Abstract

This study presents an analysis of the causes of insufficient efficiency of the nearest neighbor method, compared with deep learning networks. The primary cause is the incorrect use of Euclidean distance to the nearest neighbor for estimating the distance from the analyzed pattern to a region occupied by a class. To overcome this problem, it is necessary to construct a local estimate of the distance metric. Therefore, the proposed method combines the concepts of “tangential distance” and Mahalanobis distance. Based on this method, a modified nearest neighbor method is suggested. Validation experiments on MNIST data show that the new method reduces the recognition error rate from 3.8% to 0.8%, which is lower than the recognition error rate of 1.1% for the nearest neighbor method modified based on the concept of “tangential distance,” the calculation of which requires a priori information on allowable pattern transformations.

© 2017 Optical Society of America

Evaluation of the use of the Hopfield neural network model as a nearest-neighbor algorithm

Bruce L. Montgomery and B. V. K. Vijaya Kumar
Appl. Opt. 25(20) 3759-3766 (1986)

Testing different classification methods in airborne hyperspectral imagery processing

Vladimir V. Kozoderov and Egor V. Dmitriev
Opt. Express 24(10) A956-A965 (2016)

Flame state recognition method of a scramjet based on PLIF image fusion features and an artificial neural network

Long Gao, Jiangbo Peng, Xin Yu, Zhen Cao, Minghong Han, Guohua Wu, and Xun Yuan
Opt. Continuum 3(3) 338-353 (2024)

Nearest-neighbor median filter

Kazuyoshi Itoh, Yoshiki Ichioka, and Tatsuya Minami
Appl. Opt. 27(16) 3445-3450 (1988)

Event recognition method based on dual-augmentation for a Φ-OTDR system with a few training samples

Yi Shi, Shangwei Dai, Xinyu Liu, Yingchao Zhang, Xinjie Wu, and Tao Jiang
Opt. Express 30(17) 31232-31243 (2022)