Performance Evaluation of Clustering Based Tone Mapping Operators with State-of-Art Methods

Tone mapping operators have a major role in visualizing the real time images. There exist various methods, the performance of the selected tone mapping operators is compared with Novel clustering-based Tone mapping Operators. Performance evaluation is done by calculating its objective metrics. The performance of existing TMOs which comes in the categories of global, local, and clustering based are evaluated and compared with Novel clustering-based techniques. We measured more than 100 high dynamic images of indoor and outdoor natural scenes. Real time images are taken at different places of Hyderabad, Bahrain and Las Vegas using Nikon D5100 camera. Few images are taken from open sources. The quality metrics such as Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), Mean Absolute Error (MAE), modified Peak Signal to Noise Ratio (mPSNR) , Tone Mapped Quality Index (TMQI), Feature Similarity Index for Tone Mapped Image (FSITM) and execution time are calculated and compared with existing tone mapping operator (TMO).