Deconvolutional networks are convolutional neural networks (CNN) that work in a reversed process. Deconvolutional networks, also known as deconvolutional neural networks, are very similar in nature to CNNs run in reverse but are a distinct application of artificial intelligence (AI).
Deconvolutional networks strive to find lost features or signals that may have previously not been deemed important to a convolutional neural network’s task. A signal may be lost due to having been convoluted with other signals. The deconvolution of signals can be used in both image synthesis and analysis.
A convolutional neural network emulates the workings of a biological brain’s frontal lobe function in image processing. A deconvolutional neural network constructs upwards from processed data. This backwards function can be seen as a reverse engineering of convoluted neural networks, constructing layers captured as part of the entire image from the machine vision field of view and separating what has been convoluted.
Deconvolutional networks are related to other deep learning methods used for the extraction of features from hierarchical data such as that found in deep belief networks and hierarchy-sparse automatic encoders. Deconvolutional networks are primarily used in scientific and engineering fields of study.