Image and Pervasive Access Laboratory

Singapore, Singapore

Image and Pervasive Access Laboratory

Singapore, Singapore
Time filter
Source Type

Dao-Duc C.,National University of Singapore | Xiaohui H.,National University of Singapore | Xiaohui H.,Shanghai JiaoTong University | Morere O.,Institute for Infocomm Research | And 2 more authors.
ACM International Conference Proceeding Series | Year: 2015

The ability to identify maritime vessels and their type is an important component of modern maritime safety and security. In this work, we present the application of deep convolutional neural networks to the classification of maritime vessel images. We use the AlexNet deep convolutional neural network as our base model and propose a new model that is twice smaller then the AlexNet. We conduct experiments on different configurations of the model on commodity hardware. We comparatively evaluate and analyse the performance of different configurations the model. We measure the top-1 and top-5 accuracy rates. The contribution of this work is the implementation, tuning and evaluation of automatic image classifier for the specific domain of maritime vessels with deep convolutional neural networks under the constraints imposed by commodity hardware and size of the image collection. © 2015 ACM.

Goh H.,Institute for Infocomm Research | Goh H.,University Pierre and Marie Curie | Goh H.,Image and Pervasive Access Laboratory | Goh H.,French National Center for Scientific Research | And 7 more authors.
Proceedings - International Conference on Image Processing, ICIP | Year: 2011

Our objective is to learn invariant color features directly from data via unsupervised learning. In this paper, we introduce a method to regularize restricted Boltzmann machines during training to obtain features that are sparse and topographically organized. Upon analysis, the features learned are Gabor-like and demonstrate a coding of orientation, spatial position, frequency and color that vary smoothly with the topography of the feature map. There is also differentiation between monochrome and color filters, with some exhibiting color-opponent properties. We also found that the learned representation is more invariant to affine image transformations and changes in illumination color. © 2011 IEEE.

Goh H.,University of Paris Pantheon Sorbonne | Goh H.,Institute for Infocomm Research | Goh H.,French National Center for Scientific Research | Goh H.,Image and Pervasive Access Laboratory | And 5 more authors.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

Recently, the coding of local features (e.g. SIFT) for image categorization tasks has been extensively studied. Incorporated within the Bag of Words (BoW) framework, these techniques optimize the projection of local features into the visual codebook, leading to state-of-the-art performances in many benchmark datasets. In this work, we propose a novel visual codebook learning approach using the restricted Boltzmann machine (RBM) as our generative model. Our contribution is three-fold. Firstly, we steer the unsupervised RBM learning using a regularization scheme, which decomposes into a combined prior for the sparsity of each feature's representation as well as the selectivity for each codeword. The codewords are then fine-tuned to be discriminative through the supervised learning from top-down labels. Secondly, we evaluate the proposed method with the Caltech-101 and 15-Scenes datasets, either matching or outperforming state-of-the-art results. The codebooks are compact and inference is fast. Finally, we introduce an original method to visualize the codebooks and decipher what each visual codeword encodes. © 2012 Springer-Verlag.

Chandrasekhar V.,Institute for Infocomm Research | Lin J.,Institute for Infocomm Research | Morere O.,Institute for Infocomm Research | Morere O.,University Pierre and Marie Curie | And 5 more authors.
Data Compression Conference Proceedings | Year: 2015

The first step in an image retrieval pipeline consists of comparing global descriptors from a large database to find a short list of candidate matching images. The more compact the global descriptor, the faster the descriptors can be compared for matching. State-of-the-art global descriptors based on Fisher Vectors are represented with tens of thousands of floating point numbers. While there is significant work on compression of local descriptors, there is relatively little work on compression of high dimensional Fisher Vectors. We study the problem of global descriptor compression in the context of image retrieval, focusing on extremely compact binary representations: 64-1024 bits. Motivated by the remarkable success of deep neural networks in recent literature, we propose a compression scheme based on deeply stacked Restricted Boltzmann Machines (SRBM), which learn lower dimensional non-linear subspaces on which the data lie. We provide a thorough evaluation of several state-of-the-art compression schemes based on PCA, Locality Sensitive Hashing, Product Quantization and greedy bit selection, and show that the proposed compression scheme outperforms all existing schemes. © 2015 IEEE.

