International Research Journal of Education and Technology

Thermal imaging has become a vital tool in fields such as medical diagnosis, industrial inspection, and environmental monitoring due to its ability to capture and analyze temperature variations. This project aims to develop a system for extracting first-order statistical features from thermal images to analyze the intensity distribution of pixels. First-order features, including mean, variance, standard deviation, skewness, kurtosis, energy, and entropy, provide valuable insights into the thermal properties of objects and environments. The system will preprocess thermal images by normalizing pixel values, reducing noise, and resizing them for consistent analysis. Statistical measures will be computed directly from the pixel intensity values to reveal patterns and anomalies. The extracted features will be validated using benchmark datasets to ensure accuracy and consistency. By comparing statistical features across different thermal images, the system aims to identify anomalies, patterns, and temperature variations that could support improved decision-making in medical diagnostics, predictive maintenance, and environmental analysis. The project’s outcomes will enhance the understanding of thermal properties, improve fault detection, and support early detection of medical conditions through detailed statistical analysis of thermal data

Key Words: First-Order features, skewness, kurtosis, pixel values, anomalies, decision making in medical diagnostics