Extracting nonlinear features for multispectral images by FCMC and KPCA

doi:10.1016/j.dsp.2004.12.004

Digital Signal Processing

Volume 15, Issue 4, July 2005, Pages 331-346

https://doi.org/10.1016/j.dsp.2004.12.004 Get rights and content

Abstract

Classification is a very important task for scene interpretation and other applications of multispectral images. Feature extraction is a key step for classification. By extracting more nonlinear features than corresponding number of linear features in original feature space, classification accuracy for multispectral images can be improved greatly. Therefore, in this paper, an approach based on the fuzzy c-means clustering (FCMC) and kernel principal component analysis (KPCA) is proposed to resolve the problem of multispectral images. The main contribution of this paper is to provide a good preprocessed method for classifying these images. Finally, some experimental results demonstrate that our proposed method is effective and efficient for analyzing the multispectral images.

References (20)

A.J. Bell et al.
The independent components of natural scenes are edge filters
Vision Res.
(1997)
D. Landgrebe
D. Landgrebe
A.A. Maciejewski et al.
An example of principal component analysis applied to correlated images
S. Bannour et al.
Principal component extraction using recursive least square learning
IEEE Trans. Neural Networks
(1995)
S. Chitroub et al.
Principal component analysis of multispectral images using neural network
S. Chitroub et al.
Unsupervised learning rules for POLSAR images analysis
B. Schölkopf et al.
Nonlinear component analysis as a kernel eigenvalue
Neural Comput.
(1998)
B. Schölkopf et al.
Kernel principal component analysis
(1999)
B. Schölkopf et al.
Input space vs feature space in kernel-based methods
IEEE Trans. Neural Networks
(1999)

There are more references available in the full text version of this article.

Cited by (53)

