Feature-Level Fusion for Object Segmentation Using Mutual Information

Sharma, Vinay; Davis, James W.

doi:10.1007/978-1-84800-277-7_13

Cited by 9 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The methods for accomplishing this differ according to the application domain. For example, rough object segmentation [5] is a feature extraction technique sometimes used for image processing. Syntactic analysis [6] is a feature extraction approach used in text classification problems, while auto-correlation techniques [7] and wavelet transformations [8] can be applied in protein classification problems.…”

Section: Related Workmentioning

confidence: 99%

Applying Feature Selection to Short Time Wavelet Transformed Vibration Data for Reliability Analysis of an Ocean Turbine

Duhaney

Khoshgoftaar

Wald

2012

2012 11th International Conference on Machine Learning and Applications

View full text Add to dashboard Cite

This paper considers the use of feature selection within the state detection module for an ocean turbine condition monitoring system. The goal is to reduce the quantity of data to be processed while maintaining or improving state detection capabilities. Five feature selection techniques (Chisquared, Information Gain, Signal-To-Noise, AUC and PRC) are evaluated based on their effects on four widely used machine learning algorithms, namely Naive Bayes, k-Nearest Neighbors, Decision Tree and Logistic Regression, when each machine learner is trained on the top n features selected by each feature selection technique. Six values of n (2, 4, 6, 8, 10 and 15) were considered. Features were extracted from the raw vibration signals using a Short Time Wavelet Transform with Baselining (STWTB) technique designed to allow for reliable state detection regardless of the turbine's operating conditions, which are often reflected within its vibration readings. The condition-independent features extracted by the STWTB are then fused to combine all the data observed by all sensor sources. Models were built on data gathered at one operating condition and tested against data from a different operating condition to simulate the problem of building models which work regardless of operating condition. Results show that kNearest Neighbors, Naive Bayes and Logistic Regression have improved classification performance when using less than 11% of the 78 available features, with Logistic Regression needing just 2 features selected by the Signal-To-Noise technique to generate a perfect classification model. The Decision Tree performed best without feature selection.

show abstract

Section: Related Workmentioning

confidence: 99%

Applying Feature Selection to Short Time Wavelet Transformed Vibration Data for Reliability Analysis of an Ocean Turbine

Duhaney

Khoshgoftaar

Wald

2012

2012 11th International Conference on Machine Learning and Applications

View full text Add to dashboard Cite

show abstract

“…These methods include feature based PCA [9], [10], segment fusion [10], edge fusion [10] and contour fusion [11]. They are usually robust to noise and mis-registration.…”

Section: Introductionmentioning

confidence: 99%

Entropy dependent compressive sensing based image fusion

Jameel

Ghafoor

Riaz

2013

2013 International Symposium on Intelligent Signal Processing and Communication Systems

View full text Add to dashboard Cite

An entropy dependent compressive sensing based image fusion is proposed which adaptively adjusts the number of compressive measurements depending on the amount of information (computed using entropy). Since the computational complexity is dependent on the number of compressive measurements, the proposed scheme is more efficient and accurate. Simulation results show the significance of the proposed scheme compared to existing compressive sensing based state of art scheme.

show abstract

“…Multiscale transforms like pyramids and wavelets are also types of pixel-level fusion [11, 14]. Feature-level methods include feature based PCA [12, 13], segment fusion [13], edge fusion [13], and contour fusion [16]. They are usually robust to noise and misregistration.…”

Section: Introductionmentioning

confidence: 99%

Improved Guided Image Fusion for Magnetic Resonance and Computed Tomography Imaging

Jameel

Ghafoor

Riaz

2014

The Scientific World Journal

View full text Add to dashboard Cite

Improved guided image fusion for magnetic resonance and computed tomography imaging is proposed. Existing guided filtering scheme uses Gaussian filter and two-level weight maps due to which the scheme has limited performance for images having noise. Different modifications in filter (based on linear minimum mean square error estimator) and weight maps (with different levels) are proposed to overcome these limitations. Simulation results based on visual and quantitative analysis show the significance of proposed scheme.

show abstract

Feature-Level Fusion for Object Segmentation Using Mutual Information

Abstract: Abstract. In this chapter, a new feature-level image fusion

Cited by 9 publications

References 36 publications

Applying Feature Selection to Short Time Wavelet Transformed Vibration Data for Reliability Analysis of an Ocean Turbine

Applying Feature Selection to Short Time Wavelet Transformed Vibration Data for Reliability Analysis of an Ocean Turbine

Entropy dependent compressive sensing based image fusion

Improved Guided Image Fusion for Magnetic Resonance and Computed Tomography Imaging

Contact Info

Product

Resources

About