Traditional Classification Strategies Research Articles

Improvement of Spatio-Temporal Inconsistency of Time Series Land Cover Products

In recent years, time series land cover products have been developed rapidly. However, the traditional classification strategy rarely considers time continuity and spatial consistency, which leads to the existence of unreasonable changes among the multi-period products. In order to solve the existing problems, this paper proposes a matrix decomposition model and an optimized hidden Markov model (HMM) to improve the consistency of the time series land cover maps. It also compares the results with the spatio-temporal window filtering model. The spatial weight information is introduced into the singular value decomposition (SVD) model, and the regression model is constructed by combining the eigenvalues and eigenvectors of the image to predict the unreasonable variable pixels and complete the construction of the matrix decomposition model. To solve the two problems of reliance on expert experience and lack of spatial relationships, this paper optimizes the model and proposes the HMM Land Cover Transition (HMM_LCT) model. The overall accuracy of the matrix decomposition model and the HMM_LCT model is 90.74% and 89.87%, respectively. It is found that the matrix decomposition model has a better effect on consistency adjustment than the HMM_LCT model. The matrix decomposition model can also adjust the land cover trajectory to better express the changing trend of surface objects. After consistent adjustment by the matrix decomposition model, the cumulative proportion of the first 15 types of land cover trajectories reached 99.47%, of which 83.01% were stable land classes that had not changed for three years.

Flood Detection with SAR: A Review of Techniques and Datasets

Floods are among the most severe and impacting natural disasters. Their occurrence rate and intensity have been significantly increasing worldwide in the last years due to climate change and urbanization, bringing unprecedented effects on human lives and activities. Hence, providing a prompt response to flooding events is of crucial relevance for humanitarian, social and economic reasons. Satellite remote sensing using synthetic aperture radar (SAR) offers a great deal of support in facing flood events and mitigating their effects on a global scale. As opposed to multi-spectral sensors, SAR offers important advantages, as it enables Earth’s surface imaging regardless of weather and sunlight illumination conditions. In the last decade, the increasing availability of SAR data, even at no cost, thanks to the efforts of international and national space agencies, has been deeply stimulating research activities in every Earth observation field, including flood mapping and monitoring, where advanced processing paradigms, e.g., fuzzy logic, machine learning, data fusion, have been applied, demonstrating their superiority with respect to traditional classification strategies. However, a fair assessment of the performance and reliability of flood mapping techniques is of key importance for an efficient disasters response and, hence, should be addressed carefully and on a quantitative basis trough synthetic quality metrics and high-quality reference data. To this end, the recent development of open SAR datasets specifically covering flood events with related ground-truth reference data can support thorough and objective validation as well as reproducibility of results. Notwithstanding, SAR-based flood monitoring still suffers from severe limitations, especially in vegetated and urban areas, where complex scattering mechanisms can impair an accurate extraction of water regions. All such aspects, including classification methodologies, SAR datasets, validation strategies, challenges and future perspectives for SAR-based flood mapping are described and discussed.

Supervised segmentation for guiding peripheral revascularization with forward-viewing, robotically steered ultrasound guidewire.

