Image fusion in remote sensing based on sparse sampling method and PCNN techniques


Ho, Chiung Ching and Henesey, Lawrence and Nirala, Satish and Mishra, Deepak and Sagayam, K. Martin and Vasanth, X. Ajay and Ponraj, D. Narain (2019) Image fusion in remote sensing based on sparse sampling method and PCNN techniques. Machine Learning for Big Data Analysis, 1. pp. 149-180. ISSN 2512-8868

Full text not available from this repository.


mage fusion is a combination of multiple images that results in a fused image. It provides more information than any other input images and is base don discrete wavelet transformation (DWT) and a sparse sampling method. The sparse sampling method offers better performance than Nyquist theorem as a signal processing technique. Among the various techniques, DWT offers many advantages, such as yielding higher quality, requiring less storage and incurring lower costs, which is very useful in image applications. Image-related applications have few constraints, such as minimum data storage and lower bandwidth for communication that takes place through satellites, which actually results in the capture of low-quality images. Major technical constraints, such as minimum data storage for satellite imaging in space and lower bandwidth for communication with earth-based stations, etc., limit the ability of sat-ellite sensors to capture images with high spatial and high spectral resolution simul-taneously. To overcome this problem, image fusion has proved to be a potential tool for remote sensing applications that incorporate data from combinations of panchro-matic, multispectral images and that aim to produce a composite image having bothhigher spatial and spectral resolutions. Research in this area can be traced back to the most recent couple of decades; unfortunately, the diverse methodological approaches proposed so far by different researchers have not been discussed. Experimental re-sults indicate that a comparative analysis of two methods used in image fusion, suchas compressive sensing and pulse-coupled neural network (PCNN), for a few remote sensing location data, e.g., England, forest land, Egypt, island, Balboa and bare soil,is used in this work. The following performance metrics have two assessment proce-dures, i.e., at full- and reduced-scale resolutions, to evaluate the performance of these algorithms. Keywords: Comp

Item Type: Article
Uncontrolled Keywords: Neural network
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 05 Jan 2022 04:40
Last Modified: 05 Jan 2022 04:40


Downloads per month over past year

View ItemEdit (login required)