Design and Evaluation of Parallel Processing Techniques for 3D Liver Segmentation and Volume Rendering
Ensuring Software Quality in Engineering Environments
New 3D Face Matching Technique for an Automatic 3D Model Based Face Recognition System
Algorithmic Cost Modeling: Statistical Software Engineering Approach
Prevention of DDoS and SQL Injection Attack By Prepared Statement and IP Blocking
Face recognition systems are based on the premise that each individual has a distinct facial structure, and that by using facial symmetry, it is feasible to do computerized face matching. Since the 1960s, researchers have been working on facial recognition technology, and the results of their efforts are being utilised for security purposes in a variety of organisations and businesses all over the globe. There are a variety of algorithms available for face recognition. This article, which aims to analyze two of the face recognition method, will compare Eigenfaces and Local Binary Patterns Histogram (LBPH). Summing up the research, the accuracy of the Local Binary Patterns Histogram is higher than that of Eigenfaces, and the error rate is lower than that of Eigenfaces.
Low-light often leads to poor image visibility, which may have a substantial influence on the performance when using computer vision algorithms. The quality of low-light images can be improved by applying modified absorption light scattering model (ALSM). The reconstruction of hidden contours and features from a low-light image using an absorption light scattering picture obtained with ALSM is possible under appropriate and uniform illumination. To restrict the resemblance, superpixels may be employed as a measure. It is proposed that a mean-standard deviation (MSD) technique be used, which operates directly on patches and is depicted using superpixels. The MSD can achieve lower transmittance than the minimal technique, and it can be automatically adjusted in response to image information.
The human brain, which is made up of a white mass of cells, is the center of nervous system. A brain tumor is a collection of abnormally growing cells found in many parts of the brain, such as glial cells, neurons, lymphatic tissues, blood vessels, pituitary glands, and other parts of the brain, which leads to cancer. There are two forms of brain cancer. The first is benign, which is not cancerous and poses no threat; the second is malignant, which is a cancerous tumor that grows unnaturally, rapidly reproducing cells and eventually kills the individual if not identified. Manually detecting and identifying the tumor is more difficult. Program division method (PDM along with MRI (magnetic resonance imaging) can be used to discover and diagnose tumors. A powerful segmentation mechanism is required to provide precise results. In order to discover a patient's brain tumor, we look at their data, such as MRI images of their brain. The key concern is the segmentation, detection, and extraction of contaminated tumor areas from MR images, but this is a difficult and time-consuming task conducted by radiologists or clinical experts, and their accuracy is solely dependent on their experience. As a result, the employment of computer-assisted technology becomes increasingly important in order to overcome these constraints. This study uses a convolutional neural network to detect the type of brain tumor from MRI scans as input.
Currently, it is difficult to protect creative content and intellectual property due to technological advances. Digital media is incredibly easy to copy and manipulate, resulting in significant economic losses. As a result, digital watermarking is a feasible solution to this problem. Digital watermarking is a method of embedding a copyright mark into digital content that can be used to identify the original creator and owner of the media. It is widely used to track down copyright infringers. The original watermark image is compared with the extracted watermark image after many attacks to assess the robustness of the algorithm, and the Peak Signal to Noise Ratio (PSNR) and Normalized Correlation Coefficient (NC) are calculated. A new embedding technique for Discrete Wavelet Transform-based video watermarking is offered as an effective copyright protection algorithm. On the video, the Discrete Wavelet Transform (DWT) is used to translate the spatial data into the frequency domain, with low pass and high pass components. Using the watermark image and the binary Low frequency section (LL sub-band) of the video frame, the low frequency component is used to generate the key.
Hierarchical and Dynamic Elliptic Curve Cryptosystem (HiDE) is a self-certified public key method used in the security of medical data. HiDE offers a hierarchical cluster-based system consisting of a Backbone Cluster and numerous Area Clusters to support a high number of sensors. A Secure Access Point (SAP) in an Area Cluster receives medical data from Secure Sensors (SSs) in the sensor network and sends it to a Root SAP in the Backbone Cluster. As a result, the Root SAP may service a large number of SSs without having to create separate secure connections with each one. HiDE provides the Elliptic Curve Cryptosystem based Self-certified Public key scheme (ESP) for creating secure sessions between each pair of Cluster Head (CH) and Cluster Member (CM) to enable dynamic secure sessions for mobile SSs joining SAP (CM). Without knowing the CM's secret key, the CH may issue a public key to a CM and calculate a Shared Session Key (SSK) with that CM in ESP. This approach meets the Zero Knowledge Proof, allowing CHs to construct secure sessions with CMs without having to manage the CM's secret keys. According to our research in realistic implementations and network simulations, ESP requires fewer computational and network resources than the Rivest-Shamir-Adleman (RSA) based public key method. Furthermore, security research reveals that the keys in ESP are properly safeguarded. HiDE can therefore maintain sufficient network performance for wireless sensor networks while protecting the confidentiality of critical medical data with minimum processing cost.
