Supervised Dictionary Learning Research Articles

Energy-Aware IoT-Based Method for a Hybrid On-Wrist Fall Detection System Using a Supervised Dictionary Learning Technique.

In recent decades, falls have posed multiple critical health issues, especially for the older population, with their emerging growth. Recent research has shown that a wrist-based fall detection system offers an accessory-like comfortable solution for Internet of Things (IoT)-based monitoring. Nevertheless, an autonomous device for anywhere-anytime may present an energy consumption concern. Hence, this paper proposes a novel energy-aware IoT-based architecture for Message Queuing Telemetry Transport (MQTT)-based gateway-less monitoring for wearable fall detection. Accordingly, a hybrid double prediction technique based on Supervised Dictionary Learning was implemented to reinforce the detection efficiency of our previous works. A controlled dataset was collected for training (offline), while a real set of measurements of the proposed system was used for validation (online). It achieved a noteworthy offline and online detection performance of 99.8% and 91%, respectively, overpassing most of the related works using only an accelerometer. In the worst case, the system showed a battery consumption optimization by a minimum of 27.32 working hours, significantly higher than other research prototypes. The approach presented here proves to be promising for real applications, which require a reliable and long-term anywhere-anytime solution.

Discriminative semi-supervised learning via deep and dictionary representation for image classification

Supervised dictionary learning and deep learning have achieved promising performance in the classification task. However, in many real-world applications there usually exist very limited labeled training samples, although abundant unlabeled data is relatively easy to collect. How to effectively exploit the discrimination of unlabeled data is still an open question, hence semi-supervised learning has attracted much attention from wide fields. Semi-supervised deep feature learning has well exploited the feature discrimination from only the discriminative viewpoint, while dictionary representation-based classification has also been applied to semi-supervised learning but with shallow features. In this paper, we propose a novel discriminative semi-supervised learning via deep and dictionary representation (DSSLDDR), which jointly utilizes the discrimination of dictionary representation for data reconstruction and the distinguishing feature of each sample. To exploit the powerful discrimination of dictionary representation, class-specific dictionaries are required to discriminatively reconstruct a sample, with the reconstruction error to predict the sample’s class label. To exploit the semantic information, the deep neural network extracts discriminative features by using multiple nonlinear transformations to generate the powerful descriptor. Then the class-specific dictionary learning and deep network learning are integrated together to conduct more accurate class estimation for unlabeled data and learn a more discriminative classifier, where an entropy regularization is designed to balance and control the class estimation of unlabeled data. Furthermore, we propose the DSSLDDR++, the extension model of DSSLDDR based on consistency/contrastive learning to further improve the accuracy of class estimation for unlabeled data, making a more powerful semi-supervised learning classifier. Extensive experiments on benchmark datasets show the effectiveness of the proposed methods.

Learning discriminative visual semantic embedding for zero-shot recognition

We present a novel zero-shot learning (ZSL) method that concentrates on strengthening the discriminative visual information of the semantic embedding space for recognizing object classes. To address the ZSL problem, many previous works strive to learn a transformation to bridge the visual features and semantic representations, while ignoring that the discriminative property of the semantic embedding space can benefit zero-shot prediction tasks. Among these existing approaches, human-defined attributes are typically employed to build up the mid-level semantics. However, the discriminative capability and completeness of manually defined attributes are hard to guarantee, which may easily cause semantic ambiguity. To alleviate this issue, we propose a discriminative visual semantic embedding (DVSE) model that formulates the ZSL problem as a supervised dictionary learning framework. The proposed method is capable of exploring a set of discriminative visual attributes and ensures knowledge transfer across categories. Moreover, a unified objective is introduced to generate an augmented semantic embedding space where these learned visual attributes and human-defined attributes are incorporated jointly for consolidating the visual cues of feature representations. Finally, we treat the DVSE model as an optimization problem and further propose an iterative solver. Extensive experiments on several challenging benchmark datasets demonstrate that the proposed method achieves favorable performances compared with state-of-the-art ZSL approaches.

Deep supervised dictionary learning by algorithm unrolling-Application to fast 2D dynamic MR image reconstruction.

