No abstract

Thermal imaging is used in numerous applications, especially in security, medical and other industry which requires a non-contact temperature measurement. This proposed algorithm improves the thermal image and makes more visible the separation of the sampled object from its background. The extracted image is produced by the following techniques: the pre-processing techniques are the combination of K-means clustering, and inverted Otsu method; canny edge detection and color mapping are used for highlighting the necessary characteristics of the sampled thermal image. The experimental results of this proposed algorithm show significant distinguishable features in terms of edge and color enhancement. It outperforms the other color correction method in terms of processing time, and the implementation reduced the resource utilization. Moreover, it minimizes the misclassified pixel in different noise variance. This work is synthesized with Xilinx Zync 7000 ZED ZC702.