Performance Improvisation and Optimization Strategies for High Resolution Image Interpolation

Abstract

Image interpolation, also referred to as sampling, is a technique used to estimate new pixel values based on the surrounding pixel values in an image, resulting in a higher sampling rate. In the context of security, surveillance plays a vital role in ensuring human safety and survival, utilizing various types of cameras. However, images obtained from surveillance may suffer from low resolution due to several factors: insufficient lighting in the area of interest, a significant distance between the subject and the camera, and limited resolution of the surveillance camera. Consequently, addressing these challenges is crucial for improving the effectiveness of security monitoring systems This research addresses the challenges of low resolution images in surveillance, by proposing novel image interpolation algorithms aimed at enhancing the quality of low resolution images during the transformation to high resolution. newlineThis research introduces four innovative image interpolation methods to overcome low resolution challenges: Quantized Batch-Gradient Sharp-Edge Interpolation (Q-BG-SE), Artificial Neural Network (ANN) Quadratic Interpolator, Deep Multi-Scaled Residual Network (DMResNet), and a deep learning-based model, combining the Optimized Recursive Least Square Adaptive Filter (ORLSAF) and the Multi-Variate Dense Fusion Network (MVDFN) with Hybrid Butterfly Optimization (HBO) algorithm. newlinevi newlineThis research introduces a novel interpolation algorithm named Quantized Batch-Gradient Sharp-Edge Interpolation (Q-BG-SE). This algorithm aims to enhance the quality of image edges during the transformation of low resolution (LR) images to high resolution (HR) images, improving the overall image quality. newlineThis research introduces an innovative approach known as the Artificial Neural Network (ANN) Quadratic Interpolator, designed for the interpolation of 3-D images. The method utilizes a combination of Lagrange interpolating polynomial and Lagrange interpolating basis function within the parameter space.