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

Landforms on the earth surface are so expensive to map or monitor. Remote Sensing observations from space platforms provide a synoptic view of terrain on images. Satellite multispectral data have an advantage in that the image data in various bands can be subjected to digital enhancement techniques for highlighting contrasts in objects for improving image interpretability. Geomorphological mapping involves the partitioning of the terrain into conceptual spatial entities based upon criteria. This paper illustrates how geomorphometry and mapping approaches can be used to produce geomorphological information related to the land surface, landforms and geomorphic systems. Remote Sensing application at Razzaza–Habbaria area southwest of Razz

This paper presents a complete design and implementation of a monitoring system for the operation of the three-phase induction motors. This system is built using a personal computer and  two types of sensors (current, vibration) to detect some of the mechanical faults that may occur in the motor. The study and examination of several types of faults including (ball bearing and shaft misalignment faults) have been done through the extraction of fault data by using fast Fourier transform (FFT) technique. Results showed that the motor current signature analysis (MCSA) technique, and measurement of vibration technique have high possibility in the detection and diagnosis of most mechanical faults with high accuracy. Subsequently, diagnosi

High-resolution imaging of celestial bodies, especially the sun, is essential for understanding dynamic phenomena and surface details. However, the Earth's atmospheric turbulence distorts the incoming light wavefront, which poses a challenge for accurate solar imaging. Solar granulation, the formation of granules and intergranular lanes on the sun's surface, is important for studying solar activity. This paper investigates the impact of atmospheric turbulence-induced wavefront distortions on solar granule imaging and evaluates, both visually and statistically, the effectiveness of Zonal Adaptive Optics (AO) systems in correcting these distortions. Utilizing cellular automata for granulation modelling and Zonal AO correction methods,

Deep Learning Techniques For Skull Stripping of Brain MR Images

Back ground: Diabetic nephropathy is rapidly becoming the leading cause of end-stage renal disease (ESRD). The onset and course of DN can be ameliorated to a very significant degree if intervention institutes at a point very early in the course of the development of this complication.
Objective: The aim of this study was to characterize risk factors associated with nephropathy in type I diabetes and construct a module for early prediction of diabetic nephropathy (DN) by analyzing their risk factors.
Methods: Case control design of 400 patients with type I diabetes mellitus (IDDM), aged 19-45 years. The cases were 200 diabetic patients with overt protein urea while the controls were 200 diabetic patients with no protein urea or micr

This paper studies the adaptive coded modulation for coded OFDM system using punctured convolutional code, channel estimation, equalization and SNR estimation. The channel estimation based on block type pilot arrangement is performed by sending pilots at every sub carrier and using this estimation for a specific number of following symbols. Signal to noise ratio is estimated at receiver and then transmitted to the transmitter through feedback channel ,the transmitter according to the estimated SNR select appropriate modulation scheme and coding rate which maintain constant bit error rate
lower than the requested BER. Simulation results show that better performance is confirmed for target bit error rate (BER) of (10-3) as compared to c