Analytical study of two feature extraction methods in comparison with deep learning methods for classification of small metal objects

Abstract

This paper addresses the efficiency of two feature extraction methods for classifying small metal objects including screws, nuts, keys, and coins: the histogram of oriented gradients (HOG) and local binary pattern (LBP). The desired features for the labeled images are first extracted and saved in the form of a feature matrix. Using three different classification methods (non-parametric K-nearest neighbors algorithm, support vector machine, and naïve Bayesian method), the images are classified into four different classes. Then, by examining the resulting confusion matrix, the performances of the HOG and LBP approaches are compared for these four classes. The effectiveness of these two methods is also compared with the “You Only Look Once” and faster region-based convolutional neural network approaches, which are based on deep learning. The collected image set in this paper includes 800 labeled training images and 180 test images. The results show that the use of the HOG is more efficient than the use of the LBP. Moreover, a combination of the HOG and LBP provides better results than either alone.

Full-Resolution Lung Nodule Localization From Chest X-Ray Images Using Residual Encoder-Decoder Networks

“…Deep neural networks, especially the Convolutional Neural Network (CNN) architecture proposed by [48] and first adapted to computer vision applications by [49], have excelled at many computer vision tasks including automated medical image analysis [50]. Deep learning CNNs automatically combine feature extraction and feature classification into a single trainable algorithm [51], thereby eliminating hand-crafted feature extraction steps in the pipeline.…”