Bipolar Morphological Neural Networks: Convolution Without Multiplication

TL;DR

This paper introduces bipolar morphological neural networks that use simple operations like addition, subtraction, and maximum, enabling faster inference suitable for mobile and embedded systems, with comparable accuracy to classical CNNs.

Contribution

The paper proposes a novel bipolar morphological neuron and layer model that approximate classical computations without multiplication, along with training methods that do not require special algorithms.

Findings

Moderate accuracy decrease after converting CNN layers to bipolar morphological layers

Models enable faster inference suitable for mobile and embedded systems

Effective layer-by-layer training approach demonstrated

Abstract

In the paper we introduce a novel bipolar morphological neuron and bipolar morphological layer models. The models use only such operations as addition, subtraction and maximum inside the neuron and exponent and logarithm as activation functions for the layer. The proposed models unlike previously introduced morphological neural networks approximate the classical computations and show better recognition results. We also propose layer-by-layer approach to train the bipolar morphological networks, which can be further developed to an incremental approach for separate neurons to get higher accuracy. Both these approaches do not require special training algorithms and can use a variety of gradient descent methods. To demonstrate efficiency of the proposed model we consider classical convolutional neural networks and convert the pre-trained convolutional layers to the bipolar morphological layers. Seeing that the experiments on recognition of MNIST and MRZ symbols show only moderate decrease of accuracy after conversion and training, bipolar neuron model can provide faster inference and be very useful in mobile and embedded systems.