Single-type Features Research Articles

Magnetic resonance imaging radiomics based on artificial intelligence is helpful to evaluate the prognosis of single hepatocellular carcinoma.

Previous studies mostly use single-type features to establish a prediction model. We aim to develop a comprehensive prediction model that effectively identify patients with poor prognosis for single hepatocellular carcinoma (HCC) based on artificial intelligence (AI). Patients and methods: 236 single HCC patients were studied to establish a comprehensive prediction model. We collected the basic information of patients and used AI to extract the features of magnetic resonance (MR) images. The clinical model based on linear regression (LR) algorithm (AUC: 0.658, 95%CI: 0.5021-0.8137), the radiomics model and deep transfer learning (DTL) model based on light gradient-boosting machine (Light GBM) algorithm (AUC: 0.761, 95%CI: 0.6326-0.8886 and AUC: 0.784, 95%CI: 0.6587-0.9087, respectively) were the optimal prediction models. A comparison revealed that the integrated nomogram had the largest area under the receiver operating characteristic curve (AUC) (all P<0.05). In the training cohort, the integrated nomogram was predictive of recurrence-free survival (RFS) as well as overall survival (OS) (C-index: 0.735 and 0.712, P<0.001). In the test cohort, the integrated nomogram also can predict RFS and OS (C-index: 0.718 and 0.740, P<0.001) in patients. The integrated nomogram composed of signatures in the prediction models can not only predict the postoperative recurrence of single HCC patients but also stratify the risk of OS after the operation.

Multi-type feature fusion visual navigation for asteroid landing

In order to address the challenges encountered in visual navigation for asteroid landing using traditional point features, such as significant recognition and extraction errors, low computational efficiency, and limited navigation accuracy, a novel approach for multi-type fusion visual navigation is proposed. This method aims to overcome the limitations of single-type features and enhance navigation accuracy. Analytical criteria for selecting multi-type features are introduced, which simultaneously improve computational efficiency and system navigation accuracy. Concerning pose estimation, both absolute and relative pose estimation methods based on multi-type feature fusion are proposed, and multi-type feature normalization is established, which significantly improves system navigation accuracy and lays the groundwork for flexible application of joint absolute-relative estimation. The feasibility and effectiveness of the proposed method are validated through simulation experiments through 4769 Castalia.

Multi-feature fusion based face forgery detection with local and global characteristics.

The malicious use of deepfake videos seriously affects information security and brings great harm to society. Currently, deepfake videos are mainly generated based on deep learning methods, which are difficult to be recognized by the naked eye, therefore, it is of great significance to study accurate and efficient deepfake video detection techniques. Most of the existing detection methods focus on analyzing the discriminative information in a specific feature domain for classification from a local or global perspective. Such detection methods based on a single type feature have certain limitations in practical applications. In this paper, we propose a deepfake detection method with the ability to comprehensively analyze the forgery face features, which integrates features in the space domain, noise domain, and frequency domain, and uses the Inception Transformer to learn the mix of global and local information dynamically. We evaluate the proposed method on the DFDC, Celeb-DF, and FaceForensic++ benchmark datasets. Extensive experiments verify the effectiveness and good generalization of the proposed method. Compared with the optimal model, the proposed method with a small number of parameters does not use pre-training, distillation, or assembly, but still achieves competitive performance. The ablation experiments evaluate the role of each component.

Efficient Video Steganalytic Feature Design by Exploiting Local Optimality and Lagrangian Cost Quotient

As the opponent of motion vector (MV)-based video steganography, the corresponding symmetric steganalysis has also developed a lot in recent years, among which the logic-based steganalytic schemes, e.g., AoSO, NPELO and MVC, are the most prevailing. Although currently achieving the best detection performance, these steganalytic schemes are less effective in detecting some logic-maintaining steganographic schemes. In view of the fact that the distributions of covers’ local Lagrangian cost quotients are normally more concentrated in the small value ranges than those of stegos and “spread” to the large values ranges after modifying the motion vector, the local Lagrangian cost quotient would thus be an efficient indicator to reflect the difference between cover videos and stego ones. In this regard, combining the logic-based (Lg) and local Lagrangian cost quotient (LLCQ)-based feature, we finally proposed a more effective and general steganalysis feature, i.e., Lg-LLCQ, which is composed of diverse subfeatures and performs much better than the corresponding single-type feature. Extensive experimental results show that the proposed method exhibits detection performance superior to other state-of-the-art schemes and even works well under cover sources and steganographic scheme mismatch scenes, which indicates our proposed feature is more conducive to real-world applications.

