Showing papers in "Pattern Recognition Letters in 2019"

TL;DR: A novel deep learning framework for the detection and classification of breast cancer in breast cytology images using the concept of transfer learning is proposed and it has been observed that the proposed framework outclass all the other deep learning architectures in terms of accuracy in detection and classified of breast tumor in cytological images.

TL;DR: A deep learning model with an end-to-end structure on the standard 12-lead ECG signal for the diagnosis of MI has the potential to provide high performance on MI detection which can be used in wearable technologies and intensive care units.

TL;DR: Deep-ECG extracts significant features from one or more leads using a deep CNN and compares biometric templates by computing simple and fast distance functions, obtaining remarkable accuracy for identification, verification and periodic re-authentication.

TL;DR: This paper proposes a simple approach to implicitly select skin tissues based on their distinct pulsatility feature and shows that this method outperforms state of the art algorithms, without any critical face or skin detection.

TL;DR: The aim of this work is to classify each image into one of six facial emotion classes, based on single Deep Convolutional Neural Networks (DNNs), which contain convolution layers and deep residual blocks.

TL;DR: This paper presents a comprehensive review of the CNN-based action recognition methods according to three strategies and provides a guide for future research.

TL;DR: Two simple proofs of the triangle inequality for the Jaccard distance in terms of nonnegative, monotone, submodular functions are given and discussed in this paper, where they are shown to be equivalent.

TL;DR: It is shown that the task of phonation was more efficient than speech tasks in the detection of disease and compared with other approaches that use the same data set.

TL;DR: This work uses a fuzzy logic classifier to predict the degree of a particular emotion in AffectiveSpace and uses the combined model of deep convolutional neural networks and fuzzy logic is termed Convolutional Fuzzy Sentiment Classifier.

TL;DR: The use of recurrent neural networks (RNNs) are proposed to improve the online classification of hand gestures with Electromyography (EMG) signals acquired from the forearm muscles to achieve similar accuracy for both data sets.

TL;DR: A new hybrid technique which ensures the authenticity of the user to the system, as well as monitors whether the user has passed the biometric system as a normal or spoofed one is proposed, overcoming the limitations in normal authentication and spoofing practices.

TL;DR: A review of the literature of handwriting analysis for supporting the diagnosis of Alzheimer's and Parkinson's disease as well as of mild cognitive impairments (MCI) with the goal of providing interested researchers with the state-of-the-art research.

TL;DR: This study considered different univariate measures to produce a feature ranking and proposed a greedy search approach for choosing the feature subset able to maximize the classification results, and considered one of the most effective and widely used set of features in handwriting recognition.

TL;DR: A method to parameterize CNN based segmentation is introduced, bridging the gap between CNN based segmentsation and the rubbersheet-transform and enables the CNN segmentation as full segmentation step in any regular iris biometric system, or alternatively the segmentation can be utilized as a noise mask for other segmentation methods.

TL;DR: An attempt is made to develop an automated seizure detection method using deep neural network using the dataset collected from Bonn University, Germany and it was observed that StandardScaler and RobustScaler are equally good and are the best feature scaling techniques.

TL;DR: The discriminating power of “dynamically enhanced” static images of handwriting is investigated and a static representation that embeds dynamic information based on drawing the points of the samples, instead of linking them, so as to retain temporal/velocity information is proposed.

TL;DR: The proposed graph based learning approach, named Time based Graph Long Short-Term Memory (TGLSTM) network, is able to dynamically learn graphs when they may change during time, like in gait and action recognition.

TL;DR: Convolutional Neural Networks are employed to extract discriminating visual features from multiple representations of various graphomotor samples produced by both control and PD subjects, and these features are fed to a Support Vector Machine (SVM) classifier.

TL;DR: An effort is made to perform threat object detection by using deep neural networks based framework built upon Convolutional Neural Network (CNN) based techniques such as You Only Look Once (YOLO) and Faster Region based CNN (FRCNN) to performthreat object detection.

TL;DR: A two-layer Convolutional Neural Network is proposed to learn the high-level features which utilizes to the face identification via sparse representation via a precisely selected feature exactor to outperforms other methods on given datasets.

TL;DR: A recently developed optimally time-frequency concentrated even-length biorthogonal wavelet filter bank for automatically identifying coronary artery disease (CAD) is proposed, which has surpassed most of the state-of-art models.

TL;DR: A novel approach based on Convolutional Neural Networks to jointly predict depth maps and foreground separation masks used to condition Generative Adversarial Networks for hallucinating plausible color and depths in the initially occluded areas is proposed.

TL;DR: It is demonstrated that natural adversarial samples commonly occur and it is shown that many of these images remain misclassified even with additional training epochs, even though their correct classification may require only a small adjustment to network parameters.

TL;DR: A new algorithm that combines multiscale and multilevel evolutionary optimization (MSMLEO) with GPEI to optimize dozens of hyperparameters of neural networks with a variety of numerical types is presented.

TL;DR: An end-to-end deep-learning method for text-independent writer identification that does not require prior identification of features is proposed and achieved a better performance than the previously published best result based on handcrafted features and clustering algorithms.

TL;DR: Wang et al. as discussed by the authors proposed a depth information guided counting (DigCrowd) method to estimate the number of people in an EDOF image with a crowded scene, which first uses the depth information of an image to segment the scene into a far-view region and a nearview region.

TL;DR: This work proposes a fast density peaks algorithm that solves the time complexity problem, and maintains the generality of density peaks, which allows using it for all types of data, as long as a distance function is provided.

TL;DR: This work proposes a novel human-aware navigation strategy built upon the use of an adaptive spatial density function that efficiently cluster groups of people according to their spatial arrangement that combines a probabilistic roadmap and rapidly-exploring random tree path planners, and an adaptation of the elastic band algorithm.

TL;DR: Experimental results demonstrate that, compared with other feature extraction algorithms for multispectral images, GA-SURF can be computed much faster and are more robust and distinctive.

TL;DR: It turns out that for this kind of data, the use of weak classifiers learned over nearly incoherent features is very efficient and an empirical analysis of the parameters involved in the random subspace technique to guide the user in the choice of the appropriate hyper-parameters.