Search

An overview of the various methods and algorithms for edge detection in digital images is given in this survey paper. A fundamental issue in image processing is edge detection, which involves in locating the borders which separates various regions in an image. The survey starts with an explanation of edges and their importance in an image analysis, then reviews some of the most widely used edge detection techniques, including the canny edge detector, the Sobel operator, and the Laplacian of gaussian. The following section discusses some of the drawbacks and shortcomings of these techniques. A discussion of current trends and potential future directions in the field of edge detection rounds out the survey. [ABSTRACT FROM AUTHOR]

Xilinx System Generator (XSG) integrates with MATLAB Simulink to build a complex system through Xilinx and Simulink blocks. This paper presents a hardware implementation of various image processing techniques to modify and extract the pixel information from various input images in run time. Complex coding strategies are eradicated with integrated system. The intricacy in structural design is reduced to a huge extent using Xilinx System generator. The reconfigurable attribute of Field Programmable Gate Array (FPGA) bags the massive advantages in image processing like co simulating with selective pixel and verified on FPGA. Image processing algorithms explored in methodical manner by performing diverse operations like edge detection, and color enhancement, complement, contrast stretching etc. The proposed image processing techniques (IPT) intended for various applications. Dissimilar image processing techniques are processed by FPGA and displayed on Video Graphics Array (VGA). Image processing system is designed in XSG, hardware description language (HDL) like VHDL & Verilog code is generated by processing system by system generator token. Design is implemented on Artix-7 and targeted on Basys3 reconfigurable hardware (XC7A35T-1-CPG236C). [ABSTRACT FROM AUTHOR]

One of the most essential methods of image processing is the detection of the edge because the edges represent the most important characteristics of the image through which we can analyze these images and processing after the implementation of edge detection devices in the images is necessary to increase the speed of applications in real-time. The importance of programmable logic gate (FBGA) chipset is that it provides a platform for the processing of algorithms that can work on machines in real-time while delivering much higher performance than digital signal processors (DSPs). Through this research, we present a method we have proposed for detecting these edges using the coefficients Sobel, Robert Prewitt, and Gaussian (log) Laplace. This was done through the use of a graphical consumer interface that combines Matlab, Xilinx System, and Generator (XSG) Simulink to generate a code applied on the chip (FPGA) of the type Spartan-3E. This method differed from the other by giving the edges in the place where digital images are obtained, with a good rate of similarity compared to the software environment, with high speed of implementation with lower consumption of FPGA chip. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) becomes the backbone of the advanced countries and it has a real contribute to exchange the traditional style or way of practical life, even personal life into smart style, with (IoT) technology the life become more and more easy and professional. Internet of Things achieves various applications coordinate with sensors and standard protocols to apply what is called machine -to- machine connection (M2M), in this paper we will talk more about the concept of (M2M), the main component of internet of things and finally the common protocols that is used in network, in addition to that this work present an IOT operation with processing system using camera for capturing image and Xilinx system generator(XSG)models for designing image processing algorithms and the result of the processing is an image with black and white for edge detection, thresholding models and gray color image for gray enhancement model. [ABSTRACT FROM AUTHOR]

Angiography is an important class of imaging technique where an image is obtained using an X-ray after a contrast agent is injected into the patient's blood vessels. Coronary angiography is a critical type of angiography image as it is used to diagnose serious heart and artery diseases. However, most angiography images suffer from non-uniform illumination and noises, which makes it difficult for doctors to make accurate diagnoses. Processing of angiograms is necessary to improve their visibility. This paper investigates four classical edge detection algorithms to delineate edges of the blood vessels and compares them with a proposed Sobel-based operator. The classical techniques are Roberts' cross, Prewitt, Sobel, and Canny operator. Based on the quantitative evaluation, the classical Sobel operator was found to be the most suitable operator. The modified Sobel operator was found to produce noisy images. Further adjustments are required before the proposed operator is suitable to detect edges in coronary angiography images. [ABSTRACT FROM AUTHOR]

