ECG is a non-invasive tool used to detect cardiac arrhythmias. Many arrhythmias classification solutions with various ECG features have been reported in literature. In this work, a new method combined with a novel morphological feature is proposed for accurate recognition and classification of arrhythmias. First, the events of the ECG signals are detected. Then, parametric features of ECG morphology, i.e., amplitude, interval and duration, are extracted from selected ECG regions. Next, a novel feature for analyzing QRS complex morphology changes as visual patterns as well as a new clustering-based feature extraction algorithm is proposed. Finally, the feature vectors are applied to three well-known classifiers (neural network, SVM, and KNN) for automatic diagnosis. The proposed method was assessed with all fifteen types of heartbeats as recommended by the Association for Advancement of Medical Instrumentation from the MIT-BIH arrhythmia database and achieved the best overall accuracy of 97.70% based on KNN, using the combined parametric and visual pattern features of ECG Morphology. The accuracies for the six main types-normal (N), left bundled branch blocks (L), right bundled branch blocks (R), premature ventricular contractions (V), atrial premature beats (A) and paced beats (P) are 97.79%, 99.50%, 99.59%, 97.69%, 89.70%, and 99.92%, respectively. Comparisons with peer works prove a marginal progress in automatic heart arrhythmia classification performance.
To protect the security of vector maps, we propose a novel reversible watermarking scheme for vector maps based on a multilevel histogram modification. First, a difference histogram is constructed using the correlations of adjacent coordinates, and the histogram is divided into continuous regions and discontinuous regions by combining the characteristics of vector map data. Second, the histogram bins that require modification are determined in the continuous regions through the optimal peak value, and the peak values are chosen from the flanking discontinuous regions in both directions; the watermarks are embedded by adopting the multilevel histogram modification strategy. The watermark extraction process is the reverse of the embedding process, and after completing the watermark extraction, the carrier data can be recovered losslessly. The experimental results show that the proposed algorithm has good invisibility and is completely reversible. Compared with similar algorithms reported previously, it achieves higher watermark embedding capacity under the same embedding distortion with lower complexity, thereby having a higher application value.
Texture features are important characteristics in distinguishing collapsed buildings and intact buildings. However, texture features currently used in synthetic aperture radar (SAR) building damage assessment are extracted following the methods of optical images directly, which do not consider the statistical feature of speckles and limit the accuracy improving. Therefore, a statistical texture feature-G0-para-was proposed to reflect the homogeneity of buildings in complex urban areas after a disaster. The G0-para is arising from the G 0 distribution of SAR image and used to distinguish collapse buildings and intact buildings. First, the G0para is unified to satisfy different polarization data-single-/dual-/ quad-/compact-polarization. Second, the distinguishing ability of G0-para is under comparison in single-/dual-/quad-/compactpolarization, through the receiver operating characteristic (ROC) curve and the area under the ROC curve analysis. Then, collapsed buildings with RADARSAT-2 and ALOS-1 data are evaluated, selecting the optimal combinations of each mode and comparing with the preferable existing texture features. The results show that the statistical texture parameter-G0-para-is better than the variance of gray-level histogram and the contrast of gray level co-occurrence matrix in distinguishing intact buildings and collapsed ones and G0-para can be applied to single-/dual-/quad-/compact-polarimetric SAR data. For experimental data, VV and HH in single polarization, VH/VV and HH/HV in dual polarization, and hybrid mode in compact polarization are recommended when the best quad polarization is unavailable.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.