Unrolled neural networks (NNs) have been extensively applied to different image reconstruction problems across all imaging modalities. A key component of the latter is that they allow for physics-informed learning of the regularization method, which is parametrized by the NN. However, due to the lack of understanding of deep NNs from a theoretical point of view, unrolled NNs are still black-boxes when the regularizers are given by deep NNs, for example, U-Nets. Dictionarylearning (DL) is a well-established regularization method, which is based on learning a transform to sparsely approximate the signals of interest. Typically, DL-based image reconstruction either employs a dictionary, which was pretrained on a set of patches which were extracted from ground-truth images or a dictionary which is jointly trained during the reconstruction. However, in both cases, the used DL-algorithms are not designed to take into account the reconstruction problem or the underlying physical model, which describes the imaging process. In this work, we propose a DL-algorithm based on unrolled NNs to overcome these limitations. We construct an unrolled NN, which corresponds to an unrolled DL-based reconstruction algorithm and train the unrolled NN to optimize its weights, that is, the atoms of the dictionary, by back-propagation in a supervised manner. Further, we propose a new way to employ a 2D dictionary in the spatio-temporal domain. We tested and evaluated the method on an accelerated cardiac cine MR image reconstruction problem using 216/36/36 dynamic images for training, validation, and testing and compared it to two well-known state-of-the-art approaches for cardiac cine MRI based on deep iterative CNNs. Further, we analyze the obtained dictionaries in terms of dictionary-coherence and structure of the atoms. Last, we compare the reported methods in terms of stability by applying them to an entirely different dataset consisting of 49 different testimages. The investigated physics-informed DL-approach yields significantly more accurate reconstructions compared to the DL-method, which uses dictionaries obtained by decoupled pretraining, thereby providing an improvement of up to 4.90 dB in terms of PSNR and 5% in terms of SSIM. Further, the proposed spatio-temporal 2D dictionary outperforms the 1D and 3D dictionaries by preventing smoothing of image details while still accurately removing undersampling artifacts and noise resulting in an increase of up to 1.10 dB in terms of PSNR and 4% in terms of SSIM. Although being surpassed by the CNNs on the first dataset, the proposed NNs-based DL method is more stable compared to the latter approach and yields comparable results on the second dataset. Last, it has the advantage of being entirely interpretable in eachcomponent. The presented physics-informed NN can be used as training algorithm for a classical and interpretable data-driven regularization method based on a learned dictionary, which can then not only be linked to the considered data but also to the reconstruction method that the NNdefines.

Supervised Dictionary Learning for Block Threshold Feature in Compressive Spectrum Sensing

With the development of communication systems towards the high-frequency band, the demand for spectrum resources is ever-increasing, where the research interest has changed from narrowband spectrum sensing to wideband spectrum sensing. The Nyquist-rate-based wideband spectrum sensing with high-rate sampling is being questioned whether it is suitable for real-time applications. On the contrary, the well-known compressive spectrum sensing (CSS) is more appealing due to the compressive sensing (CS) technology, resulting in lower signal acquisition costs. However, the CS algorithms for recovering sparse spectrum generally require multiple iterations, which presents a challenge to the low complexity implementation of spectrum sensing. To address this issue, this paper proposes a novel method for solving the block CSS (BCSS) problem, where the spectrum of primary users is modeled as a block structure signal. Specifically, the block threshold feature (BTF) is utilized to reconstruct the spectrum while bypassing any iterative operations. Furthermore, to improve the performance of the BTF-based BCSS, we develop a novel supervised dictionary learning (SDL) model, based on the theoretical analysis of mutual incoherence and restricted isometry properties. Simulation results not only verify the compatibility of BTF and the SDL model, but also demonstrate the effectiveness and robustness of SDL-BTF-based BCSS for practical implementation.

Self-Balancing Dictionary Learning for Relaxed Collaborative Representation of Hyperspectral Image Classification

Supervised dictionary learning and representation learning framework has demonstrated its superiority for hyperspectral image classification. Relaxed collaborative representation (RCR) has also been acknowledged as an effective method in balancing the similarity and difference between features. In this paper, a new dictionary learning method is introduced to balance discrimination and reconstruction of training samples. In the dictionary learning stage, two new indicators are designed to measure the discriminability of items and can be improved by optimizing coding coefficients. The imposed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> -norm between independent item and the mean of class-specific samples constrains the similarity, and the calculated weights measure the difference. In label determination stage, considering that the residuals of RCR are adversely affected by the significant difference of features, a new classification approach is introduced. Class labels are assigned without calculating the reconstruction errors but calculating the levels of comprehensive contribution from all training samples instead. Since the effectiveness will be degraded when dealing with more complex circumstances, a region-based version is further introduced. It can further improve the discrimination of dictionary items due to the reduced categories in each sub-image and reduce the computation cost. The experimental results on several hyperspectral datasets demonstrate that our methods can effectively improve the classification performance.