Approximately 500000 patients present with critical limb ischemia (CLI) each year in the U.S., requiring revascularization to avoid amputation. While peripheral arteries can be revascularized via minimally invasive procedures, 25% of cases with chronic total occlusions are unsuccessful due to inability to route the guidewire beyond the proximal occlusion. Improvements to guidewire navigation would lead to limb salvage in a greater number of patients. Integrating ultrasound imaging into the guidewire could enable direct visualization of routes for guidewire advancement. In order to navigate a robotically-steerable guidewire with integrated imaging beyond a chronic occlusion proximal to the symptomatic lesion for revascularization, acquired ultrasound images must be segmented to visualize the path for guidewire advancement. The first approach for automated segmentation of viable paths through occlusions in peripheral arteries is demonstrated in simulations and experimentally-acquired data with a forward-viewing, robotically-steered guidewire imaging system. B-mode ultrasound images formed via synthetic aperture focusing (SAF) were segmented using a supervised approach (U-net architecture). A total of 2500 simulated images were used to train the classifier to distinguish the vessel wall and occlusion from viable paths for guidewire advancement. First, the size of the synthetic aperture resulting in the highest classification performance was determined in simulations (90 test images) and compared with traditional classifiers (global thresholding, local adaptive thresholding, and hierarchical classification). Next, classification performance as a function of the diameter of the remaining lumen (0.5 to 1.5mm) in the partially-occluded artery was tested using both simulated (60 test images at each of 7 diameters) and experimental data sets. Experimental test data sets were acquired in four 3D-printed phantoms from human anatomy and six ex vivo porcine arteries. Accuracy of classifying the path through the artery was evaluated using microcomputed tomography of phantoms and ex vivo arteries as a ground truth for comparison. An aperture size of 3.8mm resulted in the best-performing classification based on sensitivity and Jaccard index, with a significant increase in Jaccard index (p<0.05) as aperture diameter increased. In comparing the performance of the supervised classifier and traditional classification strategies with simulated test data, sensitivity and F1 score for U-net were 0.95±0.02 and 0.96±0.01, respectively, compared to 0.83±0.03 and 0.41±0.13 for the best-performing conventional approach, hierarchical classification. In simulated test images, sensitivity (p<0.05) and Jaccard index both increased with increasing artery diameter (p<0.05). Classification of images acquired in artery phantoms with remaining lumen diameters ≥ 0.75mm resulted in accuracies>90%, while mean accuracy decreased to 82% when artery diameter decreased to 0.5mm. For testing in ex vivo arteries, average binary accuracy, F1 score, Jaccard index, and sensitivity each exceeded 0.9. Segmentation of ultrasound images of partially-occluded peripheral arteries acquired with a forward-viewing, robotically-steered guidewire system was demonstrated for the first-time using representation learning. This could represent a fast, accurate approach for guiding peripheral revascularization.

An improved bearing fault detection strategy based on artificial bee colony algorithm

The operating state of bearing affects the performance of rotating machinery; thus, how to accurately extract features from the original vibration signals and recognise the faulty parts as early as possible is very critical. In this study, the one-dimensional ternary model which has been proved to be an effective statistical method in feature selection is introduced and shapelet transformation is proposed to calculate the parameter of one-dimensional ternary model that is usually selected by trial and error. Then XGBoost is used to recognise the faults from the obtained features, and artificial bee colony algorithm (ABC) is introduced to optimise the parameters of XGBoost. Moreover, for improving the performance of intelligent algorithm, an improved strategy where the evolution is guided by the probability that the optimal solution appears in certain solution space is proposed. The experimental results based on the failure vibration signal samples show that the average accuracy of fault signal recognition can reach 97%, which is much higher than the ones corresponding to traditional extraction strategies. And with the help of improved ABC algorithm, the performance of XGBoost classifier could be optimised; the accuracy could be improved from 97.02% to 98.60% compared with the traditional classification strategy.

Mapping and analyzing the local climate zones in China’s 32 major cities using Landsat imagery based on a novel convolutional neural network

ABSTRACT The Local Climate Zone (LCZ) scheme provides researchers with a standard method to monitor the Urban Heat Island (UHI) effect and conduct temperature studies. How to generate reliable LCZ maps has therefore become a research focus. In recent years, researchers have attempted to use Landsat imagery to delineate LCZs and generate maps worldwide based on the World Urban Database and Access Portal Tools (WUDAPT). However, the mapping results obtained by the WUDAPT method are not satisfactory. In this paper, to generate more accurate LCZ maps, we propose a novel Convolutional Neural Network (CNN) model (namely, LCZ-CNN), which is designed to cope with the issues of LCZ classification using Landsat imagery. Furthermore, in this study, we applied the LCZ-CNN model to generate LCZ mapping results for China’s 32 major cities distributed in various climatic zones, achieving a significantly better accuracy than the traditional classification strategies and a satisfactory computational efficiency. The proposed LCZ-CNN model achieved satisfactory classification accuracies in all 32 cities, and the Overall Accuracies (OAs) of more than half of the cities were higher than 80%. We also designed a series of experiments to comprehensively analyze the proposed LCZ-CNN model, with regard to the transferability of the network and the effectiveness of multi-seasonal information. It was found that the first convolutional stage, corresponding to low-level features, shows better transferability than the second and third convolutional stages, which extract high-level and more image- or task-oriented features. It was also confirmed that the multi-seasonal information can improve the accuracy of LCZ classification. The thermal characteristics of the different LCZ classes were also analyzed based on the mapping results for China’s 32 major cities, and the experimental results confirmed the close relationship between the LCZ classes and the magnitude of the Land Surface Temperature (LST).