GCN-based multi-scale dual fusion for remote sensing building change detection

ABSTRACT In recent years, Graph Neural Networks (GNN) have begun to receive extensive attention from researchers. Subsequently, ViG was proposed and its performance in learning irregular feature information in non-Euclidean data space was astonishing. Inspired by the success of ViG, we propose a GNN-based multi-scale fusion network model (GCNCD) to extract graph-level features for remote sensing building change detection (CD). GCNCD builds bitemporal images into a graph structure. It then learns richer features by aggregating the features (edge information) of neighbour vertices in the graph. To alleviate the over-smoothing problem caused by multi-layer graph convolution, the FNN module is used to improve the network’s ability to transform features and reduce the loss of spatial structure information. Compared with the traditional single-type feature fusion module, in the decoder, we perform feature fusion on adjacent-scale features and all scale features, respectively. It helps to promote information mobility and reduce spatial information loss. Our extensive experiments demonstrate the positive effects of graph convolution and fusion module in the field of remote sensing building change detection.

Estimating Human Weight From a Single Image

Body weight, as one of the biometric traits, has been studied in both the forensic and medical domains. However, estimating weight directly from 2D images is particularly challenging since visual inspection is rather sensitive to the distance between the subject and camera, even for frontal view images. In this case, the widely used body mass index (BMI), which is associated with body height and weight, can be employed as a measure of weight to indicate health conditions. Previous works on the estimation of BMI have predominantly focused on using multiple 2D images, 3D images, or facial images; however, these cues are not always available. To address this issue, we explore the feasibility of obtaining BMI from a single 2D body image with the dual-branch regression framework proposed in this work. More specifically, the framework comprises an anthropometric feature computation branch and a deep learning-based feature extraction branch. One aggregation layer maps all the features to an estimated BMI value. In addition, a new public 2D image-to-BMI dataset, which contains 4189 images (1477 males and 2712 females) from approximately 3000 subjects with attributes including gender, age, height, and weight, was collected and released to facilitate the study. Extensive experiments confirm that the proposed framework combining anthropometric features and deep features outperforms the single-type feature approaches to BMI estimation in most cases.

Estimation of Potato Chlorophyll Content from UAV Multispectral Images with Stacking Ensemble Algorithm

Rapid and accurate crop chlorophyll content estimation is crucial for guiding field management and improving crop yields. This study explored the potential for potato chlorophyll content estimation based on unmanned aerial vehicle (UAV) multispectral imagery. To search the optimal estimation method, three parts of research were conducted as following. First, a combination of support vector machines (SVM) and a gaussian mixture model (GMM) thresholding method was proposed to estimate fractional vegetation cover (FVC) during the potato growing period, and the proposed method produced efficient estimates of FVC; among all the selected vegetation indices (VIs), the soil adjusted vegetation index (SAVI) had the highest accuracy. Second, the recursive feature elimination (RFE) algorithm was utilized to screen the VIs and texture features derived from multispectral images: three Vis, including modified simple ratio (MSR), ratio vegetation index (RVI) and normalized difference vegetation index (NDVI); three texture features, including correlation in the NIR band (corr-NIR), correlation in the red-edge band (corr-Red-edge) and homogeneity in the NIR band (hom-NIR), showed higher contribution to chlorophyll content estimation. Finally, a stacking model was constructed with K-Nearest Neighbor (KNN), a light gradient boosting machine (light-GBM), SVM algorithm as the base model and linear fitting as the metamodel, and four machine learning algorithms (SVM, KNN, light-GBM and stacking) were used to build the chlorophyll content estimation model suitable for different growing seasons. The results were: (1) The performance of the estimation model could be improved based on both VIs and texture features over using single-type features, and the stacking algorithm yielded the highest estimation accuracy with an R2 value of 0.694 and an RMSE value of 0.553; (2) When FVC was added, the estimation model accuracy was further improved, and the stacking algorithm also produced the highest estimation accuracy with R2 value of 0.739, RMSE value of 0.511 (3) When comparing modeling algorithms, stacking algorithms had greater advantages in the estimation chlorophyll content with potato plants than using single machine learning algorithms. This study indicates that taking into account the combination of VIs reflecting spectral characteristics, texture features reflecting spatial information and the FVC reflecting canopy structure properties can accomplish higher chlorophyll content estimation accuracy, and the stacking algorithm can integrate the advantages of a single machine learning model, with great potential for estimation of potato chlorophyll content.

