéric Ros, Mohamed El Hajji, Rabia Riad and Khadija Gourrame

Š       “Single-Loop Architecture for JPEG 2000”. David Barina, Ondrej Klima and Pavel  Zemcik

Š       “Robust print-cam image watermarking in Fourier domain”. Khadija Gourrame, Hassan Douzi, Rachid Harba, Frederic Ros, Mo- hamed El Hajji, Rabia Riad and Meina Amar

 

June 1, 2016

 

Keynote: “Deep Learning in Image Processing ”, Roland Memisevic (9h0-10h00)

 

Deep Learning is the latest surge of wide-spread scientific interest in neural networks. Unlike previous waves of interest, this one has produced several astonishing results in various application domains, thanks to the availability of massively parallel hardware and of large labelled datasets. In this talk, I will describe the role of deep learning in image and signal processing, and I will present recent applications in these domains. I will also discuss some of the main challenges ahead, such as the need for further scaling up computation and datasets, and I will present some of the ideas and approaches developed in our own lab to address these challenges. 

 

 

Session 7: Biomedical and biological imaging (10h30-12h30)

Chair: J. Meunier

 

Š       Otolith recognition system using a normal angles contour”. Youssef El Habouz, Youssef Es-Saady, Mostafa El Yassa, Driss Mam- mass, Fathallah Nouboud, Alain Chalifour and Khalid Manchih

Š       “A hybrid combination of multiple SVM classifiers for automatic recognition of the damages and symptoms on plant leaves”. Ismail El Massi, Youssef Es-Saady, Mostafa El Yassa, Driss Mammass and Abdeslam Benazoun

Š       “Leaf Classification using Convexity Measure of Polygons”. Jules Raymond Kala, Serestina Viriri, Deshendran Moodley and Jules- Raymond Tapamo

Š       “Classification of Eukaryotic Organisms through Cepstral Analysis of Mitochondrial DNA”. Emmanuel Adetiba and Oludayo Olugbara

Š       “A Novel Geometrical Approach For a Rapid Estimation Of The HARDI Signal in Diffusion MRI”. Ines Ben Alaya, Majdi Jribi, Faouzi Ghorbel and Tarek   Kraiem

Š      “Super-Resolved Enhancement of a Single Image and Its application in Cardiac MRI”. Guang Yang, Xujiong Ye, Greg Slabaugh, Jennifer Keegan, Raad Mo- hiaddin and David Firmin.