Sparse Signal Reconstruction on Fixed and Adaptive Supervised Dictionary Learning for Transient Stability Assessment

Fixed and adaptive supervised dictionary learning (SDL) is proposed in this paper for wide-area stability assessment. Single and hybrid fixed structures are developed based on impulse dictionary (ID), discrete Haar transform (DHT), discrete cosine transform (DCT), discrete sine transform (DST), and discrete wavelet transform (DWT) for sparse features extraction and online transient stability prediction. The fixed structures performance is compared with that obtained from transient K-singular value decomposition (TK-SVD) implemented while adding a stability status term to the optimization problem. Stable and unstable dictionary learning are designed based on datasets recorded by simulating thousands of contingencies with varying faults, load, and generator switching on the IEEE 68-bus test system. This separate supervised learning of stable and unstable scenarios allows determining root mean square error (RMSE), useful for online stability status assessment of new scenarios. With respect to the RMSE performance metric in signal reconstruction-based stability prediction, the present analysis demonstrates that [DWT], [DHT|DWT] and [DST|DHT|DCT] are better stability descriptors compared to K-SVD, [DHT], [DCT], [DCT|DWT], [DHT|DCT], [ID|DCT|DST], and [DWT|DHT|DCT] on test datasets. However, the K-SVD approach is faster to execute in both off-line training and real-time playback while yielding satisfactory accuracy in transient stability prediction (i.e., 7.5-cycles decision window after fault-clearing).

Mineral quantification at deposit scale using drill-core hyperspectral data: A case study in the Iberian Pyrite Belt

Drill-core analysis is paramount for the characterization of deposits in mineral exploration. Over the past years, the use of hyperspectral (HS) sensors has rapidly increased to improve the reliability and efficiency of core logging. However, scanning drill-core samples of an entire mineral deposit entails several complex challenges, from transport logistics to large scale data management and analysis. Hence, academic studies on new applications of drill-core HS data at a mineral deposit scale remain rare.We present a semi-automated workflow for large scale interpretation of HS data, founded on a novel approach of mineral mapping based on a supervised dictionary learning technique. This approach exploits the complementary information from scanning electron microscopy based automated mineralogy and hyperspectral imaging techniques for estimating mineral quantities along all boreholes. We propose that it is effectively possible to propagate the mineral quantification to the entire borehole from small samples with high resolution mineralogical information strategically selected throughout the deposit.We showcase this approach on data acquired in the Elvira shale-hosted volcanogenic massive sulphide (VMS) deposit located at the Iberian Pyrite Belt (IPB), where 7000 m of drill-core were acquired along 80 boreholes. Resulting maps provide insights on the controls on the mineral assemblages and chemical composition of specific minerals across the whole volume at several spatial scales, from large scale variations within apparently homogeneous black shales to small scale mineral composition variations, of potential use as vectors towards mineralization. This approach adds value to the core data, allowing for a better understanding of the geological setting of the Elvira deposit and providing valuable insights for future exploration targeting in the region.This approach based on machine learning can easily be transposed to different ore deposits with a limited input from a geologist.

Finding Faults in PV Systems: Supervised and Unsupervised Dictionary Learning With SSTDR

This article explains the use of supervised and unsupervised dictionary learning approaches on spread spectrum time domain (SSTDR) data to detect and locate disconnections in a PV array consisting of five panels. The aim is to decompose an SSTDR reflection signature into different components where each component has a physical interpretation, such as noise, environmental effects, and faults. In the unsupervised dictionary learning approach, the decomposed components are inspected to detect and localize faults. The maximum difference between actual and predicted location of the fault is 0.44 m on a system with five panels connected to an SSTDR box with a leader cable of 59.13 m and total length of 67.36 m including the effective length of the panels. In the supervised dictionary learning approach, the dictionary components are used to classify the SSTDR data to their respective fault types. Our results show a 97% accuracy using the supervised learning approach.

Semisupervised Dictionary Learning with Graph Regularized and Active Points

Supervised Dictionary Learning has gained much interest in the recent decade and has shown significant performance improvements in image classification. However, in general, supervised learning needs a large number of labelled samples per class to achieve an acceptable result. In order to deal with databases which have just a few labelled samples per class, semi-supervised learning, which also exploits unlabelled samples in training phase is used. Indeed, unlabelled samples can help to regularize the learning model, yielding an improvement of classification accuracy. In this paper, we propose a new semi-supervised dictionary learning method based on two pillars: on one hand, we enforce manifold structure preservation from the original data into sparse code space using Locally Linear Embedding, which can be considered a regularization of sparse code; on the other hand, we train a semi-supervised classifier in sparse code space. We show that our approach provides an improvement over state-of-the-art semi-supervised dictionary learning methods.