Feature-level fusion based on spatial-temporal of pervasive EEG for depression recognition

Background and objectiveIn view of the depression characteristics such as high prevalence, high disability rate, high fatality rate, and high recurrence rate, early identification and early intervention are the most effective methods to prevent irreversible damage of brain function over time. The traditional method of depression recognition based on questionnaires and interviews is time-consuming and labor-intensive, and heavily depends on the doctor's subjective experience. Therefore, accurate, convenient and effective recognition of depression has important social value and scientific significance. MethodsThis paper proposes a depression recognition framework based on feature-level fusion of spatial-temporal pervasive electroencephalography (EEG). Time series EEG data were collected by portable three-electrode EEG acquisition instrument, and mapped to a spatial complex network called visibility graph (VG). Then temporal EEG features and spatial VG metric features were extracted and selected. Based on the correlation between features and categories, the differences in contribution of individual feature are explored, and different contribution coefficients are assigned to different features as the data basis of feature-level fusion to ensure the diversity of data. A cascade forest model based on three different decision forests is designed to realize the efficient depression recognition using spatial-temporal feature-level fusion data. ResultsExperimental data were obtained from 26 depressed patients and 29 healthy controls (HC). The results of multiple control experiments show that compared with single type feature, feature-level fusion without contribution coefficient, and independent classifiers, the feature-level method with contribution coefficient of spatial-temporal has a stronger recognition ability of depression, and the highest accuracy is 92.48%. ConclusionFeature-level fusion method provides an effective computer-aided tool for rapid clinical diagnosis of depression.

Cross-Subject Emotion Recognition Using Fused Entropy Features of EEG

Emotion recognition based on electroencephalography (EEG) has attracted high interest in fields such as health care, user experience evaluation, and human–computer interaction (HCI), as it plays an important role in human daily life. Although various approaches have been proposed to detect emotion states in previous studies, there is still a need to further study the dynamic changes of EEG in different emotions to detect emotion states accurately. Entropy-based features have been proved to be effective in mining the complexity information in EEG in many areas. However, different entropy features vary in revealing the implicit information of EEG. To improve system reliability, in this paper, we propose a framework for EEG-based cross-subject emotion recognition using fused entropy features and a Bidirectional Long Short-term Memory (BiLSTM) network. Features including approximate entropy (AE), fuzzy entropy (FE), Rényi entropy (RE), differential entropy (DE), and multi-scale entropy (MSE) are first calculated to study dynamic emotional information. Then, we train a BiLSTM classifier with the inputs of entropy features to identify different emotions. Our results show that MSE of EEG is more efficient than other single-entropy features in recognizing emotions. The performance of BiLSTM is further improved with an accuracy of 70.05% using fused entropy features compared with that of single-type feature.

DENet: Double-Encoder Network With Feature Refinement and Region Adaption for Terrain Segmentation in PolSAR Images

Recently, many studies exploit deep neural networks to promote terrain segmentation in polarimetric synthetic aperture radar (PolSAR) images. However, these works usually inherit the nature-scene approaches directly and may not be robust for the PolSAR image segmentation task. The main limitations include single-type feature construction, weak feature consistency, and geometry-agnostic collection of scattering information. In this article, we present the DENet, a double-encoder network with feature refinement and region adaption for the terrain segmentation in PolSAR images. First, a double-encoder architecture is proposed to leverage the multitype information of PolSAR images, which can provide more discriminative features than the previous methods using the single-type feature. Second, considering that the polarization information has strong consistency over the category-identical regions, a polarization-guided refinement module is proposed to maintain the feature consistency in the PolSAR segmentation model. This design alleviates the phenomenon of incomplete and fragmented segmentation results. Third, in view of the rich targets’ characteristics in the scattering information, a region-adaptive convolution module is developed to facilitate the scattering information collected over the geometry-irregular regions. This design can improve the segmentation accuracy on the geometry-irregular regions. Extensive experiments are conducted on six PolSAR images to verify the effectiveness of the DENet. Compared with the previous works, our method achieves competitive performance.

