Comparative Studies of Target Detection Algorithm for Hyperspectral Imagery

2015 ◽  
Vol 1 (1) ◽  
Author(s):  
B K Nagesha ◽  
M R Puttaswamy ◽  
Dsouza Hasmitha ◽  
G Hemantha Kumar
Keyword(s):  
Comparative Study ◽  
Hyperspectral Imaging ◽  
Target Detection ◽  
Spatial Resolution ◽  
Hyperspectral Image ◽  
Hyperspectral Imagery ◽  
Detection Algorithm ◽  
Hyperspectral Data ◽  
Detection Algorithms ◽  
Complex Process

<p>Target detection in hyperspectral imagery is a complex process due to many factors. Exploiting the hyperspectral image<br />for analysis is very challenging due to large information and low spatial resolution. However, hyperspectral target<br />detection has numerous applications. Hence, it is important to pursue research in target detection. In this paper, a<br />comparative study of target detection algorithms for hyperspectral imagery is presented along with scope for future<br />research. A comparative study behind the hyperspectral imaging is detailed. Also, various challenges involved in<br />exploring the hyperspectral data are discussed.</p>

scholarly journals Anomaly targets detection of hyperspectral imagery based on wavelet transform and sparse representation

2018 ◽  
Vol 232 ◽  
pp. 02054
Author(s):  
Cheng Baozhi
Keyword(s):  
Wavelet Transform ◽  
Sparse Representation ◽  
Target Detection ◽  
Hyperspectral Image ◽  
Hyperspectral Imagery ◽  
Image Data ◽  
Detection Algorithm ◽  
Discrete Wavelet ◽  
Detection Accuracy ◽  
Low Efficiency

The research of anomaly target detection algorithm in hyperspectral imagery is a hot issue, which has important research value. In order to overcome low efficiency of current anomaly target detection in hyperspectral image, an anomaly detection algorithm for hyperspectral images based on wavelet transform and sparse representation was proposed. Firstly, two-dimensional discrete wavelet transform is used to denoise the hyperspectral image, and the new hyperspectral image data are obtained. Then, the results of anomaly target detection are obtained by using sparse representation theory. The real AVIRIS hyperspectral imagery data sets are used in the experiments. The results show that the detection accuracy and false alarm rate of the propoesd algorithm are better than RX and KRX algorithm.

scholarly journals Ensemble-Based Cascaded Constrained Energy Minimization for Hyperspectral Target Detection

2019 ◽  
Vol 11 (11) ◽  
pp. 1310 ◽  
Author(s):  
Rui Zhao ◽  
Zhenwei Shi ◽  
Zhengxia Zou ◽  
Zhou Zhang
Keyword(s):  
Target Detection ◽  
Energy Minimization ◽  
Hyperspectral Image ◽  
Hyperspectral Data ◽  
Least Square ◽  
Learning System ◽  
Strong Nonlinearity ◽  
Generalization Ability ◽  
Detection Algorithms ◽  
Ensemble Techniques

Ensemble learning is an important group of machine learning techniques that aim to enhance the nonlinearity and generalization ability of a learning system by aggregating multiple learners. We found that ensemble techniques show great potential for improving the performance of traditional hyperspectral target detection algorithms, while at present, there are few previous works have been done on this topic. To this end, we propose an Ensemble based Constrained Energy Minimization (E-CEM) detector for hyperspectral image target detection. Classical hyperspectral image target detection algorithms like Constrained Energy Minimization (CEM), matched filter (MF) and adaptive coherence/cosine estimator (ACE) are usually designed based on constrained least square regression methods or hypothesis testing methods with Gaussian distribution assumption. However, remote sensing hyperspectral data captured in a real-world environment usually shows strong nonlinearity and non-Gaussianity, which will lead to performance degradation of these classical detection algorithms. Although some hierarchical detection models are able to learn strong nonlinear discrimination of spectral data, due to the spectrum changes, these models usually suffer from the instability in detection tasks. The proposed E-CEM is designed based on the classical CEM detection algorithm. To improve both of the detection nonlinearity and generalization ability, the strategies of “cascaded detection”, “random averaging” and “multi-scale scanning” are specifically designed. Experiments on one synthetic hyperspectral image and two real hyperspectral images demonstrate the effectiveness of our method. E-CEM outperforms the traditional CEM detector and other state-of-the-art detection algorithms. Our code will be made publicly available.

