The tracking of torsional eye movements, where the eye rotates about the axis of vision, is important for the diagnosis of certain vestibular disorders, but presents additional challenges when compared to the tracking...
详细信息
ISBN:
(数字)9798331518776
ISBN:
(纸本)9798331518783
The tracking of torsional eye movements, where the eye rotates about the axis of vision, is important for the diagnosis of certain vestibular disorders, but presents additional challenges when compared to the tracking of linear eye movements. This paper proposes a method for the measurement of ocular torsion that combines techniques from several previous studies, attempting to make the measurement robust to a highly off-axis centre of vision and obscurations of the pupil and iris. The proposed method detects feature points across a region of interest (ROI) including both the iris and the sclera (the white part of the eye surrounding the iris), using the Affine-SIFT (ASIFT) feature detection method, and tracks the feature points across frames using the Lucas-Kanade Optical Flow Method. The proposed method was only able to achieve an average error of 0.42 degrees compared to 0.17 degrees achieved in prior research. However, an additional 23.5% of feature points were able to be tracked compared to feature matching on the iris alone, showing that the proposed method has the potential to provide more accurate tracking than conventional methods. Future improvements proposed to the method include a deep segmentation method to automatically determine the ROI in each frame, and further testing with a calibration step to determine the source of error in the method.
Agricultural productivity is the factor on which many countries' economies heavily rely on. Identifying plant diseases is extremely crucial in the agricultural sector as they can hamper the plant's robustness ...
详细信息
A Dynamic SSE (Searchable symmetric encryption) scheme should support updates and keyword searches with outsourced symmetrically encrypted data while minimizing the amount of data revealed to the untrusted server. For...
详细信息
ISBN:
(数字)9798350375442
ISBN:
(纸本)9798350375459
A Dynamic SSE (Searchable symmetric encryption) scheme should support updates and keyword searches with outsourced symmetrically encrypted data while minimizing the amount of data revealed to the untrusted server. For this purpose, literature in this field has identified two critical security properties, i.e. forward and backward *** this work, we build upon the previously developed "Oblivious Dynamic Cross Tags" (ODXT) scheme for conjunctive keyword searches, featuring forward and backward privacy. The objective is to improve flexibility in the update operation of the given scheme, solve the open problem of s-term leakage, and perform leakage analysis of the improved scheme. We propose flexODXT, to reduce the time required for individual keyword update operations to a constant instead of linear complexity, and we propose suppODXT, to reduce the s-term leakage, an open problem found in ODXT.
Class imbalance and incompleteness are the two most serious problems faced in data science and machine learning when working on real-life datasets. Both of these cases have severe implications on the ability of classi...
详细信息
Class imbalance and incompleteness are the two most serious problems faced in data science and machine learning when working on real-life datasets. Both of these cases have severe implications on the ability of classification algorithms to make accurate predictions. When a dataset used for training classifiers is both imbalanced as well as incomplete, the traditional approach is to address the missing data first and then handle class imbalance but it could lead to some issues such as overfitting as well as amplification of some errors due to random duplication. In this paper, an alternate factor-based multiple imputation oversampling method (FB-MIO) is proposed to handle class imbalance as well as missing values in the training dataset at the same time. First, a new factor is presented to evaluate the density of missing values belonging to the majority class with respect to the minority class in a particular region. With the help of this factor, an oscillator is developed to guide how imputation based oversampling should be carried out. Then the training set is divided into multiple smaller subsets and used the oscillator to determine whether missing values for the majority class belonging to that subsets should be imputed or not. This would help in preventing exaggerated duplication when not needed. Experiments were carried out on 27 imbalanced datasets after random addition of missing value and the F1 and AUROC scores of FB-MIO were compared to other dataset level resampling methods such as SMOTE, ADASYN, B-SMOTE etc. The effectiveness of the proposed method has been validated after experiments on benchmark datasets and the comparative results are presented in the form of average rank and number of wins.
This review investigates how deep learning methods can be utilized for efficient image retrieval based on content. Obtaining accurate images from vast digital collections poses significant challenges, motivating resea...