Intervertebral disc degeneration associated with vertebral marrow fat, assessed using quantitative magnetic resonance imaging

ObjectiveTo investigate the potential clinical application of quantitative MRI in assessing the correlation between lumbar vertebrae bone marrow fat deposition and intervertebral disc degeneration.Materials and methodsA total of 104 chronic lower-back pain volunteers underwent 3.0-T MRI with T2-weighted imaging, T2 mapping, and iterative decomposition of water and fat with echo asymmetry and least squares estimation (IDEAL-IQ) between August 2018 and June 2019. Each disc was assessed with T2 value by T2 mapping, and the L1-S1 vertebral bone marrow fat fraction was assessed by IDEAL-IQ. The differences and relationship between T2 value and the adjacent vertebral bone marrow fat fraction values within the five Pfirrmann groups, five age groups, and five lumbar levels were statistically analyzed.ResultsThe vertebral bone marrow fat fraction had a significant negative correlation with T2 values of nucleus pulposus’ T2 values (p < 0.001). However, the significant negative correlation was only found between T2 values of nucleus pulposus and adjacent vertebral bone marrow fat in Pfirrmann II–III, L1/2-L5/S1 level, and 40–49 years’ age groups. Pfirrmann grades of the intervertebral disc were positively correlated with adjacent vertebrae bone marrow fat fraction (p < 0.05).ConclusionLumbar bone marrow fat deposition significantly increases during the early stages of intervertebral disc degeneration. Quantitative measurements of bone marrow fat deposition and water content of intervertebral discs have a predictive value and are an important supplement to the qualitative traditional classification strategies for the early stages of intervertebral disc degeneration.

RESEARCH ON SE-INCEPTION IN HIGH-RESOLUTION REMOTE SENSING IMAGE CLASSIFICATION

Abstract. With the deepening research and cross-fusion in the modern remote sensing image area, the classification of high spatial resolution remote sensing images has captured the attention of the researchers in the field of remote sensing. However, due to the serious phenomenon of “same object, different spectrum” and “same spectrum, different object” of high-resolution remote sensing image, the traditional classification strategy is hard to handle this challenge. In this paper, a remote sensing image scene classification model based on SENet and Inception-V3 is proposed by utilizing the deep learning method and transfer learning strategy. The model first adds a dropout layer before the full connection layer of the original Inception-V3 model to avoid over-fitting. Then we embed the SENet module into the Inception-V3 model for optimizing the network performance. In this paper, global average pooling is used as squeeze operation, and then two fully connected layers are used to construct a bottleneck structure. The model proposed in this paper is more non-linear, can better fit the complex correlation between channels, and greatly reduces the amount of parameters and computation. In the training process, this paper adopts the transfer learning strategy, makes full use of existing models and knowledge, improves training efficiency, and finally obtains scene classification results. The experimental results based on AID high-score remote sensing scene images show that SE-Inception has faster convergence speed and more stable training effect than the original Inception-V3 training. Compared with other traditional methods and deep learning networks, the improved model proposed in this paper has greater accuracy improvement.

