Alireza Rahimpour

Distributed object recognition in smart camera networks

Distributed object recognition is a significantly fast-growing research area, mainly motivated by the emergence of high performance cameras and their integration with modern wireless sensor network technologies. In wireless distributed object recognition, the bandwidth is limited and it is desirable to avoid transmitting redundant visual features from multiple cameras to the base station. In this paper, we propose a histogram compression and feature selection framework based on Sparse Non-negative Matrix Factorization (SNMF). In our proposed method, histograms of the features are modeled as linear combination of a small set of signature vectors with associated weight vectors. The recognition process in the base station is then performed based on these small sets of transmitted weights from each camera. Furthermore, we propose another novel distributed object recognition scheme based on local classification in each camera and sending the label information to the base station and making the final decision based on majority voting. Experiments on BMW dataset affirm that our approach outperforms the state of the art in accuracy and bandwidth usage.

Robust coupling in space of sparse codes for multi-view recognition

Classical dictionary learning algorithms that rely on a single source of information have been successfully used for classification tasks. Additionally, the exploitation of multiple sources has shown to be advantageous in challenging real-world situations. We propose a new framework to exploit robust modality fusion in classification in order to achieve better classification performance than single source methods. Multimodal learning is able to leverage any correlations between sensor modalities found in the data. We propose a new bilevel optimization, referred to as (MCJWDL). We perform supervised dictionary learning while forcing a coupling between the resulting sparse codes from different sources of information. Extensive experiments demonstrate that MCJWDL outperforms state-of-the-art sparse representation and dictionary learning approaches for the multi-view object and multi-view action recognition.

Non-Intrusive Energy Disaggregation Using Non-Negative Matrix Factorization With Sum-to-k Constraint

Energy disaggregation or non-intrusive load monitoring addresses the issue of extracting device-level energy consumption information by monitoring the aggregated signal at one single measurement point without installing meters on each individual device. Energy disaggregation can be formulated as a source separation problem, where the aggregated signal is expressed as linear combination of basis vectors in a matrix factorization framework. In this paper, an approach based on Sum-to-k constrained non-negative matrix factorization (S2K-NMF) is proposed. By imposing the sum-to-k constraint and the non-negative constraint, S2K-NMF is able to effectively extract perceptually meaningful sources from complex mixtures. The strength of the proposed algorithm is demonstrated through two sets of experiments: Energy disaggregation in a residential smart home; and heating, ventilating, and air conditioning components energy monitoring in an industrial building testbed maintained at the Oak Ridge National Laboratory. Extensive experimental results demonstrate the superior performance of S2K-NMF as compared to state-of-the-art decomposition-based disaggregation algorithms.

Feature encoding in band-limited distributed surveillance systems

Distributed surveillance systems have become popular in recent years due to security concerns. However, transmitting high dimensional data in bandwidth-limited distributed systems becomes a major challenge. In this paper, we address this issue by proposing a novel probabilistic algorithm based on the divergence between the probability distributions of the visual features in order to reduce their dimensionality and thus save the network bandwidth in distributed wireless smart camera networks. We demonstrate the effectiveness of the proposed approach through extensive experiments on two surveillance recognition tasks.

Non-intrusive load monitoring of HVAC components using signal unmixing

Heating, Ventilating and Air Conditioning units (HVAC) are a major electrical energy consumer in buildings. Monitoring of the operation and energy consumption of HVAC would increase the awareness of building owners and maintenance service providers of the condition and quality of performance of these units, enabling conditioned-based maintenance which would help achieving high efficiency in energy consumption. In this paper, a novel non-intrusive method based on constrained non-negative matrix factorization is proposed for monitoring the different components of HVAC unit by only having access to the whole building aggregated power signal. At the first level of this hierarchical approach, power consumption of the building is decomposed to energy consumption of the HVAC unit and all the other electrical devices operating in the building such as lighting and plug loads. Then, the estimated power signal of the HVAC is used to estimate the power consumption profile of the HVAC major electrical loads such as compressors, condenser fans and indoor blower. Experiments conducted on real data collected from a building testbed maintained at the Oak Ridge National Laboratory (ORNL) demonstrate high accuracy on disaggregation task.

Dictionary Reduction: Automatic Compact Dictionary Learning for Classification

A complete and discriminative dictionary can achieve superior performance. However, it also consumes extra processing time and memory, especially for large datasets. Most existing compact dictionary learning methods need to set the dictionary size manually, therefore an appropriate dictionary size is usually obtained in an exhaustive search manner. How to automatically learn a compact dictionary with high fidelity is still an open challenge. We propose an automatic compact dictionary learning (ACDL) method which can guarantee a more compact and discriminative dictionary while at the same time maintaining the state-of-the-art classification performance. We incorporate two innovative components in the formulation of the dictionary learning algorithm. First, an indicator function is introduced that automatically removes highly correlated dictionary atoms with weak discrimination capacity. Second, two additional constraints, namely, the sum-to-one and the non-negative constraints are imposed on the sparse coefficients. On one hand, this achieves the same functionality as the \(L_2\)-normalization on the raw data to maintain a stable sparsity threshold. On the other hand, this effectively preserves the geometric structure of the raw data which would be otherwise destroyed by the \(L_2\)-normalization. Extensive evaluations have shown that the preservation of geometric structure of the raw data plays an important role in achieving high classification performance with smallest dictionary size. Experimental results conducted on four recognition problems demonstrate the proposed ACDL can achieve competitive classification performance using a drastically reduced dictionary (https://github.com/susanqq/ACDL.git).

Multimodal weighted dictionary learning

Classical dictionary learning algorithms that rely on a single source of information have been successfully used for the discriminative tasks. However, exploiting multiple sources has demonstrated its effectiveness in solving challenging real-world situations. We propose a new framework for feature fusion to achieve better classification performance as compared to the case where individual sources are utilized. In the context of multimodal data analysis, the modality configuration induces a strong group/coupling structure. The proposed method models the coupling between different modalities in space of sparse codes while at the same time within each modality a discriminative dictionary is learned in an all-vs-all scheme whose class-specific sub-parts are non-correlated. The proposed dictionary learning scheme is referred to as the multimodal weighted dictionary learning (MWDL). We demonstrate that MWDL outperforms state-of-the-art dictionary learning approaches in various experiments.