Vidya Jyothi Institute of Technology (VJIT) is a college in Hyderabad, India, established in 1999 by A.P. Jithender Reddy and V. Purushottam Reddy. It is located on the way to Chilkur Balaji Temple from Mrugavani National Park.
This study investigates the mechanical, tribological, microstructural, and physical characteristics of Al7075–Si3N4 composites reinforced with Si3N4 particles derived from red matta rice husk. Four composite specimens B1, B2, B3, and B4 were developed with reinforcement levels of 1 vol
Paraffin wax (PW) is widely used as a phase change material (PCM) for latent heat thermal energy storage, however its low thermal conductivity limits practical performance. Consequently, researchers have extensively investigated nanoparticle enhanced PCMs to improve heat transfer characteristics. In this study, copper (Cu) nanoparticle enhanced paraffin composite PCMs (CPCMs) were prepared with Cu loadings of 5, 10, and 15 mass
Freeform surfaces are vital in aerospace, biomedical, and die/mold industries, where dimensional accuracy directly impacts performance and reliability. Reverse engineering workflows are highly sensitive to scan density, noise reduction, and reconstruction parameters, yet current practice relies on trial-and-error tuning, increasing inspection cost and uncertainty. A CNC-milled Al6061 freeform test component was fabricated and scanned under 64 parameter combinations generated using a Taguchi-based orthogonal design. Root Mean Square (RMS) deviation between CAD and scan models was quantified to establish a performance baseline. Three machine learning models, namely, Support Vector Regression (SVR), Gaussian Process Regression (GPR), and eXtreme Gradient Boosting (XGBoost) were developed to predict RMS deviation directly from reverse engineering parameters. Model hyperparameters were optimized via grid search with five-fold cross-validation. GPR achieved the highest predictive accuracy (Testing R2=0.9901, RMSE=0.0215, MAPE=4.95%), while SVR demonstrated superior robustness across folds (R2 range: 0.7717-0.9782). XGBoost captured complex nonlinearities but showed larger fold variability and outlier errors, with MAPE exceeding 20% in select cases. The proposed machine learning methodology enables tolerance-aware, first-time-right scanning of freeform components by replacing heuristic parameter tuning with predictive modeling. GPR is recommended for accuracy-driven applications, while SVR provides a stable alternative for robustness-focused scenarios, significantly advancing industrial freeform metrology planning.
Image gradient estimation is the utmost step in most image processing applications such as edge detection, feature extraction and other higher-level vision tasks, but derivative based operators also tend to increase impulse noise, resulting in unsteady gradient fields, and spurious edge responses. The proposed work presents a gradient stabilization model which utilizes the Gaussian pre-smoothing that follows gradient computation to eliminate noise extremity whilst maintaining structural data. The method is assessed with a salt and pepper noise which is controlled with a multi-scale Gaussian strategy (σ = 1.0, 1.5, 2.0) to determine the trade-off between denoising performance and feature preservation. Experimental results clearly reveal that both the quality of perceptions and gradient stability are improved substantially. Peak Signal-to-Noise Ratio (PSNR) rises to 29.07 dB at the higher levels of smoothing, compared to 18.03 dB of the noisy image and Structural Similarity (SSIM) rises to 0.8305 from 0.3073. The gradient variance (GV) declines to 0.0192 from 0.0445, which means that edge structures are better stabilized. The proposed method attained its highest possible Edge Preservation Index (EPI) of 0.7997 and the Gradient Signal-to-Noise Ratio (GSNR) of 2.38 among scales. The best trade-off between noise suppression and structural fidelity is obtained with σ = 1.5. The investigational outcomes undoubtedly illustrates that the Gaussian pre-smoothing process remarkably improves the robustness of gradientbased feature detection performance in impulse noise degraded imaging environs.
An innovative software-driven solution is designed to optimize traffic flow and expedite the transit of emergency vehicles through congested urban areas. The system leverages advanced machine learning models for both audio and image processing to dynamically adjust traffic signals and prioritize the swift passage of emergency vehicles. The audio processing module utilizes a 1D convolutional neural network (1DCNN) and the Librosa library to detect ambulance sirens in real-time audio data, triggering the activation of the visual processing module. The visual module, powered by the YOLOv5 object detection framework, identifies ambulances in live video feeds, prompting immediate adjustments to traffic signals. Extensive testing demonstrates the system’s efficacy in prioritizing emergency vehicles, with key metrics including signal transition times, ambulance detection accuracy, and overall traffic flow improvement. The holistic integration of audio and visual processing and dynamic traffic signal adjustment positions this solution as a robust and adaptable approach to addressing challenges in emergency vehicle navigation within urban traffic scenarios.