scholarly journals GPU Parallel Implementation for Real-Time Feature Extraction of Hyperspectral Images

2020 ◽  
Vol 10 (19) ◽  
pp. 6680
Author(s):  
Chunchao Li ◽  
Yuanxi Peng ◽  
Mingrui Su ◽  
Tian Jiang
Keyword(s):  
Feature Extraction ◽  
Real Time ◽  
Hyperspectral Image ◽  
Parallel Implementation ◽  
Image Data ◽  
Computer Experiments ◽  
Principal Component ◽  
Hyperspectral Images ◽  
Real Time Processing ◽  
Time Processing

As the application of real-time requirements gradually increases or real-time processing and responding become the bottleneck of the task, parallel computing in hyperspectral image applications has also become a significant research focus. In this article, a flexible and efficient method is utilized in the noise adaptive principal component (NAPC) algorithm for feature extraction of hyperspectral images. From noise estimation to feature extraction, we deploy a complete CPU-GPU collaborative computing solution. Through the computer experiments on three traditional hyperspectral datasets, our proposed improved NAPC (INAPC) has stable superiority and provides a significant speedup compared with the OpenCV and PyTorch implementation. What’s more, we creatively establish a complete set of uncrewed aerial vehicle (UAV) photoelectric platform, including UAV, hyperspectral camera, NVIDIA Jetson Xavier, etc. Flight experimental results show, considering hyperspectral image data acquisition and transmission time, the proposed algorithm meets the feature extraction of real-time processing.

scholarly journals Using Predictive and Differential Methods with K2-Raster Compact Data Structure for Hyperspectral Image Lossless Compression

2019 ◽  
Vol 11 (21) ◽  
pp. 2461 ◽  
Author(s):  
Kevin Chow ◽  
Dion Tzamarias ◽  
Ian Blanes ◽  
Joan Serra-Sagristà
Keyword(s):  
Data Structure ◽  
Hyperspectral Image ◽  
Lossless Compression ◽  
Hyperspectral Images ◽  
Direct Access ◽  
Bit Rate ◽  
Compression Algorithms ◽  
Real Time Processing ◽  
Time Processing ◽  
Differential Encoding

This paper proposes a lossless coder for real-time processing and compression of hyperspectral images. After applying either a predictor or a differential encoder to reduce the bit rate of an image by exploiting the close similarity in pixels between neighboring bands, it uses a compact data structure called k 2 -raster to further reduce the bit rate. The advantage of using such a data structure is its compactness, with a size that is comparable to that produced by some classical compression algorithms and yet still providing direct access to its content for query without any need for full decompression. Experiments show that using k 2 -raster alone already achieves much lower rates (up to 55% reduction), and with preprocessing, the rates are further reduced up to 64%. Finally, we provide experimental results that show that the predictor is able to produce higher rates reduction than differential encoding.

scholarly journals Multi-scale guided feature extraction and classification algorithm for hyperspectral images

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Shiqi Huang ◽  
Ying Lu ◽  
Wenqing Wang ◽  
Ke Sun
Keyword(s):  
Feature Extraction ◽  
Image Classification ◽  
Classification Accuracy ◽  
Hyperspectral Image ◽  
Image Data ◽  
Hyperspectral Images ◽  
Hyperspectral Image Classification ◽  
Multi Scale ◽  
Hyperspectral Image Data ◽  
Guided Filtering

AbstractTo solve the problem that the traditional hyperspectral image classification method cannot effectively distinguish the boundary of objects with a single scale feature, which leads to low classification accuracy, this paper introduces the idea of guided filtering into hyperspectral image classification, and then proposes a multi-scale guided feature extraction and classification (MGFEC) algorithm for hyperspectral images. Firstly, the principal component analysis theory is used to reduce the dimension of hyperspectral image data. Then, guided filtering algorithm is used to achieve multi-scale spatial structure extraction of hyperspectral image by setting different sizes of filtering windows, so as to retain more edge details. Finally, the extracted multi-scale features are input into the support vector machine classifier for classification. Several practical hyperspectral image datasets were used to verify the experiment, and compared with other spectral feature extraction algorithms. The experimental results show that the multi-scale features extracted by the MGFEC algorithm proposed in this paper are more accurate than those extracted by only using spectral information, which leads to the improvement of the final classification accuracy. This fully shows that the proposed method is not only effective, but also suitable for processing different hyperspectral image data.

scholarly journals AN INSTANTANEOUS POINT SOURCE DETECTION AND EXTRACTION SYSTEM FOR GEO REMOTE SENSING SATELLITE

2018 ◽  
Vol XLII-3 ◽  
pp. 451-454
Author(s):  
Y. Guo ◽  
Q. Li ◽  
W. Wu
Keyword(s):  
Remote Sensing ◽  
Point Source ◽  
Real Time ◽  
Processing System ◽  
Image Data ◽  
Source Detection ◽  
Source Information ◽  
Real Time Processing ◽  
Time Processing ◽  
Instantaneous Point