Supervised Brain Network Learning Based on Deep Recurrent Neural Networks

Task-based functional magnetic resonance imaging (tfMRI) is a widely used neuroimaging technique in exploring brain networks and functions associated with cognitive behaviors. Traditionally, the general linear model (GLM) is the most popular method in tfMRI data analysis due to its simpleness and robustness. This model-driven method adopts a canonical hemodynamic response function (HRF) and its various derivatives to construct regressors in the design matrix and estimate changes in the tfMRI data. However, a possible limitation of current model-driven methods is that the HRF is fixed and non-adaptive which may overlook other diverse and concurrent brain networks. In order to overcome these limitations, we proposed a novel hybrid framework, supervised brain network learning based on deep recurrent neural networks (SUDRNN), to reconstruct the diverse and concurrent functional brain networks. Specifically, this hybrid framework first takes advantage of the great capacity of deep recurrent neural networks (DRNN) in modeling sequential data to learn the diverse regressors from real tfMRI data. After that, it utilizes the effective supervised dictionary learning (SDL) method to reconstruct both the task-related functional brain networks and other latent brain networks simultaneously. Extensive experiment results on different tfMRI datasets from Human connectome project (HCP) demonstrated the superiority of the proposed framework.

Multi-pose facial expression recognition using rectangular HOG feature extractor and label-consistent KSVD classifier

In this paper, a new approach to the classification of facial expressions from multiple pose images is proposed. In this approach, a rectangular histogram of oriented gradient (R-HOG) algorithm is first designed to extract features of face images. The parameters of the R-HOG algorithm, which is a modification of the original HOG algorithm include cell shape, cell size, block size, and the number of orientation bins. The R-HOG is capable of capturing more discriminative texture features of different facial expressions. In addition, a supervised dictionary learning classifier, the label-consistent K-SVD (LC-KSVD) algorithm, is adopted to recognise the facial expression of the subject. To investigate its effectiveness, the proposed technique was applied to classify emotional states of the face images in the two public available facial expression datasets: KDFE and RafD. The experiment study showed that the new method outperformed in many aspects those methods reported in the literature tested with the same datasets. First, the new method handles pose variations better. Second, it is more robust in cases where the size of a training dataset is small. Finally, it's accuracy performance is more consistent measured by standard deviations.

Structured dictionary learning using mixed-norms and group-sparsity constraint

Recently, sparse representation and dictionary learning have shown significant performance in machine vision. In particular, several supervised dictionary learning methods have been proposed for classification aim and increasing its accuracy. Among them, structured dictionary learning is an interesting approach which captures the discriminative properties of each class and common features among all classes in class-specific sub-dictionaries and a distinct shared sub-dictionary, respectively. It extracts the structural information that exists in samples of each class to increase the classification accuracy. Therefore, in this paper, a group-based structured dictionary learning method is proposed that captures structural information in each class and learns class-specific and shared sub-dictionaries based on mixed $$l_{2,1}$$ norm. Also, mixed $$l_{2,1}$$ norm is used for acquiring the sparse coefficients of data samples based on the learned sub-dictionaries. Then, classification is done by finding the class with (1) minimum reconstruction error or (2) maximum number of nonzero groups based on $$l_{1,0}$$ norm. The proposed method is evaluated by conducting experiments on Extended YaleB, AR and CMU-PIE face databases and the USPS handwritten digits database. The experimental results demonstrate the effectiveness of the proposed method in data representation and classification.

Supervised dictionary learning supported classifier with feature fusion scheme to noninvasively detect TRISO-particle defects

This paper presents a novel method to support analyzing TRISO-particle failure rate by exploiting X-ray phase contrast imaging (PCI) modality to nondestructively visualize inter-defects and supervised dictionary learning to automatically distinguish cracked particles. Histogram of oriented gradient (HOG) operator was combined with local binary pattern histogram Fourier (LBP-HF) descriptor by canonical correlation analysis (CCA) method in order to extract crack features more significantly. Label consistent K-singular value decomposition (LC K-SVD) dictionary learning followed to encode features with more discriminability and learn a dictionary capable of excluding noise and intra-class variability to enforce recognition, with comparatively high recognition accuracy.

