Sparse Unmixing-Based Content Retrieval of Hyperspectral Images on Graphics Processing Units

Sevilla, Jorge; Jiménez, Luis Ignacio

doi:10.1109/lgrs.2015.2483679

Cited by 7 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Salah satu aktifitas dari perencanaan proyek perangkat lunak yaitu estimasi sumber daya, biaya serta jadwal proyek. Dalam membangun perangkat lunak sebaiknya dimulai dengan perencanaan proyek [4].…”

Section: Pendahuluanunclassified

Estimasi Biaya Software FAS (Financing Analysis System) Menggunakan Metode Function Point (Studi Kasus Pada PT BPRS Lantabur Tebuireng)

2020

View full text Add to dashboard Cite

Software cost estimation is the process of predicting software development efforts. The basic input of software cost estimation is the measurement metric. Projects often experience delays, over-budget, and are not completed due to failure to estimate software development costs. PT BPRS (Bank Perkreditan Rakyat Syariah) Lanatabur Tebuireng determines the estimated cost based on the amount of human resources, features needed, and funds owned. This study explains the estimated costs of the FAS (Financing Analysis System) software at PT BPRS Lantabur Tebuireng using the Function Point method. Function Point is a method of measuring software functionality based on the type of user function that is External Input, External Output, External Inquire, Internal Logic File, and External Interface File as well as technical calculations of software development. The final of the FAS (Financing Analysis System) study cost around IDR 94,797,120.

show abstract

Section: Pendahuluanunclassified

Estimasi Biaya Software FAS (Financing Analysis System) Menggunakan Metode Function Point (Studi Kasus Pada PT BPRS Lantabur Tebuireng)

2020

View full text Add to dashboard Cite

show abstract

“…The third order local derivative pattern is computed using the second order derivatives. The third order directional LDP [2], for the pixel 0 ∝ 3 ( 0 ), in α direction is given as in (7).…”

Section: Ldp Descriptor and Usage For Image Retrievalmentioning

confidence: 99%

“…GPUs are inexpensive, scalable and have very high computation power compared to single processor CPUs. GPUs have been widely used now for accelerating many efficient CBIR algorithms [4], [5] in the field of image vision, medical imaging [6], remote sensing [7] etc. This paper proposes a parallel implementation of the LDP algorithm on NVIDIA GPUs.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Implementation and Performance Optimization of Local Derivative Pattern Algorithm on GPUs

Chandran

Gangodkar

Mittal

2018

IJECE

View full text Add to dashboard Cite

<p><span>Pattern based texture descriptors are widely used in Content Based Image Retrieval (CBIR) for efficient retrieval of matching images. Local Derivative Pattern (LDP), a higher order local pattern operator, originally proposed for face recognition, encodes the distinctive spatial relationships contained in a local region of an image as the feature vector. LDP efficiently extracts finer details and provides efficient retrieval however, it was proposed for images of limited resolution. Over the period of time the development in the digital image sensors had paid way for capturing images at a very high resolution. LDP algorithm though very efficient in content-based image retrieval did not scale well when capturing features from such high-resolution images as it becomes computationally very expensive. This paper proposes how to efficiently extract parallelism from the LDP algorithm and strategies for optimally implementing it by exploiting some inherent General-Purpose Graphics Processing Unit (GPGPU) characteristics. By optimally configuring the GPGPU kernels, image retrieval was performed at a much faster rate. The LDP algorithm was ported on to Compute Unified Device Architecture (CUDA) supported GPGPU and a maximum speed up of around 240x was achieved as compared to its sequential counterpart.</span></p>

show abstract

“…This paper is a prospective study that brings out ways to accelerate the hyperspectral reconstruction method and determine its performance on an embedded device. This is original in the field of HS imaging in embedded systems, since most of the literature focuses on either HS data compression [17,18] or unmixing [19,20]. Our analysis presents the computational complexity of the method but also includes an estimation of the memory footprint and bandwidth, which are often overlooked in the literature but are crucial when designing a device for real use.…”

Section: Introductionmentioning

confidence: 99%

Feasibility of a Real-Time Embedded Hyperspectral Compressive Sensing Imaging System

Lim

Mancini

Mura

2022

Sensors

View full text Add to dashboard Cite

Hyperspectral imaging has been attracting considerable interest as it provides spectrally rich acquisitions useful in several applications, such as remote sensing, agriculture, astronomy, geology and medicine. Hyperspectral devices based on compressive acquisitions have appeared recently as an alternative to conventional hyperspectral imaging systems and allow for data-sampling with fewer acquisitions than classical imaging techniques, even under the Nyquist rate. However, compressive hyperspectral imaging requires a reconstruction algorithm in order to recover all the data from the raw compressed acquisition. The reconstruction process is one of the limiting factors for the spread of these devices, as it is generally time-consuming and comes with a high computational burden. Algorithmic and material acceleration with embedded and parallel architectures (e.g., GPUs and FPGAs) can considerably speed up image reconstruction, making hyperspectral compressive systems suitable for real-time applications. This paper provides an in-depth analysis of the required performance in terms of computing power, data memory and bandwidth considering a compressive hyperspectral imaging system and a state-of-the-art reconstruction algorithm as an example. The results of the analysis show that real-time application is possible by combining several approaches, namely, exploitation of system matrix sparsity and bandwidth reduction by appropriately tuning data value encoding.

show abstract

Sparse Unmixing-Based Content Retrieval of Hyperspectral Images on Graphics Processing Units

Cited by 7 publications

References 15 publications

Estimasi Biaya Software FAS (Financing Analysis System) Menggunakan Metode Function Point (Studi Kasus Pada PT BPRS Lantabur Tebuireng)

Estimasi Biaya Software FAS (Financing Analysis System) Menggunakan Metode Function Point (Studi Kasus Pada PT BPRS Lantabur Tebuireng)

Real-Time Implementation and Performance Optimization of Local Derivative Pattern Algorithm on GPUs

Feasibility of a Real-Time Embedded Hyperspectral Compressive Sensing Imaging System

Contact Info

Product

Resources

About