Gradient-based Edge Detection Research Articles

Gauss Gradient Algorithm for Edge Detection in Retinal Optical Coherence Tomography Images

Medical imaging modalities play a vital role in the healthcare sector for disease diagnosis and prediction. Optical Coherence Tomography uses near-infrared rays for generating cross-sectional images of the retina. Edge detection is a traditionalROI extractionalgorithm thatextracts the boundary of objects in an image. This research work focuses on the gauss gradient-based edge detection model for boundary detection in Optical Coherence Tomography images of the retina. The separable feature of a 2D Gaussian kernel is used, and a 1D kernel for the x and y directions is created. The Gaussian kernel utilized in this research work is the convolution result of Gaussian function and first-order derivative of Gaussian function. For performance validation, the Berkeley segmentation data set was utilized, when compared to traditional edge detection models, and better results were obtained for the gauss gradient algorithm.

Automatic 3D CT liver segmentation based on fast global minimization of probabilistic active contour.

Automatic liver segmentation from computed tomography (CT) images is an essential preprocessing step for computer-aided diagnosis of liver diseases. However, due to the large differences in liver shapes, low-contrast to adjacent tissues, and existence of tumors or other abnormalities, liver segmentation has been very challenging. This study presents an accurate and fast liver segmentation method based on a novel probabilistic active contour (PAC) model and its fast global minimization scheme (3D-FGMPAC), which is explainable as compared with deep learningmethods. The proposed method first constructs a slice-indexed-histogram to localize the volume of interest (VOI) and estimate the probability that a voxel belongs to the liver according its intensity. The probabilistic image would be used to initialize the 3D PAC model. Secondly, a new contour indicator function, which is a component of the model, is produced by combining the gradient-based edge detection and Hessian-matrix-based surface detection. Then, a fast numerical scheme derived for the 3D PAC model is performed to evolve the initial probabilistic image into the global minimizer of the model, which is a smoothed probabilistic image showing a distinctly highlighted liver. Next, a simple region-growing strategy is applied to extract the whole liver mask from the smoothed probabilistic image. Finally, a B-spline surface is constructed to fit the patch of the rib cage to prevent possible leakage into adjacent intercostaltissues. The proposed method is evaluated on two public datasets. The average Dice score, volume overlap error, volume difference, symmetric surface distance and volume processing time are 0.96, 7.35%, 0.02%, 1.17 mm and 19.8 s for the Sliver07 dataset, and 0.95, 8.89%, , 1.45 mm and 23.08 s for the 3Dircadb dataset,respectively. The proposed fully-automatic approach can effectively segment the liver from low-contrast and complex backgrounds. The quantitative and qualitative results demonstrate that the proposed segmentation method outperforms state-of-the-art traditional automatic liver segmentation algorithms and achieves very competitive performance compared with recent deep leaning-basedmethods.

Gradient-based edge detection with skeletonization (GES) segmentation for magnetic resonance optic nerve images

This study proposes a new segmentation method called gradient-based edge detection with skeletonization (GES) for the cross-sectional optic nerve on magnetic resonance (MR) images acquired with T1-weighted fast spoiled gradient-echo (FSPGR) without fat saturation. The raw optic nerve images have very poor resolution with unclear edges. Therefore, the images were first pre-processed with bicubic interpolation to improve the spatial resolution. Then, the proposed GES segmentation was applied to produce a distinct optic nerve image. The edges of the optic nerve were identified by finding the largest gradient changes in signal intensity between the optic nerve region and its surrounding cerebrospinal fluid (CSF). Particle swarm optimization (PSO) and level set method (LSM) segmentations were applied for comparison. Manual segmentation performed by a certified radiologist was used as the ground truth for the evaluation of the computerized segmentation. GES produced a higher mean Dice similarity coefficient (DSC) index of 0.81 ± 0.04 compared to the LSM with a mean DSC index of 0.67 ± 0.17. The bicubic-GES processed optic nerve images were used for the quantitative measurement on ten normal datasets. This study has reported the quantitative values of the longest length of the optic nerve up to the chiasm (37.2 mm) using MR images. The proposed GES segmentation method for the optic nerve will be useful for investigating any optic nerve-related disease that affects the area or volume of the optic nerve.