Image Denoising, Deblurring, and Enhancement techniques are most commonly used on the images to reduce or completely remove the noise. The noisy images are not capable of perfectly using the solution of the desired problems such as object edge detection, object segmentation, and object classification. Because the edges of these objects have much bluer or more noise pixels for clearly stable detection. The main motivation of this study is to solve the problem of detection and elimination of the undesired noisy pixels on the encountered images. These noisy pixels are needed to be removed from the obtained images which are the results of throughout determined image processing steps. The main reason for this study motivation come out that to find the actual inscription fault has some noisy (undesired) pixels when extracting the true difference between the two images has been subtracted from each other. The subtracted image results are used in the inscription inspection process that controls the accuracy of the inscription quality. In the inscription inspection process, subtracted image results are used to determine the accuracy of the inscription quality. These subtracted images are formed by subtraction from each other of reference images and sample images. In these subtracted images if truly exist inscriptions faults or sometimes that could be occurred undesired noisy pixels at the same time in the subtraction process. This study focused on detecting and eliminating the undesired noisy pixels in order to reach actual inscription faults in the images. Thus, the remove pixels as width algorithm (RPW) has been developed and applied to these specified images. [ABSTRACT FROM AUTHOR]

The advances of photography technology on Android smartphone cameras to develop one of its features, namely edge detection. Edge Detection is a process that produces the edges of image objects, the goal is to improve blurred image details that occur due to errors or the effects of the image acquisition process. There are several methods to support the work of edge detection, namely the Canny operator. One of the best edge detection methods is canny edge detection which has a minimum error rate compared to other edge detection methods. Canny edge detection will produce the edge line of an object in real time on the android smartphone camera. [ABSTRACT FROM AUTHOR]

3D face reconstruction from 2D images is an important research topic because it supports a wide range of applications, such as face recognition, animations, games, and AR/VR systems. 3D face reconstruction from contour features is a challenging task, because traditional edge detection algorithms produce a lot of noises, which are prone to making the reconstruction model trapped in a local optimum or even being degraded. With the development of deep learning, a lot of researcher introduce neural network into contour detection, which can extract relatively clear contours compared with previous methods. In this article, we employ a hard blended face contour feature from neural network and canny edge extractor for face reconstruction. Our method not only improves the 3D face model reconstruction accuracy on synthesis images, but performs more accurately and robustly on in-the-wild images under blurriness, makeup, occlusion and ill-illumination conditions. [ABSTRACT FROM AUTHOR]

The mass transfer of aqueous solution in liquid has been systematically investigated by using refracted laser light through diffusing solution. The camera is connected to a computer to perform image processing in the form of some image filtration process and the algorithm in terms of obtaining the digitized value of the coordinates laser beam caught on the screen. The subtract filtration process was introduced to eliminate the effect of unwanted images on the screen. Furthermore, some algorithms based on OpenCV programming languages, such as grayscale and edge detection are applied to exhibit meaningful recorded laser beam traces and firmly determined the coordinate values. For complete discussion, performance of this system is proposed based on NaCl solution as a material of observation exhibits the diffusivity value 1.3864 x 10-5 cm²/s using this method. [ABSTRACT FROM AUTHOR]

This paper deals with setting of the acquisition equipment used to visualize and measure the concentration of intermediate species chemiluminescence combustion and laser-induced fluorescence. Observed intermediates such as OH, NO, CH and others have an emitted UV spectrum and their signal is very weak. For this reason, it is necessary to use a suitable image intensifier, and its proper setting and synchronization with the camera is also very important. In view of this problem, the following text deals with the appropriate setting of selected optical amplifiers with P46 and P43. Because the recording equipment can be used for both LIF and chemiluminescence, the experimentally determined setting will be performed using the OH-chemiluminescence method. This is reported by a weaker signal and creates a more stringent setup criterion. As a qualitative factor, a gradient of the image function has been selected, which plays a significant role in edge detection and therefore also in flame front position detection. [ABSTRACT FROM AUTHOR]

The present paper proposes image edge detection algorithm utilizing a reaction-diffusion network. The network consists of two-dimensionally coupled FitzHugh-Nagumo type neurons, whose behavior is described by a set of time-evolving partial differential equations. The neurons' non-linearity and their interconnection give the image edges through an iterative computation. The proposed algorithm differs from its previously-proposed one (Nomura et al., 2016) on the pre-processing for initializing the neurons' state and the post-processing for thinning edges. The performance of the proposed algorithm is discussed with three image data-sets and two error measures, in comparison with performance of two other algorithms including Canny's edge detection method. [ABSTRACT FROM AUTHOR]

Comparing to optical images, sonar images experience more severe conditions such as degradation, light scattering, low contrast, and blurring. To guarantee the performance of the post processing such as object detection and classification, the need of a proper denoising method for sonar images has claimed. Conventional wavelet transform(WT) domain denoising method may be good candidates but they do not properly work for some cases which have ambiguity in edge and noise separation. Since image blurring after denoising is generally caused by the loss of edge information with low signal-to-noise ratio, the denoising performance may be improved by the introduction of a proper edge decision rule. In this paper, we propose a stationary WT denoising method with edge map and non-local means(NLM) filter for edge decision. Experimental results show that the proposed method has better performance than the three conventional methods(Visushrink, Sureshrink, and Bayesshrink) in terms of the edge preservation and the peak signal-to-noise ratio. [ABSTRACT FROM AUTHOR]

