local binary pattern (LBP) and its changed forms have been widely used in the many applications of different fields such as image processing, computer vision, model recognition and texture analysis etc. In this paper,...
详细信息
Brain strokes are considered a worldwide medical emergency. In this paper, we present a new feature extractor that can classify brain computed tomography (CT) scan images into normal, ischemic stroke or hemorrhagic st...
详细信息
Brain strokes are considered a worldwide medical emergency. In this paper, we present a new feature extractor that can classify brain computed tomography (CT) scan images into normal, ischemic stroke or hemorrhagic stroke. The proposed feature extractor is based on comparing neighbours with the center pixel where diagonal neighbours are thresholded with the average intensity of whole image. The remaining horizontal and vertical pixels are obtained by thresholding them with their adjacent neighbour. Thereafter, binary values of the obtained images are used to generate the pattern code for the center pixel. Further, in such a way patter code is computed for whole image which is then used to generate 1-D feature vector known as a local neighbourhood pattern (LNP) descriptor by extracting quantum information from the image. The effectiveness of our feature extracted is proved by conducting experiments on real CT scan images of patients' brain. For experiments, we have taken nine different classifiers to identify efficacy obtained by extracting LNP features. We have also compared the results obtained by LNP with the results of local binary patterns (LBP) variants, local ternary patterns (LTP), local wavelet patterns (LWP) and local diagonal extrema patterns (LDEP) descriptors. All the experimental results demonstrate that the proposed feature descriptor gives high classification accuracy as compared to other state-of-the-art feature descriptors.
Structural health monitoring (SHM) systems provide opportunities to understand the structural behaviors remotely in real-time. However, anomalous measurement data are frequently collected from structures, which greatl...
详细信息
Structural health monitoring (SHM) systems provide opportunities to understand the structural behaviors remotely in real-time. However, anomalous measurement data are frequently collected from structures, which greatly affect the results of further analyses. Hence, detecting anomalous data is crucial for SHM systems. In this article, we present a simple yet efficient approach that incorporates complementary information obtained from multi-view local binary patterns (LBP) and random forests (RF) to distinguish data anomalies. Acceleration data are first converted into gray-scale image data. The LBP texture features are extracted in three different views from the converted images, which are further aggregated as the anomaly representation for the final RF pre-diction. Consequently, multiple types of data anomalies can be accurately identified. Extensive experiments validated on an acceleration dataset acquired on a long-span cable-stayed bridge highlight the advantages of the proposed method. State-of-the-art performances are achieved by the proposed method, demonstrating its effectiveness and generalization ability.
LBP is a successful hand-crafted feature descriptor in computer vision. However, in the deep learning era, deep neural networks, especially convolutional neural networks (CNNs) can automatically learn powerful task-aw...
详细信息
Finger vein images contain rich oriented features. local line binarypattern (LLBP) is a good oriented feature representation method extended from local binary pattern (LBP), but it is limited in that it can only extr...
详细信息
Finger vein images contain rich oriented features. local line binarypattern (LLBP) is a good oriented feature representation method extended from local binary pattern (LBP), but it is limited in that it can only extract horizontal and vertical line patterns, so effective information in an image may not be exploited and fully utilized. In this paper, an orientation-selectable LLBP method, called generalized local line binarypattern (GLLBP), is proposed for finger vein recognition. GLLBP extends LLBP for line pattern extraction into any orientation. To effectually improve the matching accuracy, the soft power metric is employed to calculate the matching score. Furthermore, to fully utilize the oriented features in an image, the matching scores from the line patterns with the best discriminative ability are fused using the Hamacher rule to achieve the final matching score for the last recognition. Experimental results on our database, MMCBNU_6000, show that the proposed method performs much better than state-of-the-art algorithms that use the oriented features and local features, such as LBP, LLBP, Gabor filter, steerable filter and local direction code (LDC).
Facial expressions are the first communication channel in human interactions. This channel helps us to understand people thoughts. The accuracy of the knowledge that have obtained by this channel benefits us in decisi...
详细信息
Melanoma is of the lethal and rare types of skin *** is curable at an initial stage and the patient can survive *** is very difficult to screen all skin lesion patients due to costly *** are requiring a correct method ...
详细信息
Melanoma is of the lethal and rare types of skin *** is curable at an initial stage and the patient can survive *** is very difficult to screen all skin lesion patients due to costly *** are requiring a correct method for the right treatment for dermoscopic clinical features such as lesion borders,pigment networks,and the color of *** challenges are required an automated system to classify the clinical features of melanoma and non-melanoma *** trained clinicians can overcome the issues such as low contrast,lesions varying in size,color,and the existence of several objects like hair,reflections,air bubbles,and oils on almost all *** contour is one of the suitable methods with some drawbacks for the segmentation of irre-gular *** entropy and morphology-based automated mask selection is pro-posed for the active contour *** proposed method can improve the overall segmentation along with the boundary of melanoma *** this study,features have been extracted to perform the classification on different texture scales like Gray level co-occurrence matrix(GLCM)and local binary pattern(LBP).When four different moments pull out in six different color spaces like HSV,Lin RGB,YIQ,YCbCr,XYZ,and CIE L*a*b then global information from different colors channels have been ***,hybrid fused texture features;such as local,color feature as global,shape features,and Artificial neural network(ANN)as classifiers have been proposed for the categorization of the malignant and *** had been carried out on datasets Dermis,DermQuest,and *** results of our advanced method showed super-iority and contrast with the existing state-of-the-art techniques.
With the development of convolutional neural networks, Deep Learning (DL) has also made great breakthroughs in the field of image inpainting. Deep Learning-based models can repair large-area missing defaced images, bu...
详细信息
This paper presents a novel method for face recognition under different illumination conditions. In the proposed approach, firstly, image is pre-processed to extract an approximation sub-band and several detail sub-ba...
详细信息
This paper presents a novel method for face recognition under different illumination conditions. In the proposed approach, firstly, image is pre-processed to extract an approximation sub-band and several detail sub-bands which are robust to illumination variation using Wavelet Transform in the logarithm domain, and then directly discards almost all low frequency information to decrease the influence of the illumination. Secondly, local binary pattern(LBP) is used to decrease the noise and extract the face feature. Finally, the result can be obtained by Support Vector Machine Classifiers. The proposed method is tested on CMU-PIE and Yale B face database, and high recognition rate prove that proposed method is robust to varying lighting conditions.
Automatic detection of raveled areas and categorization of their severity are crucial for assessing maintenance requirements and guaranteeing driving safety. This paper proposes and verifies a computer vision-based me...
详细信息
Automatic detection of raveled areas and categorization of their severity are crucial for assessing maintenance requirements and guaranteeing driving safety. This paper proposes and verifies a computer vision-based method for achieving these tasks. Advanced gradient boosting machines including adaptive boosting (AdaBoost), light gradient boosting machine (LightGBM), and extreme gradient boosting machine (XGBoost) are employed as pattern classifiers. To address the need to inspect large areas of pavement surfaces, this study also focuses on lightweight texture descriptors to enhance the productivity of the classification process. The employed texture descriptors are local binary pattern (LBP), center-symmetric local binary pattern (CSLBP), completed local binary pattern (CLBP), and local ternary pattern (LTP). These texture computation methods are selected due to their high performance for texture analysis, ease of implementation, and low computational cost. Experimental results supported by statistical tests point out that XGBoost coupled with CLBP achieves the most desired classification performance with Cohen's kappa coefficients > 0.94. Therefore, the proposed approach can be a promising tool to assist pavement management authorities in the task of surveying road surface conditions.
暂无评论