Year: 2018
Type: article
Abstract: Image recognition, in the context of machine vision, is the ability of the software to identify objects, places, people, writing and actions in images.Computers can use machine vision technologies in combination with a camera and artificial intelligence software to achieve the task of image recognition.Image recognition is used to perform a large number of machine-based visual tasks, such as labeling the contents of images, performing image content search for guiding autonomous robots, self-driving cars and accidental avoidance system.While human brains recognize objects easily, computers have difficulty with the task.Software for image recognition requires deep machine learning.Performance is based on the complexity of convolutional neural network as the specific task requires massive amount of computational power for its computer-intensive nature.This work will review "CIFAR-10" dataset which has classified images in various groups.This problem is a supervised learning task which will be able to classify any new images put forward from these various groups.This work also attempts to provide an insight into "You Only Look Once (YOLO)" which is an example of unsupervised image classification.It can immediately classify the images into various objects by drawing rounded boxes around them and naming those objects.
Author: N.S. Lele
Institution: Savitribai Phule Pune University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2907586929