This paper will focus on the study and identifying various threshold values for two commonly used edge detection techniques, which are Sobel and Canny Edge detection. The idea is to determine which values are apt in giving accurate results in identifying a particular leukemic cell. In addition, evaluating suitability of edge detectors are also essential as feature extraction of the cell depends greatly on image segmentation (edge detection). Firstly, an image of M7 subtype of Acute Myelocytic Leukemia (AML) is chosen due to its diagnosing which were found lacking. Next, for an enhancement in image quality, noise filters are applied. Hence, by comparing images with no filter, median and average filter, useful information can be acquired. Each threshold value is fixed with value 0, 0.25 and 0.5. From the investigation found, without any filter, Canny with a threshold value of 0.5 yields the best result. [ABSTRACT FROM AUTHOR]

The traditional Canny algorithm has poor self-adaptability threshold, and it is more sensitive to noise. In order to overcome these drawbacks, this paper proposed a new edge detection method based on Canny algorithm. Firstly, the media filtering and filtering based on the method of Euclidean distance are adopted to process it; secondly using the Freichen algorithm to calculate gradient amplitude; finally, using the Otsu algorithm to calculate partial gradient amplitude operation to get images of thresholds value, then find the average of all thresholds that had been calculated, half of the average is high threshold value, and the half of the high threshold value is low threshold value. Experiment results show that this new method can effectively suppress noise disturbance, keep the edge information, and also improve the edge detection accuracy. [ABSTRACT FROM AUTHOR]

The problem of the automatic image calibration is considered in this paper. The most challenging task of the automatic calibration is a proper detection of the calibration object. The solving of this problem required the appliance of the methods and algorithms of the digital image processing, such as morphology, filtering, edge detection, shape approximation. The step-by-step process of the development of the algorithm and its adopting to the specific conditions of the log cuts in the image's background is presented. Testing of the automatic calibration module was carrying out under the conditions of the production process of the logging enterprise. Through the tests the average possibility of the automatic isolating of the calibration object is 86.1% in the absence of the type 1 errors. The algorithm was implemented in the automatic calibration module within the mobile software for the log deck volume measurement. [ABSTRACT FROM AUTHOR]

This paper presents a research methodology to detect fingertips from an image which contains an outstretched hand. The research is divided into three stages. In stage one, a partitioning method is developed to perform image noise reduction. The main idea for this method is to divide the image into small pieces and analyse number of skin pixels in each piece. Image pieces which contain less density in skin pixels are considered as noise and discarded. In stage two, a new hybrid convex hull algorithm is developed to extract the convex hull vertices directly from the de-noised image in the previous stage without going through the edge detection processes. Jarvis March and Bresenham Algorithm are used to perform the extraction of seven vertices of the convex hull. In stage three, every edge of convex hull is calculated by using Distance Formula and analysed for prediction of the fingertips location. The two longest edges in the convex hull are taken as the edges connecting wrist-thumb and wrist-pinky fingers. The presented methodology works for various orientations of an outstretched hand. We aim to simplify the fingertips detection processes by avoid using complex calculation or histogram analysis compare to other methodologies. [ABSTRACT FROM AUTHOR]

The Muon g-2/EDM proposed experiment at J-PARC is a promising and innovative attempt at the field of Precision Physics. The sensitivity goal of 0.1 ppm will test the limits of our current understanding, and may probe for Beyond the Standard Model observations. This paper seeks out to investigate the computational techniques required by the experiment. The GEANT4 [1] framework was used to simulate the detector setup, according to the experiment's Conceptual Design Report (CDR) [2]. This allowed to observe the event hierarchy in different energies, generate signal hit data, and construct an event-selection algorithm. ROOT and GDML enabled us to use the geometry and parsed output data in a platform-independent way. Using techniques pertaining to Machine Learning and Image Feature extraction, such as the Canny Edge detection and the Hough Transform, we were able to construct a generic representation of 'track families' from each event category. Finally, the modular GENFIT2 [3] framework was used to implement the Kalman Filter [4] along with an Deterministic Annealing Filter (DAF) [5] and the Runge-Kutta stepper to reconstruct tracks from a few digitized, smeared singular event data. [ABSTRACT FROM AUTHOR]