Supervised dictionary learning of EEG signals for mild cognitive impairment diagnosis

Mild Cognitive Impairment (MCI) is an intermediate stage of memory decline between normal aging and Alzheimer’s disease or other types of dementia. MCI diagnosis is instrumental in preventing Alzheimer’s; however, its manifestations are complicated (i.e. distinguishing the symptoms is not easy) and MCI is often confused with the normal consequences of aging. To have an accurate and reliable Electroencephalogram (EEG)-based screening tool for MCI diagnosis, we are developing a new supervised dictionary-learning-based analysis of EEG signals, namely Correlation-based Label Consistent K-SVD (CLC-KSVD), which would solve the non-repeatability problem of conventional K-SVD. The proposed method is applied on both time and frequency domains, i.e. 1) the extracted patches from EEG signals recorded at resting state with eyes closed, and 2) the extracted spectral features from these EEG signals. The final label for each EEG signal (in different channels and zones) is obtained by voting between the labels of the whole of time and spectral patches. The evaluation results for the EEG signals of 61 subjects illustrate that CLC-KSVD outperforms other methods (the best accuracy of 88.9% was obtained in F7, T3 channels and the left temporal zone). Furthermore, the results are investigated using the volumetric analysis of Magnetic Resonance (MR) images, indicating that the most significant difference between healthy and MCI groups were in the superior temporal and pars triangularis in the left hemisphere. Such location matching between EEG and MR images demonstrates that the results of CLC-KSVD are anatomically interpretable.

A semi-blind online dictionary learning approach for fMRI data

BackgroundOnline dictionary learning (ODL) has been applied to extract brain networks from functional magnetic resonance imaging (fMRI) data in recent year. Moreover, the supervised dictionary learning (SDL) that fixed the task stimulus curves as predefined atoms was proposed to improve ODL for functional networks separation. However, SDL cannot estimate the real time courses underlying the brain networks and cannot be applied to the inter-network connectivity analysis. This study aimed at investigating how to add the temporal prior information to ODL to extract the accurate task-related brain networks and the corresponding time courses. New methodTo improve the performance of ODL, we propose a semi-blind ODL (semi-ODL) method that incorporates temporal prior information of the task paradigm into the dictionary updating process and optimizes the direction of one or more specific atoms "close" to the task time courses. ResultsResults of the simulated and real fMRI experiment revealed that semi-ODL extracted more accurate task-related component and time courses than ODL and SDL. For one-task fMRI data, semi-ODL and Infomax-ICA showed similar detection power in most cases. Comparison with existing methodsThe semi-ODL outperformed ODL, SDL in robustness to noise, spatial detection power and time course estimation. Moreover, semi-ODL showed comparable performance to Infomax-ICA for one-task fMRI data and outperformed Infomax-ICA in extracting the components related to each task from multi-task fMRI data. ConclusionsThe semi-ODL method is potentially useful to reveal brain networks underlying various cognitive tasks and the interactions between task-related brain networks.

Supervised Dictionary Learning With Regularization for Near-Infrared Spectroscopy Classification

Near-infrared spectroscopy (NIRS) has been widely used in many fields due to its advantages with fast analysis speed, non-destructive testing, and on-site detection. However, NIRS has some shortcomings, such as low signal-to-noise ratio, weak absorption intensity, and overlapping peaks. The research of near-infrared spectral modeling method becomes the core of analyzing NIRS. In order to improve the accuracy of prediction model for NIRS, this paper proposes a novel sparse classification mechanism by designing appropriate regularization factors. The existing supervised dictionary learning methods have been proposed for classification aim and increasing its accuracy, the proposed method addresses some defects existing in this area through designing the representation-constrained term and the coefficients incoherence term, and the added two terms can get the reconstruction error of coding coefficients and correlations between similar samples by sharing dictionary under more stable control. Then, based on the proposed model, a supervised class-specific dictionary learning algorithm is developed by choosing appropriate samples with class labels. Finally, a classification scheme integrating the novel sparse model is designed to exploit such discriminative information. The proposed method is evaluated by conducting experiments on drug and tobacco leaves NIRS datasets. The experimental results show that the proposed sparse classification mechanism is promising for classifying NIRS and may be an alternative method to the traditional ones.

