an experiment study on optimal batch size of u-net convolutional neural network for edge detection

The u-net architecture, initially designed for biomedical image segmentation, can be repurposed for edge detection tasks by reconfiguring the network’s focus. this paper investigates the impact of batch size, a critical hyperparameter in u-net, on the network’s performance for edge detection. we conduct experiments with three different image sizes and varying batch sizes for each image size. by analyzing the trade-offs between accuracy and stability, we identify the optimal batch size that enhances the u-net model’s performance in edge detection tasks. our study contributes valuable insights into effectively configuring batch size to improve u-net’s performance in edge detection applications.