22‐4: A Novel Color Temperature Prediction Algorithm by Machine Learning

In this paper, we propose using a machine learning algorithm to predict correlated color temperature (CCT) with in‐cell ambient color sensors. Machine learning algorithms can more effectively identify the relationship between sensor data and color temperature values. Using this method, we can reduce the maximum prediction error from 250 to 150 across five light sources ranging from CCT 2300K to 6500K.