Abstract:The availability of large volumes of remote sensing data insists on higher degree of automation in feature extraction, making it a need of the <a class="link-external link-http" href="http://hour.The" rel="external noopener nofollow">this http URL</a> huge quantum of data that needs to be processed entails accelerated processing to be <a class="link-external link-http" href="http://enabled.GPUs" rel="external noopener nofollow">this http URL</a>, which were originally designed to provide efficient visualization, are being massively employed for computation intensive parallel processing environments. Image processing in general and hence automated feature extraction, is highly computation intensive, where performance improvements have a direct impact on societal needs. In this context, an algorithm has been formulated for automated feature extraction from a panchromatic or multispectral image based on image processing techniques. Two Laplacian of Guassian (LoG) masks were applied on the image individually followed by detection of zero crossing points and extracting the pixels based on their standard deviation with the surrounding pixels. The two extracted images with different LoG masks were combined together which resulted in an image with the extracted features and edges. Finally the user is at liberty to apply the image smoothing step depending on the noise content in the extracted image. The image is passed through a hybrid median filter to remove the salt and pepper noise from the image. This paper discusses the aforesaid algorithm for automated feature extraction, necessity of deployment of GPUs for the same; system-level challenges and quantifies the benefits of integrating GPUs in such environment. The results demonstrate that substantial enhancement in performance margin can be achieved with the best utilization of GPU resources and an efficient parallelization strategy. Performance results in comparison with the conventional computing scenario have provided a speedup of 20x, on realization of this parallelizing strategy.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is the efficiency and accuracy of automated feature extraction techniques when dealing with a large amount of remote - sensing data. Specifically, the paper focuses on how to use Graphics Processing Units (GPUs) to accelerate the process of automatically extracting features from satellite images in order to meet the increasing demand for processing the massive image data generated by high - resolution Earth - observation sensors. It is mentioned in the paper that traditional data - processing methods are difficult to meet the need for quickly and efficiently extracting geospatial features from these big data. Therefore, an automated feature - extraction algorithm based on image - processing techniques is proposed, and by using the Laplacian of Gaussian (LoG) filter combined with zero - crossing detection and standard - deviation statistical techniques to identify and extract edges and specific features in the image. In addition, in order to improve the execution efficiency of the algorithm, the research also explores how to effectively use the parallel - computing capabilities of the GPU, especially through optimized parallel strategies and kernel - call parameters, achieving a significant performance improvement compared to traditional CPU - processing methods. The key contributions of the paper are: 1. **Proposing an efficient automated feature - extraction algorithm**: This algorithm can extract information such as roads, buildings or water bodies from panchromatic or single - multispectral - band satellite images. 2. **Using GPU - accelerated processing**: By designing a parallelization strategy suitable for the GPU architecture, efficient execution of the algorithm on the GPU is achieved, greatly shortening the processing time. 3. **Detailed analysis of the implementation details and performance evaluation of the algorithm**: Including the design and comparison of different parallelization methods (such as row - by - row parallelism, pixel - level parallelism, etc.), and specific performance indicators for different test images, such as kernel - execution time, data - transfer time and speed - up, etc. In conclusion, this paper aims to solve the problem of automated feature extraction in remote - sensing data processing through technological innovation, especially when dealing with large - scale data sets, by using the powerful computing capabilities of the GPU to improve the processing efficiency and the accuracy of the results.

GPU Acclerated Automated Feature Extraction from Satellite Images

Hyperspectral Image Feature Extraction Accelerated By Gpu

Parallel Processing of Massive Remote Sensing Images in a GPU Architecture.

Parallel Image Resample Algorithm Based on GPU for Land Remote Sensing Data Management

GPU Accelerated Computation for Surface Topography Measurement

Papaya Tree Detection with UAV Images Using a GPU-Accelerated Scale-Space Filtering Method

GPU-based acceleration of local binary pattern algorithm for Internet applications

Pan-sharpening of spectral image with anisotropic diffusion for fine feature extraction using GPU

GPU Implementation of Graph-Regularized Sparse Unmixing with Superpixel Structures

Automatic Feature Extraction from High-Resolution Satellite Imagery using Deep Learning techniques

A Novel Parallelized Feature Extraction in Grouped Scale Space Based on Graphic Processing Units

High Performance Hyperspectral Image Classification using Graphics Processing Units

Hyperspectral Remote Sensing Image Data Processing on GPU

GPU Implementation of an Automatic Target Detection and Classification Algorithm for Hyperspectral Image Analysis

Parallel programing templates for remote sensing image processing on GPU architectures: design and implementation

Accelerating Large Scale Image Analyses on Parallel, CPU-GPU Equipped Systems

Parallel Processing of Sliding Spotlight Mode SAR Imaging Based on GPU

Parallelizing Astronomical Source Extraction on the Gpu

Feature Extraction and Object Detection Using Fast-Convolutional Neural Network for Remote Sensing Satellite Image

Parallel Implementation of Spatial–Spectral Endmember Extraction on Graphic Processing Units

Applying GPU and POSIX thread technologies in massive remote sensing image data processing