Convolutional Neural Networks and Impact of Filter Sizes on Image Classification

Abstract

Deep Neural Networks (DNN) in the past few years have revolutionized the computer vision by providing the best results on a large number of problems such as image classification, pattern recognition, and speech recognition. One of the essential models in deep learning used for image classification is convolutional neural networks. These networks can integrate a different number of features or so-called filters in a multi-layer fashion called convolutional layers. These models use convolutional, and pooling layers for feature abstraction and have neurons arranged in three dimensions: Height, Width, and Depth. Filters of 3 different sizes were used like 3×3, 5×5 and 7×7. It has been seen that the accuracy on the training data has been decreased from 100% to 97.8% as we increase the filter size and also the accuracy on the test data set decreases for 3×3 it is 98.7%, for 5×5 it is 98.5%, and for 7×7 it is 97.8%. The loss on the training data and test data per 10 epochs could be seen drastically increasing from 3.4% to 27.6% and 12.5% to 23.02%, respectively. Thus it is clear that using the filters having lesser dimensions is giving less loss than those having more dimensions. However, using the smaller filter size comes with the cost of computational complexity, which is very crucial in the case of larger data sets.

Effect of filter sizes on image classification in CNN: a case study on CFIR10 and Fashion-MNIST datasets

“…These advancements helped machines to perceive image and video recognition, natural language processing, and much more. Deep learning is enormously perfected with time, primarily over convolutional neural networks (CNN), and gives much better results than the others [4]- [7]. CNN is now a well-established machine learning tool used for image classification problems, especially in medical sciences and practical life like detecting road signs, recognizing human activity, and facial expression recognition [8], [9].…”

Toward an embedded system for gesture recognition based on artificial neural network using reconfigurable target (case study and review)

“…In recent years Artificial Neural Network (ANN) has been growing to address the challenges of the computational complexity, time, power and resources consuming. In addition, ANN achieved a remarkable performance, and it has been used in many applications related to a variety of industries such as, motion planning mobile robots and image classification [7]. In this context, the article presents a Reference Signal Generator (RSG) for the purpose of training an ANN for gesture recognition [8], [9].…”