Application of UAV Multisensor Data and Ensemble Approach for High-Throughput Estimation of Maize Phenotyping Traits.

High-throughput estimation of phenotypic traits from UAV (unmanned aerial vehicle) images is helpful to improve the screening efficiency of breeding maize. Accurately estimating phenotyping traits of breeding maize at plot scale helps to promote gene mining for specific traits and provides a guarantee for accelerating the breeding of superior varieties. Constructing an efficient and accurate estimation model is the key to the application of UAV-based multiple sensors data. This study aims to apply the ensemble learning model to improve the feasibility and accuracy of estimating maize phenotypic traits using UAV-based red-green-blue (RGB) and multispectral sensors. The UAV images of four growth stages were obtained, respectively. The reflectance of visible light bands, canopy coverage, plant height (PH), and texture information were extracted from RGB images, and the vegetation indices were calculated from multispectral images. We compared and analyzed the estimation accuracy of single-type feature and multiple features for LAI (leaf area index), fresh weight (FW), and dry weight (DW) of maize. The basic models included ridge regression (RR), support vector machine (SVM), random forest (RF), Gaussian process (GP), and K-neighbor network (K-NN). The ensemble learning models included stacking and Bayesian model averaging (BMA). The results showed that the ensemble learning model improved the accuracy and stability of maize phenotypic traits estimation. Among the features extracted from UAV RGB images, the highest accuracy was obtained by the combination of spectrum, structure, and texture features. The model had the best accuracy constructed using all features of two sensors. The estimation accuracies of ensemble learning models, including stacking and BMA, were higher than those of the basic models. The coefficient of determination (R2) of the optimal validation results were 0.852, 0.888, and 0.929 for LAI, FW, and DW, respectively. Therefore, the combination of UAV-based multisource data and ensemble learning model could accurately estimate phenotyping traits of breeding maize at plot scale.

UAV Data as an Alternative to Field Sampling to Monitor Vineyards Using Machine Learning Based on UAV/Sentinel-2 Data Fusion

Pests and diseases affect the yield and quality of grapes directly and engender noteworthy economic losses. Diagnosing “lesions” on vines as soon as possible and dynamically monitoring symptoms caused by pests and diseases at a larger scale are essential to pest control. This study has appraised the capabilities of high-resolution unmanned aerial vehicle (UAV) data as an alternative to manual field sampling to obtain sampling canopy sets and to supplement satellite-based monitoring using machine learning models including partial least squared regression (PLSR), support vector regression (SVR), random forest regression (RFR), and extreme learning regression (ELR) with a new activation function. UAV data were acquired from two flights in Turpan to determine disease severity (DS) and disease incidence (DI) and compared with field visual assessments. The UAV-derived canopy structure including canopy height (CH) and vegetation fraction cover (VFC), as well as satellite-based spectral features calculated from Sentinel-2A/B data were analyzed to evaluate the potential of UAV data to replace manual sampling data and predict DI. It was found that SVR slightly outperformed the other methods with a root mean square error (RMSE) of 1.89%. Moreover, the combination of canopy structure (CS) and vegetation index (VIs) improved prediction accuracy compared with single-type features (RMSEcs of 2.86% and RMSEVIs of 1.93%). This study tested the ability of UAV sampling to replace manual sampling on a large scale and introduced opportunities and challenges of fusing different features to monitor vineyards using machine learning. Within this framework, disease incidence can be estimated efficiently and accurately for larger area monitoring operation.

Learning Compact Multifeature Codes for Palmprint Recognition From a Single Training Image per Palm

In this article, we propose a multifeature learning method to jointly learn compact multifeature codes (LCMFCs) for palmprint recognition with a single training sample per palm. Unlike most existing hand-crafted methods that extract single-type features from raw pixels, we first form the multi-type data vectors such as the direction-data, and texture-data to completely sample the multiple information of a palmprint image. Then, we learn the discriminative multifeatures from multi-type data vectors by maximizing the inter-palm distance, and minimizing the energy loss between the learned codes, and the original data. Moreover, our LCMFC method adaptively learns the optimal weights of multi-type features to jointly learn the compact multifeature codes. Finally, we cluster the nonoverlapping blockwise histograms of the compact multifeature codes into a feature vector for palmprint representation. Extensive experimental results on six benchmark palmprint databases are presented to show the effectiveness of the proposed method.