scholarly journals A comparative study of target detection algorithms in hyperspectral imagery applied to agricultural crops in Colombia

Tecnura ◽  
2016 ◽  
Vol 20 (49) ◽  
pp. 86 ◽  
Author(s):  
Ariolfo Camacho Velasco ◽  
César Augusto Vargas García ◽  
Henry Arguello Fuentes
Keyword(s):  
Comparative Study ◽  
Target Detection ◽  
Hyperspectral Imagery ◽  
Agricultural Crops ◽  
Detection Algorithms

Contexto: Las imágenes hiperespectrales (HSI) contienen información en alta resolución espectral, en cientos de bandas contiguas sobre un rango del espectro electromagnético. Se ha tomado ventaja de la información espectral mediante algoritmos de clasificación, detección de cambios, de anomalías o detección de objetivos. Específicamente, en las dos últimas décadas han sido propuestos diferentes algoritmos para detectar objetivos en HSI. Sin embargo, encontrar un algoritmo de detección con un desempeño óptimo para diferentes escenarios y objetivos, aún es materia de investigación, debido a la alta variabilidad espectral y diversidad de escenarios del mundo real.Objetivo: La presente investigación realiza un estudio comparativo de algoritmos de detección de objetivos en imágenes hiperespectrales aplicados a la agricultura colombiana.Método: Las evaluaciones se realizaron sobre 20 HSI reales adquiridas por el sensor satelital Hyperion y 6 HSI sintéticas con diferentes niveles de ruido. En el desarrollo de la investigación se implantaron 5 objetivos sintéticos; y se extrajeron más de 115 firmas espectrales reales, 11 de ellas fueron usadas como objetivo en la evaluación de los algoritmos, permitiendo la caracterización de 5 cultivos agrícolas del nororiente colombiano en 5 áreas de estudio diferentes. Los resultados muestran que el algoritmo de Estimación de Coherencia Adaptativo (ACE) presenta un mejor desempeño con probabilidades de detección PD > 90% para diferentes escenarios y objetivos de tipo agrícola, tanto en imágenes sintéticas como reales.Conclusiones: En aplicaciones de detección de objetivos en HSI, es crítico encontrar un algoritmo que presente un rendimiento óptimo para diferentes escenarios y objetivos, debido a la variabilidad espectral generada por las diferentes condiciones geográficas de Colombia. Por otra parte, este trabajo permite mostrar que es posible realizar nuevas investigaciones y aplicaciones a nivel nacional tomando ventaja de las técnicas de imágenes hiperespectrales y algoritmos de detección espectral; específicamente en el sector de la agricultura colombiana.

scholarly journals A comparative study of hyperspectral unmixing using different algorithm approaches

2020 ◽  
Vol 20 (2) ◽  
pp. 813
Author(s):  
Ain Zat Mohd Yusof ◽  
Redzuan Abdul Manap ◽  
Abdul Majid Darsono
Keyword(s):  
Comparative Study ◽  
Multiple Scattering ◽  
Spatial Resolution ◽  
Hyperspectral Image ◽  
Hyperspectral Data ◽  
Accurate Estimation ◽  
Hyperspectral Unmixing ◽  
Propagation Of Errors ◽  
Spectral Angle ◽  
Material Mixing

<span>Hyperspectral unmixing (HU) is an important technique for remotely sensed hyperspectral data exploitation. Hyperspectral unmixing is required to get an accurate estimation due to low spatial resolution of hyperspectral cameras, microscopic material mixing, and multiple scattering that cause spectra measured by hyperspectral cameras are mixtures of spectra of materials in a scene. It is a process of estimating constituent endmembers and their fractional abundances present at each pixel in hyperspectral image. Researchers have devised and investigated many models searching for robust, stable, tractable and accurate unmixing algorithm. Such algorithm are highly desirable to avoid propagation of errors within the process. This paper presents the comparison of hyperspectral unmixing method by using different kind of algorithms. These algorithms are named VCA, NFINDR, SISAL, and CoNMF. The performance of unmixing process is evaluated by calculating the SAD (spectral angle distance) for each endmembers by using same input of hyperspectral data for different algorithm.</span>

scholarly journals Angle Distance-Based Hierarchical Background Separation Method for Hyperspectral Imagery Target Detection