Gradient Based Edge Detection of Traffic Image using Second Derivative Method Compared with First Derivative Method

Aim: To detect traffic images using Gradient based edge detection by Second derivative method compared with First derivative method. Materials and materials: A second derivative method is used in this study. Canny algorithm is the edge detection technique used in traffic images. With the help of this traffic has been detected in the images.26 is the sample size taken with p-value 0.8 and has been used to improve detection rate of Gradient based in terms of accuracy and sensitivity using better skewness. Result: According to the results obtained the Second derivative method has Accuracy 88.6% and Sensitivity is 83.4% compared with the accuracy of First derivative method has accuracy 83.22% and also sensitivity is 78.5%. It appears to be Accuracy (0.134) and Sensitivity (0.286) of the second derivative method compared with the First derivative method. Conclusion: it appears to be that the detection rate is better using a Second derivative method compared with the First derivative method in terms of Accuracy and Sensitivity.

Application of Gradient-based Edge Detection on Defined Surface Feature Boundary

In surface topography, the detection of edges or surface feature boundary should be made prior to the characterization process to ensure precise measurement of the surface feature. However, problem arises when the existing edge detection from the image processing area is being directly adopted on the surface without firstly defining the exact location of edges. Thus, an approach has been carried out with the aim of developing the definition of edges to identify its exact location. The process of developing the definition of edges is conducted based on theoretical understanding and demonstration on the simulated surface topography data with the help of MATLAB software. Gradient-based edge detection operators are applied to see their performance on the defined surface feature boundary. Overall findings of this study showed that there are slight discrepancies of the edge position signified by the gradient-based operators to manually-defined edges. Despite of such outcome, it is then recommended to use either the Sobel operator or the Prewitt operator as a tool to signify the edge before proceeding with characterization process in future study due to their stability and clarity in producing the surface gradient function.

Multi-modal Medical Image Fusion based on Two-scale Image Decomposition and Sparse Representation

Multimodality image fusion is the hot topic in medical imaging field which increases the clinical diagnosis accuracy through fusing complementary information of multimodality images. In this paper, a multimodal image fusion scheme is introduced based on two-scale image decomposition and sparse representation. In the proposed scheme, the source multimodal images are first processed through contrast enhancement technique so that the intensity distribution is improved for better visualization. A spatial gradient based edge detection technique is used for extracting the edge information from contrast stretched images. The enhanced multimodality images are then decomposed into two components: the base and detail layers. The final detail layer is extracted by using SSGSM. Finally, by using an enhanced decision maps and fusion scheme the fused image is obtained. The experimental results show that the proposed multimodal image fusion scheme outperforms with some others methods by performing qualitative and quantitative analysis.

Intensity gradient based edge detection for pixelated communication systems

Pixelated optical communication systems transmit data using a series of encoded pixelated images. The decoding of these pixelated images is important for reliable data communication. This study proposes a novel intensity gradient-based edge detection (IGED) method for the received pixelated images. In the first stage of the proposed edge detection algorithm, the intensity difference of neighbouring pixels is calculated in the horizontal or vertical direction depending on the presence of a horizontal or vertical edge. In the second stage, the false edge pixels whose intensity values are less than the threshold are removed. Next, slopes are calculated between the consecutive edge pixels and the appropriate lines are found considering the slopes that have almost the same value. The new IGED method performs edge detection without requiring any anchor pixel blocks. This study also provides a description of a misalignment correction algorithm suitable for pixelated images. Next, the effectiveness of the proposed IGED method is tested through an experimental demonstration. It is shown that the IGED method can properly detect the edges even for pixelated images affected by rotational misalignment. For a given transmitter resolution, the proposed method provides better data rates than the existing methods.

Post-Traumatic Cerebral Microhemorrhages and their Effects Upon White Matter Connectivity in the Aging Human Brain.

