Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks
The relevance of the tasks of detecting and recognizing objects in images and their sequences has only increased over the years. Over the past few decades, a huge number of approaches and methods for detecting both anomalies, that is, image areas whose characteristics differ from the predicted ones,...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Samara National Research University
2022-02-01
|
Series: | Компьютерная оптика |
Subjects: | |
Online Access: | https://computeroptics.ru/eng/KO/Annot/KO46-1/460117e.html |
_version_ | 1797867051748950016 |
---|---|
author | N.A. Andriyanov V.E. Dementiev A.G. Tashlinskiy |
author_facet | N.A. Andriyanov V.E. Dementiev A.G. Tashlinskiy |
author_sort | N.A. Andriyanov |
collection | DOAJ |
description | The relevance of the tasks of detecting and recognizing objects in images and their sequences has only increased over the years. Over the past few decades, a huge number of approaches and methods for detecting both anomalies, that is, image areas whose characteristics differ from the predicted ones, and objects of interest, about the properties of which there is a priori information, up to the library of standards, have been proposed. In this work, an attempt is made to systematically analyze trends in the development of approaches and detection methods, reasons behind these developments, as well as metrics designed to assess the quality and reliability of object detection. Detection techniques based on mathematical models of images are considered. At the same time, special attention is paid to the approaches based on models of random fields and likelihood ratios. The development of convolutional neural networks intended for solving the recognition problems is analyzed, including a number of pre-trained architectures that provide high efficiency in solving this problem. Rather than using mathematical models, such architectures are trained using libraries of real images. Among the characteristics of the detection quality assessment, probabilities of errors of the first and second kind, precision and recall of detection, intersection by union, and interpolated average precision are considered. The paper also presents typical tests that are used to compare various neural network algorithms. |
first_indexed | 2024-04-09T23:34:07Z |
format | Article |
id | doaj.art-8f1e4ec2d4e2460f87a842d60e890bcb |
institution | Directory Open Access Journal |
issn | 0134-2452 2412-6179 |
language | English |
last_indexed | 2024-04-09T23:34:07Z |
publishDate | 2022-02-01 |
publisher | Samara National Research University |
record_format | Article |
series | Компьютерная оптика |
spelling | doaj.art-8f1e4ec2d4e2460f87a842d60e890bcb2023-03-20T15:33:54ZengSamara National Research UniversityКомпьютерная оптика0134-24522412-61792022-02-0146113915910.18287/2412-6179-CO-922Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networksN.A. Andriyanov0V.E. Dementiev1A.G. Tashlinskiy2Financial University under the Government of the Russian FederationUlyanovsk State Technical UniversityUlyanovsk State Technical UniversityThe relevance of the tasks of detecting and recognizing objects in images and their sequences has only increased over the years. Over the past few decades, a huge number of approaches and methods for detecting both anomalies, that is, image areas whose characteristics differ from the predicted ones, and objects of interest, about the properties of which there is a priori information, up to the library of standards, have been proposed. In this work, an attempt is made to systematically analyze trends in the development of approaches and detection methods, reasons behind these developments, as well as metrics designed to assess the quality and reliability of object detection. Detection techniques based on mathematical models of images are considered. At the same time, special attention is paid to the approaches based on models of random fields and likelihood ratios. The development of convolutional neural networks intended for solving the recognition problems is analyzed, including a number of pre-trained architectures that provide high efficiency in solving this problem. Rather than using mathematical models, such architectures are trained using libraries of real images. Among the characteristics of the detection quality assessment, probabilities of errors of the first and second kind, precision and recall of detection, intersection by union, and interpolated average precision are considered. The paper also presents typical tests that are used to compare various neural network algorithms.https://computeroptics.ru/eng/KO/Annot/KO46-1/460117e.htmlpattern recognitionobject detectioncomputer visionimage processingrandom fieldscnnioumapprobability of correct detection |
spellingShingle | N.A. Andriyanov V.E. Dementiev A.G. Tashlinskiy Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks Компьютерная оптика pattern recognition object detection computer vision image processing random fields cnn iou map probability of correct detection |
title | Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks |
title_full | Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks |
title_fullStr | Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks |
title_full_unstemmed | Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks |
title_short | Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks |
title_sort | detection of objects in the images from likelihood relationships towards scalable and efficient neural networks |
topic | pattern recognition object detection computer vision image processing random fields cnn iou map probability of correct detection |
url | https://computeroptics.ru/eng/KO/Annot/KO46-1/460117e.html |
work_keys_str_mv | AT naandriyanov detectionofobjectsintheimagesfromlikelihoodrelationshipstowardsscalableandefficientneuralnetworks AT vedementiev detectionofobjectsintheimagesfromlikelihoodrelationshipstowardsscalableandefficientneuralnetworks AT agtashlinskiy detectionofobjectsintheimagesfromlikelihoodrelationshipstowardsscalableandefficientneuralnetworks |