Laplacian Filter Research Articles

CoronaNet: A Novel Deep Learning Model for COVID-19 Detection in CT Scans

Coronavirus disease (COVID-19) is currently the cause of a global pandemic that is affecting millions of people around the world. Inadequate testing resources have resulted in several people going undiagnosed and consequently untreated; however, using computerized tomography (CT) scans for diagnosis is an alternative to bypass this limitation. Unfortunately, CT scan analysis is time-consuming and labor intensive and rendering is generally infeasible in most diagnosis situations. In order to alleviate this problem, previous studies have utilized multiple deep learning techniques to analyze biomedical images such as CT scans. Specifically, convolutional neural networks (CNNs) have been shown to provide medical diagnosis with a high degree of accuracy. A common issue in the training of CNNs for biomedical applications is the requirement of large datasets. In this paper, we propose the use of affine transformations to artificially magnify the size of our dataset. Additionally, we propose the use of the Laplace filter to increase feature detection in CT scan analysis. We then feed the preprocessed images to a novel deep CNN structure: CoronaNet. We find that the use of the Laplace filter significantly increases the performance of CoronaNet across all metrics. Additionally, we find that affine transformations successfully magnify the dataset without resulting in high degrees of overfitting. Specifically, we achieved an accuracy of 92% and an F1 of 0.8735. Our novel research describes the potential of the Laplace filter to significantly increase deep CNN performance in biomedical applications such as COVID-19 diagnosis.

Image pan-sharpening using enhancement based approaches in remote sensing

This paper proposes to do image enhancement before pan-sharpening; that is, the image enhancement techniques are used as a pre-processing step. The image enhancement techniques are proposed in two domains, same-domain and cross-domain. In the same-domain methods, the image enhancement techniques (such as Laplacian, Unsharp) are simply applied to multispectral (MS) and panchromatic (PAN) images to sharpen both images in the spatial domain. While in cross-domain, a novel hybrid combination of Laplacian Filter (LF) and Discrete Fourier Transformation (DFT) image sharpening technique is introduced. After image enhancement, the powerful Matting Model (MM) pan-sharpening technique is used to fuse both the enhanced images and produce a resultant image with the high spatial and spectral resolutions. The experimental results of the proposed approach outperform the others as compared to the state-of-art techniques over three datasets. The results are evaluated, considering both Qualitative and Quantitative evaluation metrics.

Investigation of brain computer interface for rich multimedia environment

SummaryThe brain computer interface (BCI) can provide a direct channel of communication between an external device and the brain without including any type of muscular activity. Various applications of BCI have been stated in literature, and one of the most exciting areas which have not been extensively investigated is the field of multimedia. BCI can find huge potential in the area of command and control of video games. Video games have found huge potential for improving the gait and balance of people struggling with Parkinson's disease. Such systems make use of the brain signals which are captured through electrodes on scalp as electroencephalogram (EEG) or through implanted electrodes as electrocorticography (ECoG). The signals are then used as input commands to get desired output in an external device. To address the memory management issue arising due to the huge volume of ECoG data and to achieve distributed computation for improving the efficiency of BCI, the computations are implemented in Cloud. In this work, ECoG signals are pre‐processed, and all features are extracted by using the wavelet packet transform (WPT), common spatial patterns (CSPs), Laplacian filter, and the Kalman filter. The Adaboost classifier was employed for ECoG signal classification.

A Novel Radiometric Control Set Sample Selection Strategy for Relative Radiometric Normalization of Multitemporal Satellite Images

This article presents a new relative radiometric normalization (RRN) method for multitemporal satellite images based on the automatic selection and multistep optimization of the radiometric control set samples (RCSS). A novel image-fusion strategy based on the fast local Laplacian filter is employed to generate a difference index using the complementary information extracted from the change vector analysis and absolute gradient difference of the bitemporal satellite images. The difference index is then segmented into changed and unchanged pixels using a fast level-set method. A novel local outlier method is then applied to the unchanged pixels of the bitemporal images to identify the initial RCSS, which are then scored by a novel unchanged purity index, and the histogram of the scores is used to produce the final RCSS. The RRN between the bitemporal images is achieved by adjusting the subject image to the reference image using orthogonal linear regression on the final RCSS. The proposed method is applied to seven different data sets comprised of bitemporal images acquired by various satellites, including Landsat TM/ETM+, Sentinel 2B, Worldview 2/3, and Aster. The experimental results show that the method outperforms the state-of-the-art RRN methods. It reduces the average root-mean-square error (RMSE) of the best baseline method (IR-MAD) by up to 32% considering all data sets.