CLOUD CLASSIFICATION FOR GROUND-BASED SKY IMAGE USING RANDOM FOREST

Abstract. The use of solar power as a renewable energy has grown rapidly over the last few decades. However, the amount of solar radiation reaching the ground vary significantly in the short term. Clouds are the main factor. In this paper, a novel cloud detection method for ground-based sky images is proposed. First, the multiple features from the sky images, including spectral, texture and colour features are combined into a feature set. Then, Random Forest with this feature set is used to classify different types of cloud and clear sky. The experimental results show that cumulus and cirrus clouds can be identified from sky images. Combined with random forest, three types of features and various feature combinations are used for cloud classification, respectively. The classification accuracy with multiple features is higher than that of single-type features and dual-type features.

Incremental feature selection for dynamic hybrid data using neighborhood rough set

Feature selection with rough sets aims to delete redundant conditional features from static data by considering single type features. However, traditional feature selection methods generally ignore real-world scenarios: hybrid conditional feature set including missing, categorical and numerical ones coexists in the data, and the object set may change dynamically in one by one over time. To deal with dynamic hybrid data with mixed-type features, we propose a neighborhood entropy-based incremental feature selection framework by neighborhood rough set model. In this paper, the dynamics of an object set involves the change of a single object and multiple objects. Therefore, two incremental feature selection algorithms are developed for hybrid data with the dynamic change of a single object and multiple objects, respectively. At first, an incremental manner is utilized to compute the neighborhood entropy as feature criterion. On this basis, the incremental computations of feature significance are used to select candidate features in a descending order. Meanwhile, a deletion strategy is employed to filter out redundant features from the selection results. Finally, experimental results on different real-life data sets demonstrate the proposed incremental algorithms can outperform the non-incremental algorithm for feature selection in speed within comparable classification accuracy. Especially for multiple objects adding into and deleting from the hybrid data, the incremental algorithm is more efficient to select a subset of features than the algorithm for handling the dynamic change of a single object.

Multi-Type Interdependent Feature Analysis Based on Hybrid Neural Networks for Computer-Aided Diagnosis of Epidermal Growth Factor Receptor Mutations

The mutation status of the epidermal growth factor receptor (EGFR) is an important clinical reference indicator for lung cancer diagnosis and treatment. However, the extraction of effective discriminative features for non-invasive computer-aided EGFR mutation prediction still poses a big challenge. In this paper, multiple types of features are designed and analyzed to address this problem. These features include clinical features based on prior medical knowledge and quantitative image features extracted by convolutional neural networks (CNN). A long short-term memory (LSTM) network is also introduced to exploit the dependency between these feature types and then fuse them. In particular, a CNN is constructed to extract quantitative features of computed-tomography (CT) images. Furthermore, a LSTM is utilized to analyze the dependency between these clinical and CT image features and generate a new feature representation for computer-aided diagnosis. For samples from the same category, the proposed method deal with feature representation variabilities arising from interdependencies in multi-type features and patient specificity. The multiple feature types of the collected clinical data are used to assess the proposed approach and other relevant algorithms. Our results demonstrate that the multi-type dependency-based feature representation shows superior performance (Accuracy = 75%, AUC = 0.78) compared to single-type feature representations. The proposed method is reliable to apply for diagnosing of the EGFR mutation status.

On-line prediction of hazardous fungal contamination in stored maize by integrating Vis/NIR spectroscopy and computer vision

This work presents a fusion scheme to combine visible/near infrared spectroscopy and computer vision for on-line detection of Aspergillus spp. and Fusarium spp. contamination in stored maize. Spectroscopy and image information of 270 groups of maize kernels were collected at speed of 0.15m/s. Principal component analysis indicated fungi growth on maize could be monitored by both techniques. Spectroscopy method was found sensitive for infection level identification, while computer vision was more effective for fungal strain recognition. Linear discriminant analysis based on fusion of spectral and image features provided 100% accuracy for discrimination of samples infected by different strains after stored for 12d, which is at least 5.6% higher than single-type features. Classification rate of samples with different infection levels achieved 92.2%, also 5.5% and 10.0% higher than single technique. Moreover, data fusion improved colony counts prediction in samples by partial least squares regression, with root mean-square error of prediction value being reduced by 25.0% and 17.4%, respectively. This study demonstrated the superiority of data fusion for fungal detection in grain during on-line processing.

