Now a day's there is a significant increase in the in the duplicate copies of large original images. One of the main reason for such duplication is due to the availability of large number of image editing software...
详细信息
The pooling layer used in CNN models aims to reduce the resolution of image/feature maps while retaining their distinctive information, reducing computation time and enabling deeper models. Max and average pooling met...
详细信息
The pooling layer used in CNN models aims to reduce the resolution of image/feature maps while retaining their distinctive information, reducing computation time and enabling deeper models. Max and average pooling methods are frequently used in CNN models due to their computational efficiency;however, these methods discard the position information of the pixels. In this study, we proposed an LBP-based pooling method that generates a neighborhood-based output for any pixel, reflecting the correlation between pixels in the local area. Our proposed method reduces information loss since it considers the neighborhood and size of the pixels in the pooling region. Experimental studies were performed on four public datasets to assess the effectiveness of the LBP pooling method. In experimental studies, a toy CNN model and various transfer learning models were utilized in conducting test operations. The proposed method provided improvements of 1.56% for Fashion MNIST, 0.22% for MNIST, 3.95% for CIFAR10, and 5% for CIFAR100 dataset using the toy model. In the experimental studies conducted using the transfer learning model, performance improvements of 6.99(-/+)(0.74) and 8.3(-/+)(0.1) were achieved for CIFAR10 and CIFAR100, respectively. We observed that the proposed method outperforms the commonly used pooling layers in CNN models. Code for this paper can be publicly accessed at: https://***/cuneytozdemir/lbppooling
作者：
Kamezawa, HidemiArimura, HidetakaTeikyo Univ
Fac Fukuoka Med Technol Dept Radiol Technol 6-22 Misaki Machi Omuta Fukuoka 8368505 Japan Kyushu Univ
Fac Med Sci Dept Hlth Sci Div Med Quantum Sci 3-1-1 Maidashi Higashi Ku Fukuoka 8128582 Japan
We investigated an approach for predicting recurrence after radiation therapy using local binary pattern (LBP)-based dosiomics in patients with head and neck squamous cell carcinoma (HNSCC). Recurrence/non-recurrence ...
详细信息
We investigated an approach for predicting recurrence after radiation therapy using local binary pattern (LBP)-based dosiomics in patients with head and neck squamous cell carcinoma (HNSCC). Recurrence/non-recurrence data were collected from 131 patients after intensity-modulated radiation therapy. The cases were divided into training (80%) and test (20%) datasets. A total of 327 dosiomics features, including cold spot volume, first-order features, and texture features, were extracted from the original dose distribution (ODD) and LBP on gross tumor volume, clinical target volume, and planning target volume. The CoxNet algorithm was employed in the training dataset for feature selection and dosiomics signature construction. Based on a dosiomics score (DS)-based Cox proportional hazard model, two recurrence prediction models (DSODD and DSLBP) were constructed using the ODD and LBP dosiomics features. These models were used to evaluate the overall adequacy of the recurrence prediction using the concordance index (CI), and the prediction performance was assessed based on the accuracy and area under the receiver operating characteristic curve (AUC). The CIs for the test dataset were 0.71 and 0.76 for DSODD and DSLBP, respectively. The accuracy and AUC for the test dataset were 0.71 and 0.76 for the DSODD model and 0.79 and 0.81 for the DSLBP model, respectively. LBP-based dosiomics models may be more accurate in predicting recurrence after radiation therapy in patients with HNSCC.
Machine Learning (ML) has been widely used for Image processing. The pertinent feature extraction and feature selection techniques can help us to accomplish many complex tasks. This paper presents a framework for the ...
详细信息
Machine Learning (ML) has been widely used for Image processing. The pertinent feature extraction and feature selection techniques can help us to accomplish many complex tasks. This paper presents a framework for the classification of emotions using ML. Training and testing have been done using the JAFFE (Japanese Female Facial Expression) dataset. The work proposes a combination of Short-Time Fourier Transform (STFT) and local binary pattern (LBP) for extracting interesting features. Also, a fusion of popular feature reduction techniques namely: Fisher Discriminant Ratio (FDR), variance threshold method and chi-square test has been introduced. The selected relevant features are applied to the Support Vector Machine (SVM) classifier. Performance analysis of the existing techniques and the proposed technique has been carried out where the latter was found efficient. The proposed pipeline performs better in terms of accuracy, specificity and sensitivity as compared to the state of art.
Nowadays, face biometric based access control systems are becoming ubiquitous in our daily life while they are still vulnerable to spoofing attacks. So developing robust and reliable methods to prevent such frauds is ...
详细信息
Nowadays, face biometric based access control systems are becoming ubiquitous in our daily life while they are still vulnerable to spoofing attacks. So developing robust and reliable methods to prevent such frauds is unavoidable. As deep learning techniques have achieved satisfactory performances in computer vision, they have also been applied to face spoofing detection. However, the numerous parameters in these deep learning based detection methods cannot be updated to optimum due to limited data. local binary pattern (LBP), effective features for face recognition, have been employed in face spoofing detection and obtained promising results. Considering the similarities between LBP extraction and convolutional neural network (CNN) that the former can be accomplished by using fixed convolutional filters, we propose a novel end-to-end learnable LBP network for face spoofing detection. Our network can significantly reduce the number of network parameters by combing learnable convolutional layers with fixed-parameter LBP layers that are comprised of sparse binary filters and derivable simulated gate functions. Compared with existing deep leaning based detection methods, the parameters in our fully connected layers are up to 64x savings. Conducting extensive experiments on two standard spoofing databases, i.e., Relay-Attack and CASIA-FA, our proposed LBP network substantially outperforms the state-of-the-art methods.
In this paper, we address a real-time object tracking algorithm considering local binary pattern (LBP) as a feature descriptor. In addition to texture feature, Ohta color features are included in the feature vector of...
详细信息
To investigate the robustness of face recognition algorithms under the complicated variations of illumination, facial expression and posture, the advantages and disadvantages of seven typical algorithms on extracting ...
详细信息
To investigate the robustness of face recognition algorithms under the complicated variations of illumination, facial expression and posture, the advantages and disadvantages of seven typical algorithms on extracting global and local features are studied through the experiments respectively on the Olivetti Research Laboratory database and the other three databases （the three subsets of illumination, expression and posture that are constructed by selecting images from several existing face databases）. By taking the above experimental results into consideration, two schemes of face recognition which are based on the decision fusion of the twodimensional linear discriminant analysis （2DLDA） and local binary pattern （LBP） are proposed in this paper to heighten the recognition rates. In addition, partitioning a face nonuniformly for its LBP histograms is conducted to improve the performance. Our experimental results have shown the complementarities of the two kinds of features, the 2DLDA and LBP, and have verified the effectiveness of the proposed fusion algorithms.
In this chapter, local binary pattern (LBP) based extreme learning machine (ELM) is presented for identification of high-dimensional face images of different resolution. In this scheme, LBP is used for the representat...
详细信息
Facial recognition technology allows both the government and the general public to identify individuals based on their facial features, even in cases of significant changes. However, low lighting during the facial rec...
详细信息
Recently, the use of artificial intelligence to improve the efficiency of Covid-19 diagnosis has become a trend due to the spread and proliferation of Covid-19 and the fact that healthcare professionals alone are no l...
详细信息
暂无评论