Cyberbullying has become a widespread concern in our society, and its detection has grown in significance. In recent years, machine learning algorithms have exhibited considerable potential in detecting cyberbullying....
详细信息
A star sensor takes the stars as the reference sources for attitude measurement to output the star vectors in the coordinate system of the star sensor, providing high-precision measurement data for the attitude contro...
详细信息
Violations of the roadway leads to a decrease in the level of driver safety and the condition of the car, reduces the speed of movement. Violations include cracks, pits, ruts, the size of which can change every day, d...
详细信息
The development of computer vision systems stimulates the development of various applications in the field of image recognition. Methods and algorithms for image recognition in document processingsystems play a cruci...
The development of computer vision systems stimulates the development of various applications in the field of image recognition. Methods and algorithms for image recognition in document processingsystems play a crucial role. There is a lot of research in this direction, and one of the most difficult problems is handwritten text recognition. This includes the tasks of signature identification, which the authors of the paper consider. The article considers the main stages of signature analysis to establish authorship. They are the preprocessing of the signature, representation in a form convenient for applying author identification. A dataset containing sample signatures of various people is used. The described tools became means to develop software for preprocessingimages of signatures, extracting characteristics and identifying the author of the signature.
It is known that information security and privacy revolve around three principal keys: confidentiality, integrity, and availability. Depending on the environment, application, context, or use case, one of these princi...
详细信息
ISBN:
(纸本)9781665427357
It is known that information security and privacy revolve around three principal keys: confidentiality, integrity, and availability. Depending on the environment, application, context, or use case, one of these principles may prevail over the others. For a medical application, for example, confidentiality is essential. Therefore, any imaging transferred electronically must be encrypted to prevent unauthorized persons from seeing its contents. This paper exhibits a new efficient ciphering algorithm for the security of transmitted images in public networks such as the Internet or archived in cloud computing based on a three-dimensional S-box. The suggested method encrypts and decrypts the image block by block. Each block is 8x8 pixels. The encrypted data are obtained by replacing the original data and the key by its value in S-box. Substitution boxes (S-box) are generated using a proposed algorithm that produces a 3D S-box using the random function and permutation. The security analyses show that the presented algorithm has an important security level. The runtime of the proposed algorithms is fast compared to others image encryption algorithms.
Face recognition is used in a wide variety of applications such as surveillance systems, human-computer interaction, automatic door access control systems, and network security. One of the policies of the smart univer...
详细信息
ISBN:
(数字)9781665485104
ISBN:
(纸本)9781665485104
Face recognition is used in a wide variety of applications such as surveillance systems, human-computer interaction, automatic door access control systems, and network security. One of the policies of the smart university is to adopt technology to help with teaching and learning, especially during the Covid-19 pandemic. In this paper, a smart attendance system using face recognition algorithms with deep learning is proposed and used in the university's classroom. Instead of calling names to confirm the identity of students, our system does it automatically. The system was tested in 3 scenarios, namely, in online classes, in on-site classes, and in problematic cases using a standard dataset. The performances of the 3 scenarios were compared in the experiment in terms of precision, recall, F1 score, and percentage accuracy. Our result revealed that in online classes the recognition accuracy is as high as 100%. The implemented system is inexpensive and practical. The application can be used on any portable device such as tablets or smartphones. History viewing, multiple subjects handling, and file exporting features are also incorporated into the system.
Inspection in photovoltaic generation systems is of great importance for maintenance and corresponding generation efficiency. The combined use of unmanned aerial vehicles and deep learning computational algorithms can...
详细信息
High Dynamic Range (HDR) imaging has become a significant technological advancement in visual data processing, allowing for the capture of a wider dynamic range of luminance levels in images. This paper explores vario...
详细信息
ISBN:
(数字)9798331529505
ISBN:
(纸本)9798331529512
High Dynamic Range (HDR) imaging has become a significant technological advancement in visual data processing, allowing for the capture of a wider dynamic range of luminance levels in images. This paper explores various HDR processing techniques and their potential applications in automation and machine vision. By using methods such as multiple image fusion, image registration, and tone mapping, the paper demonstrates how HDR processing can enhance visual data in automated systems, improving accuracy in environments requiring complex lighting conditions. This work applies HDR algorithms to real-world scenarios, showcasing their potential in industrial automation and robotics, where accurate visual data plays a crucial role.
The paper discusses objects (images) segmentation algorithms that are applicable in mobile applications with augmented reality. An example of imageprocessing with virtual objects by different algorithms (the MeanShif...
详细信息
The paper discusses objects (images) segmentation algorithms that are applicable in mobile applications with augmented reality. An example of imageprocessing with virtual objects by different algorithms (the MeanShift algorithm, the GrabCut algorithm, the k- means algorithm) is considered. Different libraries, tools, and environments to implement segmentation algorithms were analyzed, such as Scikit-image, Pixellib, OpenCV, Point Cloud Library. The application was created for mobile devices running iOS 10 and higher. The GrabCut algorithm turned out to be the best algorithm for imageprocessing. The processing result was the closest to the expected one. Although the algorithm has some errors. Despite the fact that the area that was contoured turned out to be the clearest and most complete in comparison with other algorithms, this area also includes areas of the image that do not belong to the objects under study.
image segmentation is critical to object-oriented imageprocessing. Many conventional segmentation algorithms are based on the superpixel, since it integrates the pixels with similar colors and locations in prior and ...
详细信息
ISBN:
(纸本)9781665464956
image segmentation is critical to object-oriented imageprocessing. Many conventional segmentation algorithms are based on the superpixel, since it integrates the pixels with similar colors and locations in prior and is beneficial for segmentation. Recently, several segmentation algorithms based on deep learning were developed. However, due to the irregular shape and size of superpixels, it is hard to apply the superpixel directly in a leaning-based segmentation algorithm. In this paper, we propose a novel segmentation method that well integrates the techniques of the deep neural network (DNN), the superpixel, adaptive loss functions, and multi-layer feature extraction. First, different from other learning-based algorithm, which applies an image or its bounding boxes as the input, we adopt the mean and the histogram differences of the features of two superpixels as the input of the DNN to determine whether they should be merged. Moreover, to well consider both largescaled and small-scaled features, a hierarchical architecture is adopted. For different layers, the DNN models with different loss functions are applied. A larger penalty for over-merging is applied in the first layer and a larger penalty for oversegmentation is applied in the following layer. Moreover, according to human perception, the features of colors, areas, the gradient at the boundary, and the texton, which is highly related to the texture, are applied. Experiments show that the proposed method outperforms other state-of-the-art image segmentation methods and produces highly accurate segmentation results.
暂无评论