Improved Single Haze Removal using Weighted Filter and Gaussian-Laplacian

Generally computer applications use digital images. Digital image plays a vital role in the analysis and explanation of data, which is in the digital form. Images and videos of outside scenes are generally affected by the bad weather environment such as haze, fog, mist etc. It will result in bad visibility of the scene caused by the lack of quality. This paper exhibits a study about various image defogging techniques to eject the haze from the fog images caught in true world to recuperate a fast and enhanced nature of fog free images. In this paper, we propose a simple but effective the weighted median (WM) filter was first presented as an overview of the standard median filter, where a nonnegative integer weight is assigned to each position in the filter window image .Gaussian and laplacian pyramids are applying Gaussian and laplacian filter in an image in cascade order with different kernel sizes of gaussian and laplacian filter .The dark channel prior is a type of statistics of the haze-free outdoor images. It is based on a key observation - most local patches in haze-free outdoor images contain some pixels which have very low intensities in at least one-color channel. Using this prior with the haze imaging model, we can directly estimate the thickness of the haze and recover a high-quality haze-free image. Results on a variety of outdoor haze images demonstrate the power of the proposed prior. Moreover, a high-quality depth map can also be obtained as a by-product of haze removal and Calculate the PSNR and MSE of three sample images.

Constraining Euler Deconvolution Solutions Through Combined Tilt Derivative Filters

Local phase filters as the tilt derivative (TDR) and the horizontal tilt derivative (TDX) are extensively used to interpret magnetic data. We use two combinations of these filters, namely TDR − TDX and TDR + TDX, to design a constraining mask that guides the Euler deconvolution moving data window. The TDR − TDX filter produces sharp peaks over the centers of the sources, while the TDR + TDX filter generates plateaus over them. Motivated by previous approaches that make use of the Laplacian filter or the analytic signal to constrain the Euler deconvolution window, we compute the solutions for windows centered at points that (1) have positive values of TDR − TDX and (2) are contained in the plateaus of TDR + TDX. The use of both criteria improves the selection of source-related points while reducing the number of spurious ones. Our method is tested in synthetic anomalies due to interfering dike-like sources and field data from southeast Brazil. The experiments show that the use of a constraining mask based on combined tilt filters produce Euler solutions that are more contiguous and less sensitive to noise than the traditional located-Euler deconvolution.

Imaging Enhancement in Angle-Domain Common-Image-Gathers Using the Connected-Component Labeling Method

The quality of seismic migration images strongly depends on accurate velocity models. However, migration velocity models often contain errors causing artifacts which degrade seismic resolution and fidelity. Here, we propose a method to enhance seismic migration images using processes in angle-domain common-image-gathers (ADCIGs), without requiring modification to the velocity model. The ADCIGs impacted from velocity errors show non-flat gathers and scattered noise. To enhance the imaging quality in ADCIGs, these problems need to be treated prior to stacking. Ideally, layers in ADCIGs should be flat without smearing amplitudes through all angles. Using this principle, we iteratively process one reflection at a time by isolating it into a local window and then flattening it. Also, we apply internal processing steps to enhance the signal. The algorithm for segregating a reflection, referred to as the connected-component labeling (CCL) method, is the primary method for extracting any feature that has a continuous form within the ADCIGs. Moreover, this method is insensitive to scattered noise so it is a suitable approach for classifying reflections or noise. To test the efficiency of this algorithm, we create ADCIGs containing relatively poor-quality images using an inaccurate migration velocity model. The primary objective of this test is to show how well the CCL method can select various forms of reflections in the presence of various levels of noise. The results show that the CCL method is capable of individually decomposing a reflection, which can facilitate processing in the internal workflow. Finally, we evaluate the quality of processed migration images by comparing our processed Poynting-vector reverse time migration (PVRTM) images with reverse time migration applied Laplacian filter and partial stacked PVRTM. These comparable migration images are benchmarked against a baseline synthetic reflector model. The essential improvements of our method are addressed by the migration sections, namely the removal of diffraction artifacts at both small and large scales and the improvement of phase coherency. The final stacked migration image produced with our methods shows clearer geological features that are capable of delivering a more reliable seismic interpretation.