2020 ◽  
Vol 12 (4) ◽  
pp. 697 ◽  
Author(s):  
Xiaohui Hao ◽  
Yiquan Wu ◽  
Peng Wang
Keyword(s):  
Target Detection ◽  
Hyperspectral Imagery ◽  
Original Data ◽  
Separation Method ◽  
Detection Accuracy ◽  
Target Information ◽  
Detection Algorithms ◽  
Target Spectrum ◽  
Candidate Target ◽  
The Given

Traditional detectors for hyperspectral imagery (HSI) target detection (TD) output the result after processing the HSI only once. However, using the prior target information only once is not sufficient, as it causes the inaccuracy of target extraction or the unclean separation of the background. In this paper, the target pixels are located by a hierarchical background separation method, which explores the relationship between the target and the background for making better use of the prior target information more than one time. In each layer, there is an angle distance (AD) between each pixel spectrum in HSI and the given prior target spectrum. The AD between the prior target spectrum and candidate target ones is smaller than that of the background pixels. The AD metric is utilized to adjust the values of pixels in each layer to gradually increase the separability of the background and the target. For making better discrimination, the AD is calculated through the whitened data rather than the original data. Besides, an elegant and ingenious smoothing processing operation is employed to mitigate the influence of spectral variability, which is beneficial for the detection accuracy. The experimental results of three real hyperspectral images show that the proposed method outperforms other classical and recently proposed HSI target detection algorithms.

scholarly journals A Comparative Study of Moving Target Detection Algorithms

2020 ◽  
Vol 790 ◽  
pp. 012061
Author(s):  
Yuanyuan Wang ◽  
Xiaolei Zhou ◽  
Yuanyuan Zuo ◽  
Zhuang Wu
Keyword(s):  
Comparative Study ◽  
Target Detection ◽  
Moving Target ◽  
Moving Target Detection ◽  
Detection Algorithms

scholarly journals Hyperspectral Image Classification Using Parallel Autoencoding Diabolo Networks on Multi-Core and Many-Core Architectures

Electronics ◽  
2018 ◽  
Vol 7 (12) ◽  
pp. 411 ◽  
Author(s):  
Emanuele Torti ◽  
Alessandro Fontanella ◽  
Antonio Plaza ◽  
Javier Plaza ◽  
Francesco Leporati
Keyword(s):  
Deep Learning ◽  
Hyperspectral Imaging ◽  
Hyperspectral Image ◽  
Unsupervised Classification ◽  
Hyperspectral Data ◽  
Machine Learning Techniques ◽  
High Data ◽  
Learning Techniques ◽  
Speed Up ◽  
Many Core

One of the most important tasks in hyperspectral imaging is the classification of the pixels in the scene in order to produce thematic maps. This problem can be typically solved through machine learning techniques. In particular, deep learning algorithms have emerged in recent years as a suitable methodology to classify hyperspectral data. Moreover, the high dimensionality of hyperspectral data, together with the increasing availability of unlabeled samples, makes deep learning an appealing approach to process and interpret those data. However, the limited number of labeled samples often complicates the exploitation of supervised techniques. Indeed, in order to guarantee a suitable precision, a large number of labeled samples is normally required. This hurdle can be overcome by resorting to unsupervised classification algorithms. In particular, autoencoders can be used to analyze a hyperspectral image using only unlabeled data. However, the high data dimensionality leads to prohibitive training times. In this regard, it is important to realize that the operations involved in autoencoders training are intrinsically parallel. Therefore, in this paper we present an approach that exploits multi-core and many-core devices in order to achieve efficient autoencoders training in hyperspectral imaging applications. Specifically, in this paper, we present new OpenMP and CUDA frameworks for autoencoder training. The obtained results show that the CUDA framework provides a speed-up of about two orders of magnitudes as compared to an optimized serial processing chain.

Sign in / Sign up

Export Citation Format

Share Document