Propagation Filtering Research Articles

Classification of hyperspectral images using a propagation filter and convolutional neural network

ABSTRACT Deep Learning models have been successfully applied to the classification of hyperspectral images (HSIs), among which Convolutional Neural Network (CNN) can extract more efficient spatial–spectral features to improve classification accuracy compared to traditional classifiers such as Support Vector Machine. However, in the case of limited labelled samples with inevitable noise, simply applying CNN for HSIs classification could be unsatisfactory due to the issue of overfitting. To address the problem, data preprocessing techniques such as filtering methods can be used to effectively remove the noise and enhance the spatial–spectral features. Compared to typical filters like Gabor, Propagation Filter (PF) can effectively smooth the HSIs and preserve the edges without causing the cross-region mixing problem, which could significantly compromise the performance of CNN based HSIs classification models. Therefore, in this paper a new model which combine Propagation Filter with CNN (PF-CNN) is proposed. It makes use of the advantages of PF and CNN to effectively handle the cross-region mixing issue for HSIs classification. The experimental results show that the proposed algorithm outperforms other state-of-the-art HSIs classification models. The source code of the proposed model is available at https://github.com/XinweiJiang/Propagation-Filter-CNN.

LShape Partitioning: Parallel Skyline Query Processing using MapReduce

A skyline query searches the data points that are not dominated by others in the dataset. It is widely adopted for many applications which require multi-criteria decision making. However, skyline query processing is considerably time-consuming for a high-dimensional large scale dataset. Parallel computing techniques are therefore needed to address this challenge, among which MapReduce is one of the most popular frameworks to process big data. A great number of efficient MapReduce skyline algorithms have been proposed in the literature and most of their designs focus on partitioning and pruning the given dataset. However, there are still opportunities for further parallelism. In this study, we propose two parallel skyline processing algorithms using a novel LShape partitioning strategy and an effective Propagation Filtering method. These two algorithms are 2Phase LShape and 1Phase LShape, used for multiple reducers and single reducer, respectively. By extensive experiments, we verify that our algorithms outperformed the state-of-the-art approaches, especially for high-dimensional large scale datasets.

Short-term annoyance reactions to civil helicopter and propeller-driven aircraft noise: A laboratory experiment.

Helicopter noise exhibits distinctive acoustical characteristics (e.g., pulsation) compared to noise from propeller-driven aircraft which contains tonal components. Whereas, at comparable sound exposure levels (LAE), annoyance reactions to these sources might be different, knowledge of potential annoyance differences is scarce. This paper reports a comparison between short-term annoyance reactions to noise from light-weight helicopters and propeller-driven aircraft in a laboratory setup. Stimuli were presented with a 3D sound reproduction system in a listening test facility based on field recordings of takeoffs and landings. Propagation filtering and amplitude changes were carried out to simulate various propagation distances and source levels, covering a reasonable LAE range from 64 to 85 dB(A) for a stimuli length of 24 s. Fifty-six subjects rated their short-term annoyance reactions on the ICBEN 11-point numerical scale. Associations between design variables (source type, procedure, and LAE) and short-term annoyance were explored by means of a linear mixed-effect model. LAE was found to be the major predictor. For the major range of LAE, no significant difference was found between annoyance to noise from the two aircraft types. Observed level differences at equal annoyance ratings were below 1 dB. Furthermore, helicopter landings were found slightly more annoying than helicopter takeoffs.

Fusion of infrared and visible images with propagation filtering

Edge-preserving filters have been widely applied to fusion of infrared and visible images in high quality. However, there exists the performance degradation caused by explicit spatial kernel functions in the difficulty of determining predefined pixel neighborhood regions with traditional edge-preserving filtering decomposition. To remedy such deficiency, a novel method with propagation filtering is proposed for fusion of infrared and visible images, which consists of three steps: (1) Source images are decomposed with gaussian and propagation filters. (2) Rules are designed to achieve base layer combination and detail layer combination. (3) Reconstruction is performed to acquire the fused image. Compared with traditional multi-scale decomposition based on edge-preserving, spatial kernels are not included in the proposed decomposition task, which solves the difficulty of determining predefined pixel neighborhood regions. Moreover, saliency growth and artifact suppression are reached to combine base layers. In addition, by weighted least square optimization, more visual details and less irrelevant infrared information are added into combined detail layers to satisfy human’s perception. As demonstrated in the experimental results, the proposed method achieves much fusion performance compared with other commonly used image fusion algorithms.

Spectral-Spatial Feature Extraction of Hyperspectral Images Based on Propagation Filter.

Recently, image-filtering based hyperspectral image (HSI) feature extraction has been widely studied. However, due to limited spatial resolution and feature distribution complexity, the problems of cross-region mixing after filtering and spectral discriminative reduction still remain. To address these issues, this paper proposes a spectral-spatial propagation filter (PF) based HSI feature extraction method that can effectively address the above problems. The dimensionality/band of an HSI is typically high; therefore, principal component analysis (PCA) is first used to reduce the HSI dimensionality. Then, the principal components of the HSI are filtered with the PF. When cross-region mixture occurs in the image, the filter template reduces the weight assignments of the cross-region mixed pixels to handle the issue of cross-region mixed pixels simply and effectively. To validate the effectiveness of the proposed method, experiments are carried out on three common HSIs using support vector machine (SVM) classifiers with features learned by the PF. The experimental results demonstrate that the proposed method effectively extracts the spectral-spatial features of HSIs and significantly improves the accuracy of HSI classification.

Multi-Modal Ship Target Image Smoothing Based on Adaptive Mean Shift

In this paper, we propose an adaptive image smoothing method for infrared (IR) and visual ship target images, aiming to effectively suppress noise as well as preserve important target structures, thus benefiting image segmentation. First, by analyzing the specific features of ship target images, a block based method combining local region mean and standard deviation is developed to highlight ship target regions. It is helpful to distinguish the ship target region from the background. Then, by associating the range bandwidth with local image properties of the ship target region and the background region, we develop an adaptive range bandwidth mean shift filtering method for IR and visual ship target image smoothing. With this proposed method, we can obtain a small bandwidth for ship target region and a large one for the background region. Therefore, we can effectively smooth the background while preserving the details of the targets. Experimental results show that this method works well for IR and visual ship target images with different backgrounds. The method demonstrates superior performance for image smoothing and target preservation compared with the four well-known edge-preserving denoising methods, including the anisotropic diffusion filtering, the bilateral filtering, the propagation filtering, and the mean shift filtering with fixed range bandwidth.

Consistent Coding Scheme for Single-Image Super-Resolution Via Independent Dictionaries

In this paper, we present a unified frame based on collaborative representation (CR) for single-image super-resolution (SR), which learns low-resolution (LR) and high-resolution (HR) dictionaries independently in the training stage and adopts a consistent coding scheme (CCS) to guarantee the prediction accuracy of HR coding coefficients during SR reconstruction. The independent LR and HR dictionaries are learned based on CR with $l_{2}$ -norm regularization, which can well describe the corresponding LR and HR patch space, respectively. Furthermore, a mapping function is learned to map LR coding coefficients onto the corresponding HR coding coefficients. Propagation filtering can achieve smoothing over an image while preserving image context like edges or textural regions. Moreover, to preserve the edge structures of a super-resolved image and suppress artifacts, a propagation filtering-based constraint and image nonlocal self-similarity regularization are introduced into the SR reconstruction framework. Experimental comparison with state-of-the-art single image SR algorithms validates the effectiveness of proposed approach.