High-Performance CML approximate full adders for image processing application of laplace transform

PurposeThe high demand for fast, energy-efficient, compact computational blocks in digital electronics has led the researchers to use approximate computing in applications where inaccuracy of outputs is tolerable. The purpose of this paper is to present two ultra-high-speed current-mode approximate full adders (FA) by using carbon nanotube field-effect transistors.Design/methodology/approachInstead of using threshold detectors, which are common elements in current-mode logic, diodes are used to stabilize voltage. Zener diodes and ultra-low-power diodes are used within the first and second proposed designs, respectively. This innovation eliminates threshold detectors from critical path and makes it shorter. Then, the new adders are employed in the image processing application of Laplace filter, which detects edges in an image.FindingsSimulation results demonstrate very high-speed operation for the first and second proposed designs, which are, respectively, 44.7 per cent and 21.6 per cent faster than the next high-speed adder cell. In addition, they make a reasonable compromise between power-delay product (PDP) and other important evaluating factors in the context of approximate computing. They have very few transistors and very low total error distance. In addition, they do not propagate error to higher bit positions by generating output carry correctly. According to the investigations, up to four inexact FA can be used in the Laplace filter computations without a significant image quality loss. The employment of the first and second proposed designs results in 42.4 per cent and 32.2 per cent PDP reduction compared to when no approximate FA are used in an 8-bit ripple adder.Originality/valueTwo new current-mode inexact FA are presented. They use diodes as voltage regulators to design current-mode approximate full-adders with very short critical path for the first time.

Real-world fence removal from a single-image via deep neural network

At public space such as a zoo and sports facilities, the presence of fence often annoys tourists and professional photographers. There is a demand for a post-processing tool to produce a non-occluded view from an image or video. This “de-fencing” task is divided into two stages: one is to detect fence regions and the other is to fill the missing part. For a decade or more, various methods have been proposed for video-based de-fencing. However, only a few single-image-based methods are proposed. In this paper, we mainly focus on single-image fence removal. Conventional approaches suffer from inaccurate and non-robust fence detection and inpainting due to less content information. To solve these problems, we combine novel methods based on a deep convolutional neural network (CNN) and classical domain knowledge in image processing. In the training process, we are required to obtain both fence images and corresponding non-fence ground truth images. Therefore, we synthesize natural fence image from real images. Moreover, spacial filtering processing (e.g. a Laplacian filter and a Gaussian filter) improves the performance of the CNN for detecting and inpainting. Our proposed method can automatically detect a fence and generate a clean image without any user input. Experimental results demonstrate that our method is effective for a broad range of fence images.

A brain connectivity characterization of children with different levels of mathematical achievement based on graph metrics.

Recent studies aiming to facilitate mathematical skill development in primary school children have explored the electrophysiological characteristics associated with different levels of arithmetic achievement. The present work introduces an alternative EEG signal characterization using graph metrics and, based on such features, a classification analysis using a decision tree model. This proposal aims to identify group differences in brain connectivity networks with respect to mathematical skills in elementary school children. The methods of analysis utilized were signal-processing (EEG artifact removal, Laplacian filtering, and magnitude square coherence measurement) and the characterization (Graph metrics) and classification (Decision Tree) of EEG signals recorded during performance of a numerical comparison task. Our results suggest that the analysis of quantitative EEG frequency-band parameters can be used successfully to discriminate several levels of arithmetic achievement. Specifically, the most significant results showed an accuracy of 80.00% (α band), 78.33% (δ band), and 76.67% (θ band) in differentiating high-skilled participants from low-skilled ones, averaged-skilled subjects from all others, and averaged-skilled participants from low-skilled ones, respectively. The use of a decision tree tool during the classification stage allows the identification of several brain areas that seem to be more specialized in numerical processing.