Distributed Analytics For Big Data: A Survey
2024, Neurocomputing
In recent years, a constant and fast information growing has characterized digital applications in the majority of real-life scenarios. Thus, a new information asset, namely Big Data, has been defined and lead to different challenges, mainly related to data storage, management and analysis. Focusing on the last challenge, several Big Data analytics techniques have been developed, based on Machine Learning and Deep Learning paradigms. When dealing with Big Data, traditional approaches often take a lot of time to produce even a single predictive model, due to the extremely high demand of computational resources.
The design of approaches specifically oriented to Big Data is required to overcome these computational issues. Most solutions rely on the deployment of Big Data analytics infrastructures on a cluster of machines and/or on parallelization techniques. When deployment and parallelization apply to Machine Learning and Deep Learning, we can refer to the terms Distributed Machine Learning and Distributed Deep Learning, respectively.
We here discuss the main principles and features of Distributed Machine Learning and Distributed Deep Learning frameworks. The main contribution of this work is a survey of solutions proposed in the literature, through the investigation of selected features and capabilities. In particular, the survey provides a comparative analysis according to the following classification criteria: implemented parallelization technique, supporting device, supported architecture, implemented communication mode, working mode, and class of algorithms.
The paper also gives an overview of the most commonly used criteria and metrics for the performance evaluation of analyzed frameworks; finally, some emerging but promising optimization techniques are reviewed apart from our classification.
A comprehensive survey on segmentation techniques for retinal vessel segmentation
2023, Neurocomputing
In recent years, enormous research has been carried out on the segmentation of blood vessels. Segmentation of blood vessels in retinal images is crucial for diagnosing, treating, evaluating clinical results, and early detection of eye disorders. A successful segmentation accurately reflects the blood vessels’ structure and helps obtain patterns that can be used to identify retinal disorders and diseases. Most recent research on vessel segmentation employs multiple processes to determine the appropriate segmentation. Finding the best techniques for segmentation is a complex process. In certain circumstances, it requires a thorough understanding of every step that can only be acquired through years of training. Comprehension and expertise in segmentation procedures are essential for accurate segmentation. This paper briefly introduces the segmentation of blood vessels in retinal images, describes many preprocessing and segmentation techniques, and summarizes challenges and trends. Furthermore, the limitations of the current systems will be identified.
Comprehensive review on vehicle Detection, classification and counting on highways
2023, Neurocomputing
Vehicle detection, counting and finally classification has been an important aspect of traffic analysis specially on highways in many developed and developing nations. This has vitalized the monitoring of freeways and reduced the reliance on human traffic monitors specially in developed nations. However, much research carried out in this regard since 1995 have been slow to progress until around 2000. Since then, much more encouraging outcomes have been achieved. This has been mainly due to the advances on vision-based computing and the miniaturizing of hardware since the early 2000. Initial vision-based systems used basic computer vision approaches such as background subtraction and edge detection to detect and count vehicles. Early progress of classification had little success until around 2010. Since 2010, many computer vision approaches using Neural Networks have been gradually increasing the efficiency and the realtime operation of such systems. Since then, many deep learning based neural network approaches have produced remarkable outcomes specially in classification. We have also reported highly accurate and efficient deep learning based YOLOv5 system that has recorded an astounding success. We have used models YOLOv5l, YOLOv5m, YOLOv5n and YOLOv5s in our research to ascertain the best model for the task.
However, low light conditions associated with overcast, and dusk have significantly reduced the accuracy in many deep learning-based systems. There are other techniques based on approaches such as headlight detection have appeared to increase the accuracy of counting however, classification at night without adequate lighting still pauses a formidable challenge. The work presented here analyses the vehicle counting problem using computer vision providing many different viewpoints in a view to provide the reader with accurate problem definition.
An overview on density peaks clustering
2023, Neurocomputing
Density peaks clustering (DPC) algorithm is a new algorithm based on density clustering analysis, which can quickly obtain the cluster centers by drawing the decision diagram by using the calculation of local density and relative distance. Without prior knowledge and iteration, the parameters and structure are simple and easy to implement. Since it was proposed in 2014, it has attracted a large number of researchers to explore experiments and improve applications in recent years. In this paper, we first analyze the theory of DPC and its performance advantages and disadvantages. Secondly, it summarizes the improvement of DPC in recent years, analyzes the improvement effect, and shows it with experimental data. The related application research of DPC in different fields is introduced. Finally, the clustering results of DPC, LCDPC, DCHDPC and NADPC algorithms in different data sets are analyzed. Experiments show that the improved algorithm can divide the clustering more accurately, which provides new ideas for improving DPC algorithm in the future. At the same time, this paper summarizes and prospects the improvement and development of DPC.
A survey of machine learning approaches in animal behaviour
2022, Neurocomputing
Animal activity recognition is an important topic that facilitates understanding of animal behavior that is useful for analyzing and classifying their wellbeing. Research studies have been reporting the use of animal activity as an effective indicator of their health state. This survey focuses on recent advancements in machine intelligence utilizing wearable devices for sheep activity recognition. We summarise existing works focusing on various types of sensors used in agricultural sheep activity recognition. Furthermore, data segmentation methods used in each study, followed by the potential recommendations on window size and sample rate selection are addressed in detail. Finally, we present the features being identified as significant along with an overview of machine learning algorithms used in the domain of sheep activity recognition using accelerometer data.
An overview on twin support vector regression
2022, Neurocomputing
Citation Excerpt :
Owing to its high good performance, SVR immediately attracted significant attention after first being proposed [6-11]. Compared with some traditional regression algorithms, SVR has the following advantages [12-18]. Specifically for the problem of limited samples, the optimal solution under the existing sample information is obtained, not only when the sample tends toward infinity.
Twin support vector regression (TSVR) is a useful extension of traditional support vector regression (SVR). As a new regression model, the basic idea of TSVR is generating a pair of nonparallel functions on both sides of the training data points, such that the $ε$ -insensitive upper and lower bounds of the regression function can be determined. Owing to its excellent learning ability, TSVR has become a research hotspot in the field of machine learning. With the deepening of such research, scholars have found that TSVR also has certain limitations, and thus various improved models have been proposed. This review aims to report the recent developments in twin support vector regression. First, the basic concepts and basic models of TSVR are introduced. Second, the improved algorithms and applications of TSVR in recent years are summarized, and the advantages and disadvantages of its representative algorithms are analyzed and compared with the experiments. Finally, we discuss the research conducted on TSVR.

View all citing articles on Scopus

^☆: This work was supported by the NSF of China (Nos. 60472111 and 60405002).

View full text

Extracting nonlinear features for multispectral images by FCMC and KPCA☆

Abstract

Vision Res.

An example of principal component analysis applied to correlated images

Principal component extraction using recursive least square learning

IEEE Trans. Neural Networks

Principal component analysis of multispectral images using neural network

Unsupervised learning rules for POLSAR images analysis

Nonlinear component analysis as a kernel eigenvalue

Neural Comput.

Kernel principal component analysis

Input space vs feature space in kernel-based methods

IEEE Trans. Neural Networks