Cerebral microbleeds (CMBs), a common manifestation of mild traumatic brain injury (mTBI), have been sporadically implicated in the neurocognitive deficits of mTBI victims but their clinical significance has not been established adequately. Here we investigate the longitudinal effects of post-mTBI CMBs upon the fractional anisotropy (FA) of white matter (WM) in 21 older mTBI patients across the first ~6 months post-injury. CMBs were segmented automatically from susceptibility-weighted imaging (SWI) by leveraging the intensity gradient properties of SWI to identify CMB-related hypointensities using gradient-based edge detection. A detailed diffusion magnetic resonance imaging (dMRI) atlas of WM was used to segment and cluster tractography streamlines whose prototypes were then identified. The correlation coefficient was calculated between (A) FA values at vertices along streamline prototypes and (B) topological (along-streamline) distances between these vertices and the nearest CMB. Across subjects, the CMB identification approach achieved a sensitivity of 97.1% ± 4.7% and a precision of 72.4% ± 11.0% across subjects. The correlation coefficient was found to be negative and, additionally, statistically significant for 12.3% ± 3.5% of WM clusters (p <; 0.05, corrected), whose FA was found to decrease, on average, by 11.8% ± 5.3% across the first 6 months post-injury. These results suggest that CMBs can be associated with deleterious effects upon peri-lesional WM and highlight the vulnerability of older mTBI patients to neurovascular injury.

Determination of the penumbra width of Elekta SRS cone collimator for 6 MV FF and 6 MV FFF energies using gradient-based edge detection

The dosimetric measurements of small fields of radiation are undertaken with detectors such as: GafchromicTM films, diodes, diamond detectors and ionization chambers of small volumes. The penumbra width, e.g. the spatial distance between 80% and 20% dose, is smaller in small fields. This fact increases the curvature and hence the volume effect in the penumbra region. The accuracy in the penumbra calculation is important at the QA test for implementation of the cranial Stereotactic Radiosurgery. The objective of this work was to apply a Gradient-Based Edge Detection plugin to calculate the penumbra of a radiation beam defined with the Elekta SRS Cone Collimators for beams of 6 MV (Flattening Filter – FF) and 6 MV (Flattening Filter Free – FFF) energies using the beam profile obtained with GafchromicTM film. The results corroborate those from the literature, and they allow a quantitative evaluation of the width of dosimetric penumbra by Gradient-Based Edge Detection.

A Resolution-Free Parallel Algorithm for Image Edge Detection within the Framework of Enzymatic Numerical P Systems.

Image edge detection is a fundamental problem in image processing and computer vision, particularly in the area of feature extraction. However, the time complexity increases squarely with the increase of image resolution in conventional serial computing mode. This results in being unbearably time consuming when dealing with a large amount of image data. In this paper, a novel resolution free parallel implementation algorithm for gradient based edge detection, namely EDENP, is proposed. The key point of our method is the introduction of an enzymatic numerical P system (ENPS) to design the parallel computing algorithm for image processing for the first time. The proposed algorithm is based on a cell-like P system with a nested membrane structure containing four membranes. The start and stop of the system is controlled by the variables in the skin membrane. The calculation of edge detection is performed in the inner three membranes in a parallel way. The performance and efficiency of this algorithm are evaluated on the CUDA platform. The main advantage of EDENP is that the time complexity of can be achieved regardless of image resolution theoretically.

VHDL IMPLEMENTATION FOR EDGE DETECTION USING LOG GABOR FILTER FOR DISEASE DETECTION

Edge detection is first and essential step in the field of image processing. Detected edges play a very important role such as image enhancement, object detection, focus area selection and many more. In medical application like tonsillitis, tumor, fracture can be detected in its early stage by detecting edges of disease. There are different and many ways for edge detection, However, the most may be grouped into three categories, first order gradient, second order and optimal edge detection.. Sobel edge detection is gradient based edge detection method used for finding edges of image. Also Sobel edge detection method provide one more advantage that it having better noise sensitivity as compared to other edge detection method. Here new concept Log-Gabor filter is used for best contrast ridges, efficient noise reduction and improved edges of an images. Most image processing tools such LabVIEW are not suited for strong real-time constraints, so to overcome this problem hardware implementation FPGA used. Proposed model for disease detection is design in LabVIEW platform with NI Vision Assistant tool 14.0 .

