Performance Evaluation of various Enhancement Algorithm on Thermal Images by Implementation on PYNQ-Z2 Board

This paper presents an efficient image enhancement technique based on detail guided image filtering algorithm implemented on the Python Productivity for ZYNQ (PYNQ-Z2) board, a flexible platform for developing real-time applications in diverse domains. The proposed work utilizes detail guided image filtering (DGIF) method to enhance details of thermal image while preserving essential structures in the thermal image. It provides a flexible environment using the PYNQ-Z2 for controlling the enhancement process by improving the visual appeal and identifying key elements in an image. Performance is evaluated by comparing the techniques with the other variants such as Global histogram equalization (GHE), Contrast limited adaptive histogram equalization (CLAHE), and Brightness Preserving Bi-Histogram Equalization (BBHE). For the quality assessment, computing the parameter such as PSNR, SSIM, and EDGE SCORE along with the computation time. It is observed that the guided image filtering algorithm for the thermal image enhancement and its hardware implementation on the PYNQ-Z2 board, offering faster processing compared to traditional software-based approaches, significantly reducing computational time. Application of this thermal image enhancement boosts surveillance systems by delivering more explicit images, better detection, and enhanced analysis, improving security and more efficient monitoring across various application.