In electrical equipment excessive temperature leads to failures, therefore thermal monitoring plays a major role. For preventing equipment failures, thermal imaging cameras are routinely used for monitoring the condition of electrical equipment. Initial cost for installing thermal imaging cameras is very expensive. This paper introduces the extraction of thermal images from a low-cost web camera. The webcam modified in such a way that IR filter is removed, and a color filter is introduced to get more precise thermal image data. Captured image is in black and white format. Temperature is measured manually, and a software introduced to compute corresponding the RGB values. Regression technology based on Artificial Neural Network (ANN) is used to predict the exact temperature of the corresponding thermal image. A PC speaker is utilized to announce the predicted output in audio format.