A novel image edge detection technique is proposed in this paper. The considered boundary detection approach uses a nonlinear parabolic partial differential equation (PDE) - based model. We develop a nonlinear anisotropic diffusion scheme consisting in a second-order parabolic equation and some boundary conditions. An effective edgestopping function is constructed for this PDE model. Then, we construct an explicit iterative numerical approximation scheme for this model, by using the finite-difference method. This numerical discretization scheme is applied on the analyzed image and two results, achieved after different iterations, are considered. The image representing the absolute difference between those two states is then determined and a thresholding procedure is then applied to it. Some morphology-based operations are applied on that image, its edges being determined from these morphologically processes. Numerous experiments and method comparison are performed using the proposed PDE-based edge detection framework. [ABSTRACT FROM AUTHOR]

The Gaussian function is essential in the construction of the Frangi and COSFIRE (combination of shifted filter responses) filters. The connection of the broken vessels and an accurate extraction of the vascular structure is the main goal of this study. Thus, the outcome of the Frangi and COSFIRE edge detection algorithms are fused using the Dempster-Shafer algorithm with the aim to improve detection and to enhance the retinal vascular structure. For objective results, the average diameters of the retinal vessels provided by Frangi, COSFIRE and Dempster-Shafer fusion algorithms are measured. These experimental values are compared to the ground truth values provided by manually segmented retinal images. We prove the superiority of the fusion algorithm in terms of image quality by using the figure of merit objective metric that correlates the effects of all post-processing techniques. [ABSTRACT FROM AUTHOR]

Recently. Computer vision is playing an important role in many essential applications, such as medical image analysis, visual surveillance, etc. Many of these applications are subject to a "real-time constraint", therefore it requires a fast and reliable computation system. Edge detection is the approach used most frequently for segmenting images based on changes in intensity, it extracts important structural information needed for high-level functionality and reduces the amount of data that needs to be processed. There are various kernels employed to achieve edge detection, such as Sobel, Robert, and Prewitt, upon which, the most commonly used is Sobel. This paper introduces a novel type of operator cells on the Reconfigurable Static Data-flow Architecture (RSDA), which is a scalable architecture optimized for the computation of image and video. This enhancement shows significant improvement, as it decreases the computational 26%, compared to using the conventional adder cells, and also decreases the LUTs and hardware resources of the architecture. A comparison between the conventional adders and different types of compressors has been exploited based on results from simulation on Isim simulator and a flooring plan using PlanAhead tool. [ABSTRACT FROM AUTHOR]

Convolution serves as the basic computational primitive for various associative computing tasks ranging from edge detection to image matching. CMOS implementation of such computations entails significant bottlenecks in area and energy consumption due to the large number of multiplication and addition operations involved. In this paper, we propose an ultra-low power and compact hybrid spintronic-CMOS design for the convolution computing unit. Low-voltage operation of domain-wall motion based magneto-metallic "Spin-Memristor"s interfaced with CMOS circuits is able to perform the convolution operation with reasonable accuracy. Simulation results of Gabor filtering for edge detection reveal ∼ 2.5× lower energy consumption compared to a baseline 45nm-CMOS implementation. [ABSTRACT FROM AUTHOR]

Historical document restoration has always been a challenging need for information acquisition and conversion to a stable and portable format. Document image binarization, is an important pre-processing phase in document image analysis and processing of such historical documents. It has never been an easy task accounting to the difficulty in choosing a threshold value when the document image suffers from noises due present in the original document due to its degradation, ageing of the paper, other unwanted impurities, background images, bleed-through, and so on. Inconsistency in the darkness of the letters in handwritten documents adds to the difficulty. Though there are many algorithms that have been proposed for document image binarization, there has not been an effective algorithm which can efficiently binarize all kind of documents. In this paper we introduce a novel method to perform document image binarization using bit-plane slicing. As per this approach, the 8 bit planes of the gray image are extracted and processed separately, at the end of which, the results are combined to give the final binarized output. This method works well for all kinds of historical documents except those with both, less noise and low contrast. [ABSTRACT FROM AUTHOR]

Deep learning and deep belief networks (DBNs) are one of the most used topics in machine learning and pattern recognition area in recent years. DBNs consist of stacked Restricted Boltzmann Machine (RBM) structure. RBM is energy based stochastic neural networks. DBNs have many hidden layers and it is optimized after fine-tuning process with Autoencoder (AE) architecture. AE transforms input space to new space. Edge detection is also one of the important issues in machine vision. It is generally done with gradient or Laplacian methods. Some of classical techniques, used in the literature, are canny, differential, sobel, prewitt, roberts or fuzzy logic methods. In this paper, we propose deep learning based edge detection method. Some hidden features is discovered with suitable DBNs architecture. In order to evaluate the performance of presented method we use handwritten character images from MNIST data set. Experimental results show that our approach improves the performance of edge detection process. [ABSTRACT FROM AUTHOR]