Towards feature representation for steganalysis of spatial steganography

Feature separability analysis supports the feature selection and reduction and also guides the feature construction. This paper addresses the feature representation for steganalysis of spatial steganography. The paper analyzes the effects of spatial steganography on image residual histogram, and gives the feature separability of histogram features represented in both spatial and frequential domains, which explains the effectiveness of the typical spatial features, such as SRM and TLBP features. Based on the feature separability analysis, TLBP feature is improved by using a new texture pattern mapping to replace the ‘riu2’ mapping in the feature construction. Besides, features in the frequency domain are constructed by applying DFT to the submodels of SRM. Experimental results show that both the two schemes improved the steganalysis performances. Furthermore, features in spatial and frequency domains are merged, which obtains better results than the corresponding single type features for steganalysis of S-UNIWARD, HILL, and MiPOD. For example, using HILL for information embedding with payload 0.4 bpp on BOSSbase dataset, the detection accuracy of the best merged feature is 3.64% and 1.13% higher than the corresponding single type features.

A Novel Feature Optimization for Wearable Human-Computer Interfaces Using Surface Electromyography Sensors.

The novel human-computer interface (HCI) using bioelectrical signals as input is a valuable tool to improve the lives of people with disabilities. In this paper, surface electromyography (sEMG) signals induced by four classes of wrist movements were acquired from four sites on the lower arm with our designed system. Forty-two features were extracted from the time, frequency and time-frequency domains. Optimal channels were determined from single-channel classification performance rank. The optimal-feature selection was according to a modified entropy criteria (EC) and Fisher discrimination (FD) criteria. The feature selection results were evaluated by four different classifiers, and compared with other conventional feature subsets. In online tests, the wearable system acquired real-time sEMG signals. The selected features and trained classifier model were used to control a telecar through four different paradigms in a designed environment with simple obstacles. Performance was evaluated based on travel time (TT) and recognition rate (RR). The results of hardware evaluation verified the feasibility of our acquisition systems, and ensured signal quality. Single-channel analysis results indicated that the channel located on the extensor carpi ulnaris (ECU) performed best with mean classification accuracy of 97.45% for all movement’s pairs. Channels placed on ECU and the extensor carpi radialis (ECR) were selected according to the accuracy rank. Experimental results showed that the proposed FD method was better than other feature selection methods and single-type features. The combination of FD and random forest (RF) performed best in offline analysis, with 96.77% multi-class RR. Online results illustrated that the state-machine paradigm with a 125 ms window had the highest maneuverability and was closest to real-life control. Subjects could accomplish online sessions by three sEMG-based paradigms, with average times of 46.02, 49.06 and 48.08 s, respectively. These experiments validate the feasibility of proposed real-time wearable HCI system and algorithms, providing a potential assistive device interface for persons with disabilities.

Kernel Embedding Multiorientation Local Pattern for Image Representation.

Local feature descriptor plays a key role in different image classification applications. Some of these methods such as local binary pattern and image gradient orientations have been proven effective to some extent. However, such traditional descriptors which only utilize single-type features, are deficient to capture the edges and orientations information and intrinsic structure information of images. In this paper, we propose a kernel embedding multiorientation local pattern (MOLP) to address this problem. For a given image, it is first transformed by gradient operators in local regions, which generate multiorientation gradient images containing edges and orientations information of different directions. Then the histogram feature which takes into account the sign component and magnitude component, is extracted to form the refined feature from each orientation gradient image. The refined feature captures more information of the intrinsic structure, and is effective for image representation and classification. Finally, the multiorientation refined features are automatically fused in the kernel embedding discriminant subspace learning model. The extensive experiments on various image classification tasks, such as face recognition, texture classification, object categorization, and palmprint recognition show that MOLP could achieve competitive performance with those state-of-the art methods.