Impact of Tuning Parameters in Deep Convolutional Neural Network Using a Crack Image Dataset

TL;DR

This study investigates how different tuning parameters affect the performance of a deep convolutional neural network in crack image classification, highlighting optimal configurations for accuracy.

Contribution

It provides an experimental analysis of parameter tuning effects on DCNN performance using a crack image dataset, which is less explored in existing literature.

Findings

Max pooling improves performance.

Adam optimizer yields better results.

Tanh activation function is more effective.

Abstract

The performance of a classifier depends on the tuning of its parame ters. In this paper, we have experimented the impact of various tuning parameters on the performance of a deep convolutional neural network (DCNN). In the ex perimental evaluation, we have considered a DCNN classifier that consists of 2 convolutional layers (CL), 2 pooling layers (PL), 1 dropout, and a dense layer. To observe the impact of pooling, activation function, and optimizer tuning pa rameters, we utilized a crack image dataset having two classes: negative and pos itive. The experimental results demonstrate that with the maxpooling, the DCNN demonstrates its better performance for adam optimizer and tanh activation func tion.