Mohammed Amin Belarbi

PCA as Dimensionality Reduction for Large-Scale Image Retrieval Systems

Dimensionalityufeffreductionufeffinufefflarge-scaleufeffimageufeffresearchufeffplaysufeffanufeffimportantufeffroleufeffforufefftheirufeffperformanceufeff inufeff differentufeff applications.ufeff Inufeff thisufeff paper,ufeff weufeff exploreufeff Principalufeff Componentufeff Analysisufeff (PCA)ufeff asufeff aufeff dimensionalityufeff reductionufeff method.ufeff Forufeff thisufeff purpose,ufeff first,ufeff theufeff Scaleufeff Invariantufeff Featureufeff Transformufeff (SIFT)ufefffeaturesufeffandufeffSpeededufeffUpufeffRobustufeffFeaturesufeff(SURF)ufeffareufeffextractedufeffasufeffimageufefffeatures.ufeffSecond,ufeff theufeffPCAufeffisufeffappliedufefftoufeffreduceufefftheufeffdimensionsufeffofufeffSIFTufeffandufeffSURFufefffeatureufeffdescriptors.ufeffByufeffcomparingufeff multipleufeffsetsufeffofufeffexperimentalufeffdataufeffwithufeffdifferentufeffimageufeffdatabases,ufeffweufeffhaveufeffconcludedufeffthatufeffPCAufeffwithufeffaufeff reductionufeffinufefftheufeffrange,ufeffcanufeffeffectivelyufeffreduceufefftheufeffcomputationalufeffcostufeffofufeffimageufefffeatures,ufeffandufeffmaintainufeff theufeffhighufeffretrievalufeffperformanceufeffasufeffwell KeywoRDS Dimensionality Reduction, Large-Scale Image Retrieval, Principal Component Analysis, Scale Invariant Feature Transform, Speeded Up Robust Features

Towards a smart selection of resources in the cloud for low-energy multimedia processing: Towards a Smart Selection of Cloud Resources

Nowadays, image and video processing applications have become widely used in many domains related to computer vision. Indeed, they can come from cameras, smartphones, social networks, or from medical devices. Generally, these images and videos are used for illustrating people or objects (cars, trains, planes, etc) in many situations such as airports, train stations, public areas, sport events, and hospitals. Thus, image and video processing algorithms have got increasing importance, they are required from various computer vision applications such as motion tracking, real time event detection, database (images and videos) indexation, and medical computer‐aided diagnosis methods. The main inconvenient of image and video processing applications is the high intensity of computation and the complex configuration and installation of the related materials and libraries. In this paper, we propose a new framework that allows users to select in a smart and efficient way the computing units (CPU or/and GPU) in a cloud‐based platform, in case of processing one image (or one video in real time) or many images (or videos). This framework enables to affect the local or remote computing units for calculation after analyzing the type of media and the algorithm complexity. The framework disposes of a set of selected CPU and GPU‐based computer vision methods, such as image denoising, histogram computation, features descriptors (SIFT, SURF), points of interest extraction, edges detection, silhouette extraction, and sparse and dense optical flow estimation. These primitive functions are exploited in various applications such as medical image segmentation, videos indexation, real time motion analysis, and left ventricle segmentation and tracking from 2D echocardiography. Experimental results showed a global speedup ranging from 5× to 273×(compared to CPU versions) as result of the application of our framework for the above‐mentioned methods. In addition to these performances, the parallel and heterogeneous implementations offered lower power consumption as result of the fast treatment.

Indexing Video by the Content

Indexing video by content represents an important research area that one can find in the field of intelligent search of videos. Visual characteristics such as color are one of the most relevant components used to achieve this task. We are proposing in this paper the basics of indexing by content, the various symbolic features and our approach. Our project is composed of a system based on two phases: an indexing process, which can take long time, and a search engine, which is done in real time because features are already extracted at the indexing phase.

A New Parallel and Distributed Approach for Large Scale Images Retrieval

The process of image retrieval presents a great interest in the domains of computer vision, video-surveillance, etc. Visual characteristics of image such as color, texture, shape are used to identify the content of images. However, the retrieving process becomes very challenging due to the hard management of large databases in terms of storage, computation complexity, performance and similarity representation.

Cloud-based platform for computer vision applications

During last years, images and videos have become widely used in many daily applications. Indeed, they can come from cameras, smartphones, social networks of from medical devices. Generally, these images and videos are used for illustrating people or objects (cars, trains, planes, etc.) in many situations such as airports, train stations, public areas, sport events, hospitals, etc. Thus, image and video processing algorithms have got increasing importance, they are required from various computer visions applications such as motion tracking, real time event detection, database (images and videos) indexation and medical computer aided diagnosis methods. In this paper, we propose a cloud platform that integrates the above-mentioned methods, which are generally developed with popular open source image and video processing libraries (OpenCV1, OpenGL2, ITK3, VTK4, etc.). Theses modules are automatically integrated and configured in the cloud application. Thus, the platform users will have access to different computer vision techniques without the need to download, install and configure the corresponding software. Each guest can select the required application, load its data and get the output results in a safe and simple way. The cloud platform can handle the variety of Operating Systems and programming languages (C++, Java, Python, etc.). Experimentations were conducted within two kinds of applications. The first represents medical methods such as image segmentation in MR images, 3D image reconstruction from 2D radiographs, left ventricle segmentation and tracking from 2D echocardiography. The second kind of applications is related to video processing such as face, people and cars tracking, and abnormal event detection in crowd videos.