Single-Image Fence Removal Using Deep Convolutional Neural Network

In public spaces such as zoos and sports facilities, the presence of fences often annoys tourists and professional photographers. There is a demand for a post-processing tool to produce a non-occluded view from an image or video. This “de-fencing” task is divided into two stages: one to detect fence regions and the other to fill the missing part. For over a decade, various methods have been proposed for video-based de-fencing. However, only a few single-image-based methods are proposed. In this paper, we focus on single-image fence removal. Conventional approaches suffer from inaccurate and non-robust fence detection and inpainting due to less content information. To solve these problems, we combine novel methods based on a deep convolutional neural network (CNN) and classical domain knowledge in image processing. In the training process, we are required to obtain both fence images and corresponding non-fence ground truth images. Therefore, we synthesize natural fence images from real images. Moreover, spacial filtering processing (e.g. a Laplacian filter and a Gaussian filter) improves the performance of the CNN for detection and inpainting. Our proposed method can automatically detect a fence and generate a clean image without any user input. Experimental results demonstrate that our method is effective for a broad range of fence images.

Preprocessing and Feature Extraction for Psoriasis Images Based on Discrete Wavelet Transform

This paper, proposes the pre-processing methods use a small neighborhood of a pixel in the input image to get a new brightness value in the output image. Such pre-processing operations are also called filtration. in this paper, we start with the image using a medical case for psoriasis image after change it to gray state implemented under the transform domain (i.e frequency), using wavelet transform then use three filters sharpening, Sobel, and Laplace filter. after make proposed by computing PSNR for each state to show the effect of it. Then extract features through an apply a set of measures (Energy, Entropy, Standard deviation, Variance, Mean) of low low sub-image. The proposed system was implemented on the medical case for psoriasis image dataset, some of them were obtained from the hospitals and the other was obtained from the dataset (Light Field Image of Dataset skin Lesions), available on the Internet and the proposed system implemented in programing language Visual Basic 6.0.

Multi Focus Image Fusion using Image Enhancement Techniques with Wavelet Transformation

Multi-focus image fusion produces a unification of multiple images having different areas in focus, which contain necessary and detailed information in the individual image. The paper is proposing a novel idea in a pre-processing step in the image fusion environment in which sharpening techniques applied before fusion in the pre-processing step. This article is proposing multi-focus hybrid techniques for fusion, based on image enhancement, which helps to identify the key features and minor details and then fusion performed on the enhanced images. In image enhancement, we introduced a new hybrid sharpening method that combines Laplacian Filter (LF) with a Discrete Fourier Transform (DFT) and also performs sharpening using the Unsharp sharpen approach. Then fusion is performed using Stationary Wavelet Transformation (SWT) technique to fused the enhanced images and obtaining more detail of the resultant image. The proposed approach is applied to two image sets, e., the “planes” and “clocks” image sets. The quality of the output image evaluated using both qualitative and quantitative approaches. Four will know quantitative metrics used to assess the performance of the novel technique. The experimental results of the novel methods showed efficient, improved outcomes and better for multi-focused image fusion. The SWT (LF+DFT) and SWT (Unsharp Mask) are 2.6 %, 1.8%, and 0.62%, 0.61% better than the best baseline measure, i.e., SWT, considering RMSE (Root Mean Square Error) for both image sets.