To accomplish the task of detecting the instantaneous point source, an on-board information real-time processing system is designed which can process the point-source detection with reconfigurable function. The system has the algorithm reconfigurable function, which can detect and extract the instantaneous point source from the remote sensing image. By using FPGA programming, the satellite target detection and processing algorithm can be update easily. At the same time, the software can be reconfigured to improve the system's information processing capabilities. The system has been verified by simulating real instantaneous source point target image data to meet the real-time processing requirements of instantaneous point source information detection.

Fast Pedestrian Detection Based on GPU+CPU

2013 ◽  
Vol 718-720 ◽  
pp. 2291-2295
Author(s):  
Heng Li Liu ◽  
Si Yue Zhou ◽  
Zheng Peng Yuan ◽  
Yu Chi Wang
Keyword(s):  
Real Time ◽  
Processing Speed ◽  
Parallel Implementation ◽  
Pedestrian Detection ◽  
Frame Rate ◽  
Real Time Processing ◽  
Time Processing ◽  
Practical Applications ◽  
Security Applications

Accurate pedestrian detection is required for practical applications such as automotive and security applications. However, the implementation does not have enough performance because the present schemes are not sufficient. In this paper, the authors proposed parallel implementation of HOG-based pedestrian detection on GPU to obtain real-time processing results. By the proposed implementation, the total processing speed becomes 60 times faster than that of original one on frame rate and real-time processing is achieved.

GPU Acceleration of Algorithm to Design Directional Volumetric Display for Real-time Processing

2019 ◽  
pp. 989
Author(s):  
Daiki Matsumoto ◽  
Ryuji Hirayama ◽  
Naoto Hoshikawa ◽  
Hirotaka Nakayama ◽  
Tomoyoshi Shimobaba ◽  
...  
Keyword(s):  
Real Time ◽  
Gpu Acceleration ◽  
Real Time Processing ◽  
Time Processing ◽  
Volumetric Display

Putting up barriers

2017 ◽  
Author(s):  
David J. Lobina
Keyword(s):  
Knowledge Base ◽  
Real Time ◽  
Language Processing ◽  
Cognitive Domain ◽  
Linguistic Knowledge ◽  
Levels Of Explanation ◽  
Real Time Processing ◽  
Time Processing ◽  
Neural Underpinnings ◽  
Different Levels

The study of cognitive phenomena is best approached in an orderly manner. It must begin with an analysis of the function in intension at the heart of any cognitive domain (its knowledge base), then proceed to the manner in which such knowledge is put into use in real-time processing, concluding with a domain’s neural underpinnings, its development in ontogeny, etc. Such an approach to the study of cognition involves the adoption of different levels of explanation/description, as prescribed by David Marr and many others, each level requiring its own methodology and supplying its own data to be accounted for. The study of recursion in cognition is badly in need of a systematic and well-ordered approach, and this chapter lays out the blueprint to be followed in the book by focusing on a strict separation between how this notion applies in linguistic knowledge and how it manifests itself in language processing.

scholarly journals Limits on expectation-based processing: Use of grammatical aspect for co-reference in L2

2020 ◽  
pp. 1-25
Author(s):  
Theres Grüter ◽  
Hannah Rohde
Keyword(s):  
Real Time ◽  
Language Processing ◽  
Judgment Task ◽  
Linguistic Knowledge ◽  
Real Time Processing ◽  
Time Processing ◽  
Grammatical Aspect ◽  
Level Information ◽  
Native Speakers Of English ◽  
Reference Processing

Abstract This study examines the use of discourse-level information to create expectations about reference in real-time processing, testing whether patterns previously observed among native speakers of English generalize to nonnative speakers. Findings from a visual-world eye-tracking experiment show that native (L1; N = 53) but not nonnative (L2; N = 52) listeners’ proactive coreference expectations are modulated by grammatical aspect in transfer-of-possession events. Results from an offline judgment task show these L2 participants did not differ from L1 speakers in their interpretation of aspect marking on transfer-of-possession predicates in English, indicating it is not lack of linguistic knowledge but utilization of this knowledge in real-time processing that distinguishes the groups. English proficiency, although varying substantially within the L2 group, did not modulate L2 listeners’ use of grammatical aspect for reference processing. These findings contribute to the broader endeavor of delineating the role of prediction in human language processing in general, and in the processing of discourse-level information among L2 users in particular.

Sign in / Sign up

Export Citation Format

Share Document