Nassim Ammour

Deep Learning Approach for Car Detection in UAV Imagery

This paper presents an automatic solution to the problem of detecting and counting cars in unmanned aerial vehicle (UAV) images. This is a challenging task given the very high spatial resolution of UAV images (on the order of a few centimetres) and the extremely high level of detail, which require suitable automatic analysis methods. Our proposed method begins by segmenting the input image into small homogeneous regions, which can be used as candidate locations for car detection. Next, a window is extracted around each region, and deep learning is used to mine highly descriptive features from these windows. We use a deep convolutional neural network (CNN) system that is already pre-trained on huge auxiliary data as a feature extraction tool, combined with a linear support vector machine (SVM) classifier to classify regions into “car” and “no-car” classes. The final step is devoted to a fine-tuning procedure which performs morphological dilation to smooth the detected regions and fill any holes. In addition, small isolated regions are analysed further using a few sliding rectangular windows to locate cars more accurately and remove false positives. To evaluate our method, experiments were conducted on a challenging set of real UAV images acquired over an urban area. The experimental results have proven that the proposed method outperforms the state-of-the-art methods, both in terms of accuracy and computational time.

Premature Ventricular Contraction Arrhythmia Detection and Classification with Gaussian Process and S Transform

This paper presents an efficient Bayesian classification system based on Gaussian process classifiers (GPC) for detecting premature ventricular contraction (PVC) beats in electrocardiographic (ECG) signals. GPC have the advantage over SVM classifiers in that the parameters of its kernel are automatically selected according to the Bayesian estimation procedure based on Laplace approximation. We also propose to feed the classifier with different representations of the ECG signals based on morphology, discrete wavelet transform, and S-transform. The latter representation has never been used for ECG signals before. The experimental results obtained on 48 records (i.e., 109887 heart beats) of the MIT-BIH arrhythmia database showed that for all feature representations adopted in this work, the proposed GP classifier combined with the S-transform and trained with only 600 beats from PVC and Non-PVC classes can provide an overall accuracy and a sensitivity above 96% on the whole 48 recordings.

Fusion of fingerprint and heartbeat biometrics using fuzzy adaptive genetic algorithm

The fusion of fingerprint and heartbeat is a promising multimodal biometric method especially for remote authentication using mobile computing devices. We propose the use of fuzzy adaptive genetic algorithm for the improvement of authentication performance of this multimodal biometrics. This method computes the optimal weights required for the fusion of matching scores from two modalities. This method is tested by a large database of multimodal biometrics composed of fingerprint and heartbeat signal captured from fingers. The proposed method outperforms existing methods in terms of authentication performances such as equal error rate (EER) and area under the ROC curve (AUR).

A hierarchical learning paradigm for semi-supervised classification of remote sensing images

In this paper, we present a new semi-supervised method for the classification of hyperspectral and VHR remote sensing images. The method is based on a hierarchical learning paradigm which is composed of multiple layers feeding into each other: 1) feature extraction layer, 2) classification layer, and 3) spatial regularization layer. In the feature extraction layer, the method employs morphological operators. In case of hyperspectral images, a dimensionality reduction step is first applied using an algorithm such PCA. In layer 2, the Extreme Learning Machine is trained and used to build an initial classification map of the image. Finally, in layer 3, a regularization step is applied to exploit spatial information between all pixels in the image. The Random Walker (RW) algorithm is used for this purpose, which uses the output results of layer 2, such as the class map and the posterior probabilities, as inputs. Initial results are obtained using the PAVIA dataset, which outperform the state-of-the-art methods in terms of accuracy and execution times.

Rhythm-based heartbeat duration normalization for atrial fibrillation detection

BACKGROUND Screening of atrial fibrillation (AF) for high-risk patients including all patients aged 65 years and older is important for prevention of risk of stroke. Different technologies such as modified blood pressure monitor, single lead ECG-based finger-probe, and smart phone using plethysmogram signal have been emerging for this purpose. All these technologies use irregularity of heartbeat duration as a feature for AF detection. We have investigated a normalization method of heartbeat duration for improved AF detection. METHOD AF is an arrhythmia in which heartbeat duration generally becomes irregularly irregular. From a window of heartbeat duration, we estimate the possible rhythm of the majority of heartbeats and normalize duration of all heartbeats in the window based on the rhythm so that we can measure the irregularity of heartbeats for both AF and non-AF rhythms in the same scale. Irregularity is measured by the entropy of distribution of the normalized duration. Then we classify a window of heartbeats as AF or non-AF by thresholding the measured irregularity. The effect of this normalization is evaluated by comparing AF detection performances using duration with the normalization, without normalization, and with other existing normalizations. RESULTS Sensitivity and specificity of AF detection using normalized heartbeat duration were tested on two landmark databases available online and compared with results of other methods (with/without normalization) by receiver operating characteristic (ROC) curves. ROC analysis showed that the normalization was able to improve the performance of AF detection and it was consistent for a wide range of sensitivity and specificity for use of different thresholds. Detection accuracy was also computed for equal rates of sensitivity and specificity for different methods. Using normalized heartbeat duration, we obtained 96.38% accuracy which is more than 4% improvement compared to AF detection without normalization. CONCLUSIONS The proposed normalization method was found useful for improving performance and robustness of AF detection. Incorporation of this method in a screening device could be crucial to reduce the risk of AF-related stroke. In general, the incorporation of the rhythm-based normalization in an AF detection method seems important for developing a robust AF screening device.