An Efficient Gradient based Algorithm for Improving Performance of Image Edge Detection

Quality and execution time are two important factors for evaluation of edge detection algorithms. In these algorithms, there is a trade-off between quality and execution time. Some algorithms only concentrate on quality and some of them are fast and low quality. Efficient methods try to achieve high quality in a low time. This research concentrates on improvement of gradient based edge detection that is fast method and appropriate for real-time processing. The proposed algorithm reduces execution time by removing many pixels from computations. It calculates gradient and angle class of remaining pixels in a very efficient way so that it reinforces quality and locality of edges. The results of this algorithm indicated improvement of performance in comparison to Canny and LOG algorithms. General Terms Computer Vision, Image segmentation, Edge detection.

An Advance Approach to Select Initial Seed Pixel using Edge Detection

This paper proposes a method of initial seed pixel selection used for image segmentation based on edge detection technique. Initial pixel seed selection is the crucial and starting stage of image segmentation. This method is based on RGB color model. In this paper author use the gradient magnitude of the Red, Green and Blue components of true color image. Two type of information are used: non-edge information and similarity behavior of pixels to its neighbors. Edge detection technique is used to select initial seed pixel selection by computing gradient of the image intensity function. The Gradient magnitude provides image with regions having high intensity variation. The threshold value is used to obtain seed pixels and its value depends on the nature of the image. In this paper the comparative analysis of various images with different Edge Detection techniques is presented. The analysis of images with different edge detection techniques is presented using image processing tool MATLAB 7.5.0. It has been shown that gradient based edge detection techniques provides similar and better result than other edge detection techniques.

Separation of faults of electromechanical drive chain using artificial intelligence methods

Abstract In this paper a complex drive chain is modelled with Local Linear Neuro-Fuzzy Model (LLNF). The developed models were used for detecting different faults that may occur in the system. The models were developed based on measurements carried out on the real system. Using feed-forward neural networks with perceptron neuron structure, model-based fault diagnosis of the analysed system was developed to separate the different faults. The performance and efficiency of the developed different types of artificial neural network's structures were compared using gradient based edge detection method.

Segmenting images with gradient-based edge detection using Membrane Computing

In this paper, we present a parallel implementation of a new algorithm for segmenting images with gradient-based edge detection by using techniques from Natural Computing. This bio-inspired parallel algorithm has been implemented in a novel device architecture called CUDA™(Compute Unified Device Architecture). The implementation has been designed via tissue P systems on the framework of Membrane Computing. Some examples and experimental results are also presented.

Accuracy and resolution of in vitro imaging based porcine lens volumetric measurements

There is considerable interest in determining lens volume in the living eye. Lens volume is of interest to understand accommodative changes in the lens and to size accommodative IOLs (A-IOLs) to fit the capsular bag. Some studies have suggested lens volume change during accommodation. Magnetic Resonance Imaging (MRI) is the only method available to determine lens volume in vivo. MRI is, by its nature, relatively low in temporal and spatial resolution. Therefore analysis often requires determining lens volume from single image slices with relatively low resolution on which only simple image analysis methods can be used and without repeated measures. In this study, 7 T MRI scans encompassing the full lens volume were performed on 19 enucleated pig eyes. The eyes were then dissected to isolate and photograph the lens in profile and the lens volumes were measured empirically using a fluid displacement method. Lens volumes were calculated from two- and three-dimensional (2D and 3D) MR and 2D photographic profile images of the isolated lenses using several different analysis methods. Image based and actual measured lens volumes were compared. The average image-based volume of all lenses varied from the average measured volume of all lenses by 0.6%–6.4% depending on the image analysis method. Image analysis methods that use gradient based edge detection showed higher precision with actual volumes (r2: 0.957–0.990), while threshold based segmentation had poorer correlations (r2: 0.759–0.828). The root-mean-square (RMS) difference between image analysis based volumes and fluid displacement measured volumes ranged from 8.51 μl to 25.79 μl. This provides an estimate of the error of previously published methods used to calculate lens volume. Immobilized, enucleated porcine eyes permit improved MR image resolution relative to living eyes and therefore improved image analysis methods to calculate lens volume. The results show that some of the accommodative changes in lens volume reported in the literature are likely below the resolution limits of imaging methods used. MRI, even with detailed image analysis methods used here, is unlikely to achieve the resolution required to accurately size an A-IOL to the capsular bag.

