Fig. 4: Influence of state dataset list size on DRLAF trained by randomly sampling performance. | Microsystems & Nanoengineering

Fig. 4: Influence of state dataset list size on DRLAF trained by randomly sampling performance.

From: Precision autofocus in optical microscopy with liquid lenses controlled by deep reinforcement learning

Fig. 4

a Impact of state dataset list size on the distribution of automatic focusing deviation. As the number of state datasets increases, the focusing deviation of the model significantly decreases. b Effect of state dataset list size on automatic focusing success rate. With an increase in the number of state datasets, both the accuracy and success rate of automatic focusing notably improve. c Influence of state dataset list size on automatic focusing time steps and accuracy. The trend shows that as the number of state datasets increases, the accuracy of automatic focusing gradually improves, but the average step exhibits a trend of initially decreasing and then increasing. Appropriate sizes of state dataset lists can be configured based on practical requirements

Back to article page