Convolution Confusion

Robert, a computer vision specialist working for an auto manufacturer, is frustrated because he can't seem to get the output size of his convolutional layers right.

He is designing a convolutional neural network, and the first layer operates on grayscale images of size 32×32. Robert is using a convolutional layer with a kernel of size 5, no padding, and a stride of 2.

What is the size of the output of the convolutional layer?