Aloulou H.,Image and Pervasive Access Laboratory | Aloulou H.,Orange Group | Mokhtari M.,Image and Pervasive Access Laboratory | Mokhtari M.,Orange Group | And 7 more authors.
BMC Medical Informatics and Decision Making | Year: 2013

Background: With an ever-growing ageing population, dementia is fast becoming the chronic disease of the 21st century. Elderly people affected with dementia progressively lose their autonomy as they encounter problems in their Activities of Daily Living (ADLs). Hence, they need supervision and assistance from their family members or professional caregivers, which can often lead to underestimated psychological and financial stress for all parties. The use of Ambient Assistive Living (AAL) technologies aims to empower people with dementia and relieve the burden of their caregivers.The aim of this paper is to present the approach we have adopted to develop and deploy a system for ambient assistive living in an operating nursing home, and evaluate its performance and usability in real conditions. Based on this approach, we emphasise on the importance of deployments in real world settings as opposed to prototype testing in laboratories. Methods. We chose to conduct this work in close partnership with end-users (dementia patients) and specialists in dementia care (professional caregivers). Our trial was conducted during a period of 14 months within three rooms in a nursing home in Singapore, and with the participation of eight dementia patients and two caregivers. A technical ambient assistive living solution, consisting of a set of sensors and devices controlled by a software platform, was deployed in the collaborating nursing home. The trial was preceded by a pre-deployment period to organise several observation sessions with dementia patients and focus group discussions with professional caregivers. A process of ground truth and system's log data gathering was also planned prior to the trial and a system performance evaluation was realised during the deployment period with the help of caregivers. An ethical approval was obtained prior to real life deployment of our solution. Results: Patients' observations and discussions allowed us to gather a set of requirements that a system for elders with mild-dementia should fulfil. In fact, our deployment has exposed more concrete requirements and problems that need to be addressed, and which cannot be identified in laboratory testing. Issues that were neither forecasted during the design phase nor during the laboratory testing surfaced during deployment, thus affecting the effectiveness of the proposed solution. Results of the system performance evaluation show the evolution of system precision and uptime over the deployment phases, while data analysis demonstrates the ability to provide early detection of the degradation of patients' conditions. A qualitative feedback was collected from caregivers and doctors and a set of lessons learned emerged from this deployment experience. (Continued on next page) (Continued from previous page). Conclusion: Lessons learned from this study were very useful for our research work and can serve as inspiration for developers and providers of assistive living services. They confirmed the importance of real deployment to evaluate assistive solutions especially with the involvement of professional caregivers. They also asserted the need for larger deployments. Larger deployments will allow to conduct surveys on assistive solutions social and health impact, even though they are time and manpower consuming during their first phases. © 2013 Aloulou et al.; licensee BioMed Central Ltd.

Aloulou H.,CNRS Laboratory for Informatics | Aloulou H.,Orange Group | Mokhtari M.,CNRS Laboratory for Informatics | Mokhtari M.,Orange Group | And 5 more authors.
IEEE Journal of Biomedical and Health Informatics | Year: 2014

On account of chronic neurocognitive disorders, many people progressively lose their autonomy and become more dependent on others, finally reaching the stage when they need round-the-clock care from caregivers. Over time, as patients' needs increase with the evolution of their diseases, caregivers experience increasing levels of stress and burden. Therefore, an assistive solution that is able to adapt to the changing needs of the end-users is needed. This need was considered as a major requirement that emerged from our field work and deployment experience in Singapore. In this paper, we focus on the technical aspects of our deployment, where we were interested in solving the technical requirement of adaptability and extendibility of the framework that has emerged from our predeployment analysis and discussions with professional caregivers. We expose our approach for dynamic integration of assistive services with their related sensing technologies and interaction devices and provide the technical results of the deployment of this solution. We also provide guidelines for real-world deployment of assistive solutions. © 2013 IEEE.

