e-ISSN:0976-5166
p-ISSN:2231-3850


INDIAN JOURNAL OF COMPUTER SCIENCE AND ENGINEERING

Call for Papers 2020

Dec 2020 - Volume 11, Issue 6
Deadline: 15 Nov 2020
Publication: 20 Dec 2020

Feb 2021 - Volume 12, Issue 1
Deadline: 15 Jan 2021
Publication: 20 Feb 2021

More

Indexed in

IJCSE Indexed in Scopus

ABSTRACT

Title : RETRIEVING STEREO IMAGE USING VISUAL WORD PAIRS WITH CONVOLUTION NEURAL NETWORK
Authors : Rachakonda Swetha, Ganta Raghotham Reddy
Keywords : Image Retrieval; Visually word pairs; Stereo Images and Convolutional Neural Network (CNN).
Issue Date : Sep-Oct 2020
Abstract :
The exponential increase and omnipresent access to visual data on the Internet contributed to the growth of image search or recovery science. Ignoring the visual content as a ranking indicator can result in inconsistencies between the words and the visual content with the selection technique. The failure of current methods and characteristics of HrosSIs therefore does not provide a reasonable retrieval of images. In addition, nothing of the recent remote sensing image retrieval techniques takes the basic features of high-resolution optical satellite stereo images (HrosSIs) such as the display number and the viewing angles into account. The article offers a specific structure to leverage the unique features of HrosSIs data in such a way that HrosSI's information is collected efficiently and accurately. The recovery of HrosSI is achieved through the similitude among the structures attained from DSMs and orthographs, together mined from the HrosSIs. Moreover, the value of pinnacle evidence stood examined for HrosSI recovery. For process authentication spending the ISPRS stereo test dataset, a prototype framework has been developed and implemented. Investigational results demonstration that the methods projected for HrosSI recovery are successful. The proposed system is successful and tailored to stereo images conveyed across space but may also be sufficient for stereo images transmitted by air.
Page(s) : 651-657
ISSN : 0976-5166
Source : Vol. 11, No.5
PDF : Download
DOI : 10.21817/indjcse/2020/v11i5/201105148