A Lung Dense Deep Convolution Neural Network for Robust Lung Parenchyma Segmentation

Lung parenchyma segmentation is the prerequisite for an automatic diagnosis system to analyze lung CT (computed tomography) images. However, traditional lung segmentation algorithms have poor adaptability and are not effectively robust regarding lung databases with blood vessels and small voids which can interfere the segmentation. The main work of this paper is as follows: Firstly, a lung dense deep convolutional neural network (LDDNet) is proposed, which adopts some popular optimizer methods, such as dense block, batch normalization (BN) and dropout. The performance of LDDNet is tested on the public lung database LIDC-IDRI which contains many cases of interference for segmentation. Secondly, the labeled with blood vessels and small voids are not contained by the public ground-truth masks of the LIDC-IDRI database, therefore these regions are labeled by us with LabelMe software. Thirdly, for the aim of exploring the effect of image preprocessing on segmenting lung CT images with deep neural network, contrast enhancing, median filtering and Laplacian filtering are used to preprocess the image as comparative experiments. Finally, dataset is classified into four classes by the geometrical shapes to test the performance of LDDNet. The accuracy of the segmentation experiment reaches over 99% and the four classes can all reach over 95%. Additionally, blood vessels and small voids are segmented out from the lung parenchyma which is not achieved by other methods. Experimental results confirm that the proposed LDDNet can segment the lung parenchymal area more accurately and has better robustness in comparison with other neural networks and most of the traditional methods.

A Novel Image Dehazing Algorithm via Adaptive Gamma-Correction and Modified AMEF

Captured images are usually influenced by fog or haze. In reality, image dehazing is challenging. This paper proposes a modified artificial multiple-exposure image fusion (AMEF) algorithm to remove the haze from an image. In the algorithm, first, an adaptive gamma-correction transform with the mean and standard deviation values of each component of a hazy image is utilized to verify the intensities. Second, the homomorphic filtering algorithm is introduced into the Gaussian pyramid and Laplacian pyramid to compute the exposed accessible images. Last, a modified Laplacian filter method is presented to calculate the contrast of the exposed accessible images. Further, extensive experimental results demonstrate that the proposed algorithm has superior performance compared with that of some state-of-the-art methods, including higher contrast, richer details and a better visual effect in the dehazed image.

A New Fast Local Laplacian Completed Local Ternary Count (FLL-CLTC) for Facial Image Classification

Face recognition is one of the most interesting areas of research areas because of its importance in authentication and security. Differentiating between different facial images is not easy because of the similarities in facial features. Human faces can also be covered obscured by eyeglasses, facial expressions and hairstyles can also be changed causing difficulty in finding similar faces. Thus, the need for powerful image features has become a critical issue in the face recognition systems. Many texture features have been used in these systems, including Local Binary Pattern (LBP), Local Ternary Pattern (LTP), Completed Local Binary Pattern (CLBP), Completed Local Binary Count (CLBC) and Completed Local Ternary Pattern (CLTP). In this paper, a new texture descriptor, namely, Completed Local Ternary Count (CLTC), is proposed by adding a threshold value for the CLBC to overcome its sensitivity to noise drawback. The CLTC is also enhanced by adding the Fast-Local Laplacian filter during the pre-processing stage to increase the discriminative property of the proposed descriptor. The proposed Fast-Local Laplacian CLTC (FLL-CLTC) texture descriptor is evaluated for face recognition task using five different face image datasets. The experimental results of the FLL-CLTC showed that the proposed FLL-CLTC outperformed the CLBP and CLTP texture descriptors in term of recognition accuracy. The FLL-CLTC achieved 99.1%, 86.93%, 93.21%, 84.92% and 99.15% with JAFFE, YALE, Georgia Tech, Caltech and ORL face image datasets, respectively.

Graph Signal Processing Approach to QSAR/QSPR Model Learning of Compounds.