Irshad H.,Joseph Fourier University | Irshad H.,French National Center for Scientific Research | Veillard A.,University Pierre and Marie Curie | Veillard A.,Image and Pervasive Access Laboratory | And 4 more authors.
IEEE Reviews in Biomedical Engineering | Year: 2014

Digital pathology represents one of the major evolutions in modern medicine. Pathological examinations constitute the gold standard in many medical protocols, and also play a critical and legal role in the diagnosis process. In the conventional cancer diagnosis, pathologists analyze biopsies to make diagnostic and prognostic assessments, mainly based on the cell morphology and architecture distribution. Recently, computerized methods have been rapidly evolving in the area of digital pathology, with growing applications related to nuclei detection, segmentation, and classification. In cancer research, these approaches have played, and will continue to play a key (often bottleneck) role in minimizing human intervention, consolidating pertinent second opinions, and providing traceable clinical information. Pathological studies have been conducted for numerous cancer detection and grading applications, including brain, breast, cervix, lung, and prostate cancer grading. Our study presents, discusses, and extracts the major trends from an exhaustive overview of various nuclei detection, segmentation, feature computation, and classification techniques used in histopathology imagery, specifically in hematoxylin-eosin and immunohistochemical staining protocols. This study also enables us to measure the challenges that remain, in order to reach robust analysis of whole slide images, essential high content imaging with diagnostic biomarkers and prognosis support in digital pathology. © 2014 IEEE.

Goh H.,Agency for Science, Technology and Research Singapore | Goh H.,Image and Pervasive Access Laboratory | Thome N.,University Pierre and Marie Curie | Cord M.,University Pierre and Marie Curie | And 2 more authors.
IEEE Transactions on Neural Networks and Learning Systems | Year: 2014

In this paper, we propose a hybrid architecture that combines the image modeling strengths of the bag of words framework with the representational power and adaptability of learning deep architectures. Local gradient-based descriptors, such as SIFT, are encoded via a hierarchical coding scheme composed of spatial aggregating restricted Boltzmann machines (RBM). For each coding layer, we regularize the RBM by encouraging representations to fit both sparse and selective distributions. Supervised fine-tuning is used to enhance the quality of the visual representation for the categorization task. We performed a thorough experimental evaluation using three image categorization data sets. The hierarchical coding scheme achieved competitive categorization accuracies of 79.7% and 86.4% on the Caltech-101 and 15-Scenes data sets, respectively. The visual representations learned are compact and the model's inference is fast, as compared with sparse coding methods. The low-level representations of descriptors that were learned using this method result in generic features that we empirically found to be transferrable between different image data sets. Further analysis reveal the significance of supervised fine-tuning when the architecture has two layers of representations as opposed to a single layer. © 2012 IEEE.

Elawady M.,CNRS Hubert Curien Laboratory | Sadek I.,Image and Pervasive Access Laboratory | Shabayek A.E.R.,Suez Canal University | Pons G.,University of Girona | Ganau S.,Center Diagnostic
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016

Breast cancer is one of the leading causes of cancer death among women worldwide. The proposed approach comprises three steps as follows. Firstly, the image is preprocessed to remove speckle noise while preserving important features of the image. Three methods are investigated, i.e., Frost Filter, Detail Preserving Anisotropic Diffusion, and Probabilistic Patch-Based Filter. Secondly, Normalized Cut or Quick Shift is used to provide an initial segmentation map for breast lesions. Thirdly, a postprocessing step is proposed to select the correct region from a set of candidate regions. This approach is implemented on a dataset containing 20 B-mode ultrasound images, acquired from UDIAT Diagnostic Center of Sabadell, Spain. The overall system performance is determined against the ground truth images. The best system performance is achieved through the following combinations: Frost Filter with Quick Shift, Detail Preserving Anisotropic Diffusion with Normalized Cut and Probabilistic Patch-Based with Normalized Cut. © Springer International Publishing Switzerland 2016.

Aloulou H.,Orange Group | Aloulou H.,Laboratory of Informatics | Abdulrazak B.,Laboratory of Informatics | Abdulrazak B.,Université de Sherbrooke | And 8 more authors.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016

Simplify deployment and maintenance of Ambient Intelligence solutions is important to enable large-scale deployment and maximize the use/benefit of these solutions. More mature Ambient Intelligence solutions emerge on the market as a result of an intensive investment in research. This research targets mainly the accuracy, usefulness, and usability aspects of the solutions. Still, possibility to adapt to different environments, ease of deployment and maintenance are ongoing problems of Ambient Intelligence. Existing solutions require an expert to move on-site in order to install or maintain systems. Therefore, we present in this paper our attempt to enable quick large scale deployment. We discuss lessons learned from our approach for automating the deployment process in order to be performed by ordinary people. We also introduce a solution for simplifying the monitoring and maintenance of installed systems. © Springer International Publishing Switzerland 2016.

Loading Image and Pervasive Access Laboratory collaborators
Loading Image and Pervasive Access Laboratory collaborators