详细信息
ISBN:
(数字)9798350378092
ISBN:
(纸本)9798350378108
This review investigates how deep learning methods can be utilized for efficient image retrieval based on content. Obtaining accurate images from vast digital collections poses significant challenges, motivating research in CBIR. The effectiveness of these methods varies depending on the dataset’s type and size, with certain algorithms excelling with specific dataset characteristics. An extensive and well-structured review of successful image retrieval techniques is given in this paper. Our primary objective is to evaluate various deep learning models applied by researchers and compare their performance based on the outcome of evaluation matrices. These models encompass CNNs, DBNs, and other deep architectures tailored for image retrieval tasks. By synthesizing insights from this review, researchers can make informed decisions regarding model selection and potentially enhance retrieval performance by leveraging advanced deep learning features. The importance of deep features in image retrieval is the ability to capture complex visual patterns and semantic information that cannot be easily extracted by traditional handcrafted features. With the increasing volume of online images, Image retrieval using deep learning has become crucial for applications like object recognition, image retrieval, and image search engines.
Synthetic Reduced Nearest Neighbor (SRNN) models, operating exclusively on synthetic samples or prototypes, represent a significant stride in the field of nearest-neighbor algorithms. Central to this innovation is enh...
详细信息
ISBN:
(数字)9798350373752
ISBN:
(纸本)9798350373769
Synthetic Reduced Nearest Neighbor (SRNN) models, operating exclusively on synthetic samples or prototypes, represent a significant stride in the field of nearest-neighbor algorithms. Central to this innovation is enhancing the inter-pretability and optimization of the model, achieved through specialized techniques. This study introduces a novel Two-Layer Neural-SRNN model for classification tasks, diverging from traditional Expectation Maximization (EM) methodologies. The TLN-SRNN model significantly advances efficiency and scalability, outperforming traditional methods in speed while maintaining accuracy. Our empirical findings highlight the model's rapid convergence and robust performance across diverse datasets, establishing it as a notable innovation in the field of machine learning.
Working with large scale microservice based applications may be hard to maintain and control. The services in those applications must be isolated from each other and they must work in harmony. There must be an intelli...
详细信息
This paper presents a comprehensive investigation of 5nm double gate MOSFETs (DG MOSFETs) circuits using different substrate semiconductor materials for low-power devices. A comparative analysis of characteristics of ...
详细信息
ISBN:
(数字)9798350374643
ISBN:
(纸本)9798350374650
This paper presents a comprehensive investigation of 5nm double gate MOSFETs (DG MOSFETs) circuits using different substrate semiconductor materials for low-power devices. A comparative analysis of characteristics of SiGe, InGaAs and GaN as substrate materials is performed. The study reveals that InGaAs offers superior performance due to its higher electron mobility, lower threshold voltage and thermal stability than other materials studied here for low-power applications. The electrical properties of DG MOSFETs on these substrates have been investigated by calculating key parameters such as threshold voltages (Vth), Ion and Ioff currents. The results show that InGaAs-based DG MOSFETs exhibit superior performance with threshold voltage, improved leakage current and Ion/Ioff ratio compared to other substrate materials. Additionally, a lower threshold voltage has been observed for InGaAs-based devices, indicating reduced power consumption. These findings demonstrate the potential of InGaAs as a preferred substrate material for low-power, high-performance DG MOSFET devices. Double gate MOSFET circuits are designed using InGaAs and power dissipation is calculated. The results show a significant reduction in power consumption and improved circuit performance compared to traditional Si-based devices. This research demonstrates the potential of InGaAs-based DG MOSFETs for low-power electronic devices, enabling the development of energy-efficient circuits for various applications.
Inventory management is very important in any industry. In the past, people were directly involved in inventory management. However, this method is time consuming and requires a lot of labor. Therefore, in recent year...
详细信息
Water bodies are essential component of Earth’s surface and accurate water region mapping is crucial for disaster management and environmental monitoring. Existing methods to map water bodies relies on traditional mo...
详细信息
ISBN:
(数字)9798350369175
ISBN:
(纸本)9798350369182
Water bodies are essential component of Earth’s surface and accurate water region mapping is crucial for disaster management and environmental monitoring. Existing methods to map water bodies relies on traditional modelling methods that need extensive manual calibration struggle with scalability and lack advanced pattern recognition. This paper evaluates the effectiveness of deep learning-based semantic segmentation models, DeepLabV3+ and U-Net, using aerial images resized to $\mathbf{2 5 6 x} 256$ pixels in mapping water regions in aerial images. DeepLabV3+, with a ResNet50 backbone, and U-Net are trained on a comprehensive flood-affected dataset with preprocessing techniques such as resizing and augmentation. Model performance is assessed using metrics like IoU, F1 score, accuracy, precision, and recall. The results indicate that DeepLabV3+ outperforms U-Net, showing promise for automating water region mapping and improving flood surveys and urban planning.
暂无评论