Multiscale Incremental Dictionary Learning With Label Constraint for SAR Object Recognition

In this letter, a novel nonlinear supervised dictionary learning (DL) scheme called multiscale incremental DL, whose objective function contains reconstruction error terms and a classification error term, is proposed for synthetic aperture radar (SAR) object recognition. In the reconstruction error terms, considering the local and global features of SAR images, Gaussian functions with different blurring parameters are exploited to extract SAR images' multiscale features, and all features can be reconstructed according to the weights assigned to these features at different scales. In the classification error term, a linear combination of classification vectors close to the labels of samples restricts sparse codes from different classes to be almost independent. Furthermore, an incremental method is utilized to address the memory consumption problem, and the optimal solution is obtained. Experiments on the moving and stationary target automatic recognition database demonstrate that the proposed algorithm outperforms several representative DL, support vector machine, and k-nearest neighbor methods in the case of a small training sample set size and exhibits strong antinoise performance.

Unified No-Reference Quality Assessment of Singly and Multiply Distorted Stereoscopic Images.

A challenging problem in the no-reference quality assessment of multiply distorted stereoscopic images (MDSIs) is to simulate the monocular and binocular visual properties under a mixed type of distortions. Due to the joint effects of multiple distortions in MDSIs, the underlying monocular and binocular visual mechanisms have different manifestations with those of singly distorted stereoscopic images (SDSIs). This paper presents a unified no-reference quality evaluator for SDSIs and MDSIs by learning monocular and binocular local visual primitives (MB-LVPs). The main idea is to learn MB-LVPs to characterize the local receptive field properties of the visual cortex in response to SDSIs and MDSIs. Furthermore, we also consider that the learning of primitives should be performed in a task-driven manner. For this, two penalty terms including reconstruction error and quality inconsistency are jointly minimized within a supervised dictionary learning framework, generating a set of quality-oriented MB-LVPs for each single and multiple distortion modality. Given an input stereoscopic image, feature encoding is performed using the learned MB-LVPs as codebooks, resulting in the corresponding monocular and binocular responses. Finally, responses across all the modalities are fused with probabilistic weights which are determined by the modality-specific sparse reconstruction errors, yielding the final monocular and binocular features for quality regression. The superiority of our method has been verified on several SDSI and MDSI databases.

S184. Sparse representation and classification of neural spikes using supervised dictionary learning

Neural spike sorting is curial for neuroscience applications. For example, spike sorting allows designing of efficient human-computer interfaces. Therefore, many investigations, aiming for accurate computer-aided algorithms for spike detection and classification, have emerged. However, many of the available algorithms require human intervention during the classification step. This study proposes decomposition and classification of spike waveforms using dictionary learning. In sparse decomposition, each spike is represented as a unique sparse combination of the dictionary basis vectors. This can lead to interesting sorting performance. Spiking activity was detected using amplitude thresholding and a 2.5-ms window around the peak value. Under the assumption that each neuron has a consistent waveform, the dictionary atoms were learned using training samples. This learned dictionary was used to get a sparse code representation for the test samples. These codes can be used to reconstruct the original spikes and therefore it provides data compression capability. Classification of the neural spikes was done using the sparse coefficients alongside with the reconstruction error. To ensure high discrimination, the dictionary was learned such that the resultant atoms have correspondence to class labels. In other words, the dictionary is structured of sub-dictionaries each corresponds to a different spiking pattern. Subsequently, an optimization problem was solved to make sure that (i) the sub-dictionaries accurately represent the training data, and (ii) the coding coefficients are sparse and minimize the reconstruction error. A testing sample was classified by sparsely coding it over each sub-dictionary. The reconstruction error for each sub-dictionary is computed. An error function is defined for each class as the summation of the reconstruction error and the difference between the sparse code and that class average. The testing sample was assigned to the class that gave the minimum error. The algorithm was tested on four simulated data sets, each contains spiking activities from three different neurons. Approximately 10% of each class data was used to construct the dictionary. The reconstruction mean-squared error for all the training samples was 3.49. This confirms the reconstruction power and accuracy of the learned dictionary. We measured the accuracy of our algorithm at different noise levels. The classification accuracy of the proposed method is 97% ± 5%. This accuracy outperforms the existing state-of-the-art algorithms. A dictionary learning based spike sorting algorithm has been developed. The results suggest that dictionary-based representation enables us to use the sparse representation of signals without a severe loss of data; besides that, this sparse representation can be also used for classification. This could also enable estimation of missing data points or clipped signals. The reconstruction error can be used to efficiently sort spikes.