Search

Advance in technology world has lots of contributions from artificial intelligence which is a highly growing area. The failure of traditional algorithms has led to the employment of deep learning algorithms in various fields like pattern recognition, recommendation systems and classification systems. Removal of noise from images can be done using traditional noise removal filters. These filters can either remove more noise that wanted or leave unwanted noise than what is needed in the data. Utilization of Convolutional neural networks designed based on the dataset requirements along with the noise removal filter can yield better results. In this work, evaluation of the performance of convolutional neural network (CNN) against existing image denoising algorithms has been successfully executed . The proposed model is a generalized CNN model which can recognize and classify any type of noisy image given. Two types of model were compared where one model 1 uses the Adam optimizer and model 2 uses the Stochastic Gradient Descent (SGD) optimizer. The image dataset used here is MNIST handwritten dataset, which is trained, tested and validated with both the models by adding three different types of noise viz, Poisson, Salt and Pepper as well as Gaussian Noise. More accuracy and better results were given by the model 2 which uses the SGD optimizer.

This paper presents a study on the line heating simulation process with an efficient "finite element analysis (FEA)" simulation model. The simulation model is conceived using concepts from three dimensional (3D) FEA of heat and mass transfer, plate theory and geometry, and implemented with commercially available software (Ansys*™ Mechanical APDL). The source of thermal energy is assumed to be "Gaussian" distributed heat flux, and we consider the temperature dependency of the material properties. The computer simulation model (CSM) is validated using the published experimental results. We study the discretization requirements for thick plates (>14 mm) that are commonly used in the shipbuilding processes and suggest guidelines for referential use in the industry. Also, we investigate the minimum time step size requirement during the cooling phase of the simulation process and based on the experimental numerical simulations, few efficient simulation practices have been coined, and we show that if they are followed then a noticeable gain in reduction of computational cost and time can be attained. We report extensive simulations of the estimated lines and then compare them with the target shape, and show a close agreement. Finally, our presented results are utilized to derive techniques for the improvement in existing FEA simulation models, and we show that the presented CSM is efficient for the application in ships and offshore structures for high bending of thick plates. 1. Introduction In ship and offshore building industries, geometric shapes are generated by forming of metal plates or sheets into simple or complex curvatures as demanded by the design requirements. Although, the doubly curved structures are used in offshore structures (e.g., spherical tanks and facades), in general, the offshore structures have planar, low curvature formations and developable surfaces, and the ships have a combination of developable and nondevelopable surfaces, low and high curvatures, and planar and nonplanar formations. In engineering sciences, a surface can either be "single curved (developable surface)" or "doubly curved (nondevelopable)" or a combination of them in various regions, e.g., doubly curved—bulbous bow and stern bulb, single curved—parallel middle body, combination—in the zones of transitions. In metal based industries, a single-curved surface can be manufactured with simple mechanical processes—bending, rolling, pressing, and/or a combination of them. For doubly curved surface, after the process of rolling and/or pressing, complex mechanical processes are needed:either the plate is shrunk along its longer edges using line/curve/pattern heating,or the plate is pressed against a base shape restricting its movement and then applying the high forces by pressing.