In a Convolutional Neural Network, the operation helps detect spatial hierarchies in the input image.

The operation that helps detect spatial hierarchies in the input image in a Convolutional Neural Network is the Convolution operation.

Here's a step-by-step explanation:

1. Convolution operation: This is the first step in a Convolutional Neural Network. The input image is convolved with a filter or kernel (a smaller-sized matrix) to generate a feature map or convolved feature. This operation is applied to the input data using a convolution filter to produce a feature map.

2. ReLU (Rectified Linear Unit) operation: After the convolution operation, the feature map may contain negative values. To remove these, the ReLU operation is applied which replaces all negative pixel values in the feature map with zero.

3. Pooling or Sub Sampling: This operation reduces the dimensionality of each feature map but retains the most important information. Pooling can be of different types: Max, Average, Sum etc.

4. Fully Connected Layer: After several convolutional, ReLU and pooling layers, the high-level reasoning in the neural network is done via fully connected layers. Neurons in a fully connected layer have connections to all activations in the previous layer.

These operations help in detecting spatial hierarchies in the input image by moving from simple to complex patterns.