A cluster ensemble method for robust unsupervised classification of VHR remote sensing images

This paper present a novel ensemble method for clustering very high spatial resolution (VHR) images that is composed of four main steps. Firstly, because of the important role of the spatial component in VHR imagery, a set of morphological features are extracted from the original image using many openings and closings with increasing structural element sizes. Secondly, we construct the ensemble by running the k-means algorithm several times with different initializations. In order to increase the diversity, different subsets of features are randomly selected at each time. Third, an optimal relabeling of the ensemble with respect to a representative partition is made via a pairwise relabeling procedure. Finally, the relabeled maps are fused with a Markov Random Field (MRF) method. The Experimental results obtained on two real VHR images acquired by the sensors IKONOS-2 and GeoEye-1 over urban areas confirmed the promising capabilities of the proposed approach.

Selection of Heart-Biometric Templates for Fusion

The heart is potentially a highly secured biometric modality. Although many templates have been proposed to be extracted from heart-signal for biometric authentication, they have yet to reach a single digit equal error rate (EER) of false matches and false non-matches when applied on large across-session data sets, where gallery and probe data are taken from different sessions. However, since different templates possess different strengths, the fusion of them has a great potential to improve the authentication performance. We propose an efficient template selection algorithm to select a suitable subset of templates from a given set to obtain a minimal EER. The fusion of the subset of templates selected by this algorithm from a set of seven state-of-the-art templates has obtained a significant 5% reduction of EER in authentication in our experiments on a large database of finger-based ECG signals captured in two different sessions.

Tile-Based Semisupervised Classification of Large-Scale VHR Remote Sensing Images

This paper deals with the problem of the classification of large-scale very high-resolution (VHR) remote sensing (RS) images in a semisupervised scenario, where we have a limited training set (less than ten training samples per class). Typical pixel-based classification methods are unfeasible for large-scale VHR images. Thus, as a practical and efficient solution, we propose to subdivide the large image into a grid of tiles and then classify the tiles instead of classifying pixels. Our proposed method uses the power of a pretrained convolutional neural network (CNN) to first extract descriptive features from each tile. Next, a neural network classifier (composed of 2 fully connected layers) is trained in a semisupervised fashion and used to classify all remaining tiles in the image. This basically presents a coarse classification of the image, which is sufficient for many RS application. The second contribution deals with the employment of the semisupervised learning to improve the classification accuracy. We present a novel semisupervised approach which exploits both the spectral and spatial relationships embedded in the remaining unlabelled tiles. In particular, we embed a spectral graph Laplacian in the hidden layer of the neural network. In addition, we apply regularization of the output labels using a spatial graph Laplacian and the random Walker algorithm. Experimental results obtained by testing the method on two large-scale images acquired by the IKONOS2 sensor reveal promising capabilities of this method in terms of classification accuracy even with less than ten training samples per class.

A dynamic weights OWA fusion for ensemble clustering

In this work, a new image segmentation algorithm is introduced. The proposed algorithm combines the results of an hybrid clustering ensemble. The ensemble clustering is composed of fuzzy c-means (FCM) algorithm and fuzzy local information c-means (FCM_S1 and FCM_S2) algorithms with different values of the neighbors effect. The consensus technique is performed by the ordered weighted averaging (OWA) method. The weight attributed to each classifier can be modified during the process of classification and is determined by the classifier results of the pixel neighbors classification. Dynamic weights give the method the ability to select the temporal best performance during the classification process. Experiments performed on a synthetic image, and real images show that the proposed algorithm is effective and efficient and provides good noise elimination effect.

Siamese-GAN: Learning Invariant Representations for Aerial Vehicle Image Categorization

In this paper, we present a new algorithm for cross-domain classification in aerial vehicle images based on generative adversarial networks (GANs). The proposed method, called Siamese-GAN, learns invariant feature representations for both labeled and unlabeled images coming from two different domains. To this end, we train in an adversarial manner a Siamese encoder–decoder architecture coupled with a discriminator network. The encoder–decoder network has the task of matching the distributions of both domains in a shared space regularized by the reconstruction ability, while the discriminator seeks to distinguish between them. After this phase, we feed the resulting encoded labeled and unlabeled features to another network composed of two fully-connected layers for training and classification, respectively. Experiments on several cross-domain datasets composed of extremely high resolution (EHR) images acquired by manned/unmanned aerial vehicles (MAV/UAV) over the cities of Vaihingen, Toronto, Potsdam, and Trento are reported and discussed.