Development and Application of a Universal Filter in Image Processing for Automatic Interpretation of Temperature-Sensing Thermal Paints

Abstract

Machines like Internal Combustion Engines and gas turbines work on the principle of converting the heat energy into mechanical energy. Every effort is taken to elevate the operating temperature which in turn increases the output efficiency. However, the engine components cannot withstand such high temperatures and their life is seriously affected. The true gradient of the temperature to which these engine components are exposed is a wealth of knowledge for an engine designer. Conventional thermometry has got many limitations from measuring and mounting point of view and is unable to give the exact thermal gradient. Advancement in thermal mapping technique is required to enhance the development of these components. Comparatively thermal paints provide a better alternative to be used as temperature sensors. These paints change their color permanently when exposed to temperature. A proper calibration and automatic interpretation of these paints using digital image processing gives a reliable thermal contour. During the engine testing the carbon exhaust from the engine blackens the color of the thermal paint applied on the various components as it flows across it creating an obstacle in the color pattern (temperature) recognition. This article highlights the efforts taken in development and application of a filter which removes the effect of the carbon soot to denoise the degraded image and recover the original image to get the required temperature profile.

Analysis of Image Noise Reduction Using Neural Network

“…(1) The input image to process in Figure explain f(x,y) and the processed image is g(x,y) and T is an operator on f, `defined over some neighborhood of (x, y) In addition, T can operate on a set of input images, such as performing the pixel-by-pixel and the principal approach in defining a neighborhood about a point (x, y) is to use a square or rectangular sub image area centered at (x, y) [ [4], shown in Figure 1. This operation uses only the pixels in the area of the image expanded by the neighborhood in spite of other neighborhood shapes , precisely a circle sometimes are used square and rectangular arrays which is more popular because of their ease of implementation and the simplest form where T is when the neighborhood is of size 1x1 (that is, a single pixel) In this case, g depends only on the value of f at (x, y), and T becomes a gray-level in transformation function of the form [5]:…”

“…Though, noise remained to be data in image. Thus, loud activity may be referred to as noisy [1,4]. 1) Adaptive Gaussian Noise.…”

A Color Restoration Method for Irreversible Thermal Paint Based on Atmospheric Scattering Model

“…These paints can change their colors according to the peak working temperatures and will not change back after it cools down, thus providing an off-line temperature measurement. They can be applied to the large area components or complex surface shapes and do not interfere with the thermal behavior [1].…”