Question about how merge patchs work

[bisnow33]

Hi,
Thank for your nice work !
I have a question about how you compute the average in the merge_patchs function. I'm not sure to understand how you proceed when I look at your code. In those following lines you sum value of different patches rigth ? and why you divide value by 2 ? :

elif dims == 4:
                    if rgb:
                        empty_data[indice[0]:indice[1], indice[2]:indice[3], :] = np.where(empty_data[indice[0]:indice[1], indice[2]:indice[3], :] == 0,
                                                                                            data_patches[i], 
                                                                                            np.add(data_patches[i],empty_data[indice[0]:indice[1], indice[2]:indice[3], :])/2)

If I'm after the sum where do you divide by the number of by the number of occurrences ?

Best,

Tristan