Multi-scale edge detection on range and intensity images

Multi-scale feature extraction has become prominent in recent years. Additionally, processing images containing sparse or irregularly distributed data has become increasingly important, in particular with respect to the use of range image data. We present a family of multi-scale gradient-based edge detection algorithms that are suitable for use on either regularly or irregularly distributed image data; these algorithms can be applied directly to the range and intensity images without any image pre-processing. We quantitatively evaluate our algorithms on synthetic intensity and range images and also provide comparative visual output, using real images. The results demonstrate that this approach can be successfully applied to both range and intensity images, providing results that for intensity images are more accurate than from traditional gradient operators and for range images are more accurate than from the scan-line approximation.

An FPGA implementation of chaotic and edge enhanced error diffusion

Digital halftoning plays a central role in getting more observable grey-levels for either the innovative electronic paper or other less level devices. The hardware implementation of digital halftoning is, however, seldom fully explored. In this paper, we propose a novel implementation of digital halftoning by means of error diffusion. The proposed scheme not only can perform a new method called chaotic and edge enhanced error diffusion, but also can be reduced to perform the conventional Floyd-Steinberg error diffusion. Best of all, our new scheme can produce halftone images with lower worm-like artifacts and sharper image edges. This scheme is mainly composed of four components: gradient-based edge detection, chaotic threshold generation, edge enhanced quantization, and error diffusion. Each circuit design of the four components is illustrated for the first time. Besides, we demonstrate the hardware performance of our scheme by using a field programmable gate array (FPGA) chip to offer possibly further applications.

Gradient histogram: Thresholding in a region of interest for edge detection

Selecting a threshold from the gradient histogram, a histogram of gradient magnitudes, of an image plays a crucial role in a gradient based edge detection system. This paper presents a methodology to determine the threshold from a gradient histogram generated using any kind of linear gradient operator on an image. We consider the image as a random process with dependent samples, model the gradient histogram using theories of random process and random input to a system, and determine a region of interest in the gradient histogram using certain properties of a probability density function. Standard histogram thresholding techniques are then used within the region of interest to get the threshold value. To obtain the edges, this threshold value is then used as the upper threshold of the hysteresis thresholding technique that follows the non-maximum suppression operation applied on the gradient magnitude image. The proposed methodology of determining a threshold in a gradient histogram is deduced through rigorous analysis and hence it helps in achieving consistently appreciable edge detection performance. Experimental results using different real-life and benchmark images are shown to demonstrate the effectiveness of the proposed technique.

Methods to Automate Print Quality Assessment

This paper demonstrates how image processing techniques can be combined to automate print quality assessment and reduce the time to market. Typically, significant engineering resource is required to evaluate and optimize print quality over a range of environmental conditions. This method is not only tedious, but its subjective nature induces error. An automated process has been developed to quantify print quality for a variety of digitized print samples. The process uses neighborhood analysis and edge detection to automatically align the digitized image. Techniques such as gradient-based edge detection, neighborhood analysis, and thresholding are used to quantify print quality for each region of interest. Calibration curves are used to relate manually graded images to computationally graded images. Finally, a weighted normalized least-squares fitting routine transforms print quality metrics for each environmental condition into a table of optimized electrophotographic voltage settings.