General Background: Deep image matting is a fundamental task in computer vision, enabling precise foreground extraction from complex backgrounds, with applications in augmented reality, computer graphics, and video processing. Specific Background: Despite advancements in deep learning-based methods, preserving fine details such as hair and transparency remains a challenge. Knowledge Gap: Existing approaches struggle with accuracy and efficiency, necessitating novel techniques to enhance matting precision. Aims: This study integrates deep learning with fusion techniques to improve alpha matte estimation, proposing a lightweight U-Net model incorporating color-space fusion and preprocessing. Results: Experiments using the AdobeComposition-1k

This paper presents a modified training method for Recurrent Neural Networks. This method depends on the Non linear Auto Regressive (NARX) model with Modified Wavelet Function as activation function (MSLOG) in the hidden layer. The modified model is known as Modified Recurrent Neural (MRN). It is used for identification Forward dynamics of four Degrees of Freedom (4-DOF) Selective Compliance Assembly Robot Arm (SCARA) manipulator robot. This model is also used in the design of Direct Inverse Control (DIC). This method is compared with Recurrent Neural Networks that used Sigmoid activation function (RS) in the hidden layer and Recurrent Neural Networks with Wavelet activation function (RW). Simulation results shows that the MRN model is bett

 

This work implements the face recognition system based on two stages, the first stage is feature extraction stage and the second stage is the classification stage. The feature extraction stage consists of Self-Organizing Maps (SOM) in a hierarchical format in conjunction with Gabor Filters and local image sampling. Different types of SOM’s were used and a comparison between the results from these SOM’s was given.

The next stage is the classification stage, and consists of self-organizing map neural network; the goal of this stage is to find the similar image to the input image. The proposal method algorithm implemented by using C++ packages, this work is successful classifier for a face database consist of 20

Gold nanoparticles AuNPs have proven to be powerful tools in various nanomedicine applications, because of their photo-optical distinctiveness and biocompatibility. Noble metal gold nanoparticles was prepared by pulsed laser ablation method (1064-Nd: YAG with various Laser power from 200 to 800 mJ and 1 Hz frequency) in distil water. The process was characterized using UV-VIS absorption spectroscopy. Morphology and average size of nanoparticles were estimated using AFM and X-ray diffraction (XRD) analysis which show the nature of gold nanoparticles (AuNPs). Antibacterial activity of gold nanoparticles as a function of particles concentration against gram negative bacterium Escherichia coli and gram positive bacterial Staphylococcus aureu

In this work we investigate and calculate theoretically the variation in a number of optoelectronic properties of AlGaAs/GaAs quantum wire laser, with emphasis on the effect of wire radius on the confinement factor, density of states and gain factor have been calculated. It is found that there exist a critical wire radius (rc) under which the confinement of carriers are very weak. Whereas, above rc the confinement factor and hence the gain increase with increasing the wire radius.

Contours extraction from two dimensional echocardiographic images has been a challenge in digital image processing. This is essentially due to the heavy noise, poor quality of these images and some artifacts like papillary muscles, intra-cavity structures as chordate, and valves that can interfere with the endocardial border tracking. In this paper, we will present a technique to extract the contours of heart boundaries from a sequence of echocardiographic images, where it started with pre-processing to reduce noise and produce better image quality. By pre-processing the images, the unclear edges are avoided, and we can get an accurate detection of both heart boundary and movement of heart valves.

Modeling forward kinematics with neural networks allows for efficient handling of nonlinear relationships and realistic error correction in time-critical applications by relying on accurate training data. This paper presents a Multi-Layer Feed-Forward Neural Network (MLFFNN) to solve the forward kinematics of a 3-DOF robot. The proposed MLFFNN consists of 50 hidden neurons and was trained using 628319 samples to find only the position (x, y, z) of the end-effector. Data were generated by MATLAB, assuming an incremental motion of joints. The joint variables ( , , and ) are the inputs of the NN, which outputs the positions of the end effector (x, y, z) calculated using the Denavit-Hartenberg (DH) method. The results demonstrate that t