Quantitative relationship between the activity/property and the structure of compound is critical in chemical applications. To learn this quantitative relationship, hundreds of molecular descriptors have been designed to describe the structure, mainly based on the properties of vertices and edges of molecular graph. However, many descriptors degenerate to the same values for different compounds with the same molecular graph, resulting in model failure. In this paper, we design a multidimensional signal for each vertex of the molecular graph to derive new descriptors with higher discriminability. We treat the new and traditional descriptors as the signals on the descriptor graph learned from the descriptor data, and enhance descriptor dissimilarity using the Laplacian filter derived from the descriptor graph. Combining these with model learning techniques, we propose a graph signal processing based approach to obtain reliable new models for learning the quantitative relationship and predicting the properties of compounds. We also provide insights from chemistry for the boiling point model. Several experiments are presented to demonstrate the validity, effectiveness and advantages of the proposed approach.

Saliency Improvement in Feature-Poor Surgical Environments Using Local Laplacian of Specified Histograms

Navigation in endoscopic environments requires an accurate and robust localisation system. A key challenge in such environments is the paucity of visual features that hinders accurate tracking. This article examines the performance of three image enhancement techniques for tracking under such feature-poor conditions including Contrast Limited Adaptive Histogram Specification (CLAHS), Fast Local Laplacian Filtering (LLAP) and a new combination of the two coined Local Laplacian of Specified Histograms (LLSH). Two cadaveric knee arthroscopic datasets and an underwater seabed inspection dataset are used for the analysis, where results are interpreted by defining visual saliency as the number of correctly matched key-point (SIFT and SURF) features. Experimental results show a significant improvement in contrast quality and feature matching performance when image enhancement techniques are used. Results also demonstrate the LLSHs ability to vastly improve SURF tracking performance indicating more than 87% of successfully matched frames. A comparative analysis provides some important insights useful in the design of vision-based navigation for autonomous agents in feature-poor environments.

Capabilities of UAV-Based Watershed Segmentation Method for Estimating Tree Crown: A Case Study of Oil Palm Tree

Currently, the tree crown delineation has become an important part of the tree parameter measurement in particular for the forest management plan. Moreover, the requirement for the advanced information on forest resources has headed researchers to built-up more effective approaches. With the advanced in the technology of imagery, an Unmanned Aerial Vehicle (UAV) has become one of the appearance tools that offers an affordable, cheaper and faster technology in acquiring data for numerous applications. More importantly, this technology has its own efficient approaches for real-time acquisition of high resolution to produce three-dimensional (3D) information. The digital surface model was extracted and used to estimate tree crown. The influences of different filtering method such as Gaussian and Laplacian filter were investigated on tree crown estimation. The result shows the differences in tree crown delineation that were confirmed by processing methods. This study has potential for future use by tree parameters that related researcher, industries and farmers to estimate tree height and tree counting.

Integration of Solar PV With Low-Voltage Weak Grid System: Using Normalized Laplacian Kernel Adaptive Kalman Filter and Learning Based InC Algorithm

This paper proposes a novel normalized Laplacian kernel adaptive Kalman filter (NLKAKF) based control technique and learning based incremental conductance (LIC) maximum power point tracking (MPPT) algorithm, for low-voltage weak grid-integrated solar photovoltaic (PV) system. Here, a two-stage topology of three-phase grid integrated solar PV system is implemented, where the loads are connected at the point of common coupling. Proposed LIC is the improved form of incremental conductance (InC) algorithm, where inherent problems of traditional InC technique, such as steady-state oscillation, slow dynamic responses, and fixed step size issues, are successfully mitigated. The prime objective of proposed NLKAKF control is to meet the active power requirement of the loads from generated solar PV power, and after feeding load, excess power is fed to the grid. However, when generated PV power is less than the required load power, then NLKAKF control meets the load by taking extra required power from the grid. During this process, power quality is improved at the grid. The controller action provides reactive power compensation, power factor correction, and harmonics filtering and mitigation of other power quality issues. Moreover, when the solar irradiation is zero than voltage source converter acts as a distribution static compensator (DSTATCOM), which enhances the utilization factor of the system. The proposed techniques are modeled and their performances are verified experimentally on a developed prototype, in solar insolation variation conditions, unbalanced loading, as well as in different grid disturbances such as overvoltage, undervoltage, phase imbalance, harmonics distortion in the grid voltage, etc.