Early detection of diseases in crops for human consumption is crucial for agricultural production in the world. Nowadays, Deep learning algorithms are used for the analysis and classification of images such as the convolutional neural network used to detect diseases in plants. Unfortunately, these algorithms still need a lot of computation to obtain results that could take hours, days, or maybe weeks for a prediction. This paper aims to provide an analysis of how the Wavelet transform in distributed platforms can provide competitive results with a fraction of the resources used comparatively in time and memory used to detect diseases in tomato leaves.