Hierarchy-associated semantic-rule inference framework for classifying indoor scenes

Typically, the initial task of classifying indoor scenes is challenging, because the spatial layout and decoration of a scene can vary considerably. Recent efforts at classifying object relationships commonly depend on the results of scene annotation and predefined rules, making classification inflexible. Furthermore, annotation results are easily affected by external factors. Inspired by human cognition, a scene-classification framework was proposed using the empirically based annotation (EBA) and a match-over rule-based (MRB) inference system. The semantic hierarchy of images is exploited by EBA to construct rules empirically for MRB classification. The problem of scene classification is divided into low-level annotation and high-level inference from a macro perspective. Low-level annotation involves detecting the semantic hierarchy and annotating the scene with a deformable-parts model and a bag-of-visual-words model. In high-level inference, hierarchical rules are extracted to train the decision tree for classification. The categories of testing samples are generated from the parts to the whole. Compared with traditional classification strategies, the proposed semantic hierarchy and corresponding rules reduce the effect of a variable background and improve the classification performance. The proposed framework was evaluated on a popular indoor scene dataset, and the experimental results demonstrate its effectiveness.

A Hybrid Approach of Dimensionality Reduction and Classification for Hyperspectral Imager

Hyperspectral imagery widely used in remote sensing applications provides detailed data for the sensed materials than multispectral imagery. In this image, the identification of every ground surface picture element is represented by its corresponding spectral signatures which are unique but difficult to process as a result of the large volume of information. Traditional classification strategies might not be used along with the dimension reduction technique. This is often as a result of large dimension, that refers to the very fact that the sample size required to estimate a perform of many variables to a given degree of accuracy grows exponentially with the quantity of variables. Principal Component Analysis (PCA) has been the technique of alternative for dimension reduction. However, PCA is computationally extensive and not eliminate anomalies which will be seen at an arbitrary band. The high dimensional nature of the information collected by such sensors not solely will increase procedure complexness however can also degrade classification accuracy. To deal with this issue, dimension reduction (DR) has become a crucial aid to rising classifier potency on these imagery. In this paper, we summarized the hybrid approach of dimensionality reduction of HSI for the purpose of effective classification. The key findings revealed that the classification accuracy has been increased significantly on the application of dimensionality reduction techniques.

Band Grouping versus Band Clustering in SVM Ensemble Classification of Hyperspectral Imagery

Due to the dense sampling of spectral signatures of land covers, hyperspectral images have a better discrimination among similar ground cover classes than traditional remote sensing data. However, these images are usually composed of tens or hundreds of spectrally close bands, which result in high redundancy and great amount of computation time in hyperspectral image classifi cation. In addition, the large number of spectral bands, but limited availability of training samples creates the problem of Hughes phenomenon. Consequently, traditional classifi cation strategies have often limited performance in classifi cation of hyperspectral imagery. Referring to the limitation of single classifi ers in these situations, classifi er ensemble system may exhibit better performance. This paper presents a method for classifi cation of hyperspectral data based on two concepts of Band Clustering (BC) and Band Grouping (BG) through a Support Vector machine (SVM) ensemble system. The proposed method uses the BC\BG strategies to split data into few band portions. After this step, we applied SVM on each band cluster\group that is produced in previous step. Finally, Naive Bayes as a classifi er fusion method combines the decisions of SVM classifi ers. Experimental results show that the proposed method improves the classification accuracy in comparison to the standard SVM and to feature selection methods.

A Multiple SVM System for Classification of Hyperspectral Remote Sensing Data

With recent technological advances in remote sensing sensors and systems, very high-dimensional hyperspectral data are available for a better discrimination among different complex land-cover classes. However, the large number of spectral bands, but limited availability of training samples creates the problem of Hughes phenomenon or ‘curse of dimensionality’ in hyperspectral data sets. Moreover, these high numbers of bands are usually highly correlated. Because of these complexities of hyperspectral data, traditional classification strategies have often limited performance in classification of hyperspectral imagery. Referring to the limitation of single classifier in these situations, Multiple Classifier Systems (MCS) may have better performance than single classifier. This paper presents a new method for classification of hyperspectral data based on a band clustering strategy through a multiple Support Vector Machine system. The proposed method uses the band grouping process based on a modified mutual information strategy to split data into few band groups. After the band grouping step, the proposed algorithm aims at benefiting from the capabilities of SVM as classification method. So, the proposed approach applies SVM on each band group that is produced in a previous step. Finally, Naive Bayes (NB) as a classifier fusion method combines decisions of SVM classifiers. Experimental results on two common hyperspectral data sets show that the proposed method improves the classification accuracy in comparison with the standard SVM on entire bands of data and feature selection methods.