Automatic Skin Lesion Diagnosis (ASLD) allows skin lesion diagnosis by using a computer or mobile devices. The idea of using a computer to assist in diagnosis of skin lesions was first proposed in the literature around 1985. Images of skin lesions are analyzed by the computer to capture certain features thought to be characteristic of skin diseases. These features (expressed as numeric values) are then used to classify the image and report a diagnosis. Image segmentation is often a critical step in image analysis and it may use statistical classification, thresholding, edge detection, region detection, or any combination of these techniques. Nevertheless, image segmentation of skin lesion images is yet limited to superficial evaluations which merely display images of the segmentation results and appeal to the reader's intuition for evaluation. There is a consistent lack of quantitative measure, thus, it is difficult to know which segmentation present useful results and in which situations they do so. If segmentation is done well, then, all other stages in image analysis are made simpler. If significant features (that are crucial for diagnosis) are not extracted from images, it will affect the accuracy of the automated diagnosis. This paper explore the existing quantitative measure in image segmentation ranging in the application of pattern recognition for example hand writing, plat number, and colour. Selecting the most suitable segmentation measure is highly important so that as much relevant features can be identified and extracted. [ABSTRACT FROM AUTHOR]

This paper aims to define the best suited combination of color components for edge detection. This is done by a selection of the most relevant components from each color space. First, we compare the performances of the conventional canny edge detector applied to this color spaces. Then, we tend to assess the accuracy of detected edges by using two quantitative methods namely, Fram and Deutsch in addition to robust statistical error measures. Experiments have been conducted on images from BSDS300 database using perceptual and correlated color spaces. We additionally perform these experiments on a progressively contaminated versions of each processed image. Relevant color components are defined according to their detection accuracy and gaussian noise robustness. The obtained results show that Y(luma of YIQ), Y(luminance of YCbCr) and V(brightness of HSV) perceptual color components remains the best candidates for an eventual combination. [ABSTRACT FROM AUTHOR]

One of the problems in genome assembly is overlap graph complexity. Despite the existence of many overlap graph simplification methods, it is still tangled after applying many of them. A novel approach to solve this problem is presented. The proposed algorithm uses several types of information about edges in the overlap graph to simplify it. We define a reliability measure of an edge based on this information. The method has been tested on three bacteria with different characteristics; the results show the applicability and efficiency of the suggested approach. [ABSTRACT FROM AUTHOR]

Optical Proximity Correction (OPC) has been widely adopted for resolution enhancement to achieve nanolithography. However, conventional rule-based and model-based OPCs encounter severe difficulties at advanced technology nodes. Inverse Lithography Technique (ILT) that solves the inverse problem of the imaging system becomes a promising solution for OPC. In this paper, we consider simultaneously 1) the design target optimization under nominal process condition and 2) process window minimization with different process corners, and solve the mask optimization problem based on ILT. The proposed method is tested on 32nm designs released by IBM for the ICCAD 2013 contest. Our optimization is implemented in two modes, MOSAIC fast and MOSAIC exact, which outperform the first place winner of the ICCAD 2013 contest by 7% and 11%, respectively. [ABSTRACT FROM AUTHOR]

Due to the resolution limitations of optical lithography equipment, 1D gridded layout design is gaining steam. Self-aligned double patterning (SADP) is a mature technology for printing 1D layouts. However, for 20nm and beyond, SADP using a single trim mask becomes insufficient for printing all 1D layouts. A viable solution is to complement SADP with e-beam lithography. In this paper, in order to increase the throughput of printing a 1D layout, we consider the problem of e-beam shot count minimization subject to bounded line end extension constraints. Two different approaches of utilizing the trim mask and e-beam to print a layout are considered. The first approach is under the assumption that the trim mask and e-beam are used for end cutting. The second is under the assumption that the trim mask and e-beam are used to rid of all unnecessary portions. We propose elegant ILP formulations for both approaches. Experimental results show that both ILP formulations can be solved efficiently. The pros and cons of the two approaches for manufacturing 1D layout are discussed. [ABSTRACT FROM AUTHOR]

A stacked-die product integrates multiple dies on interposers. In this paper, we first discuss the difficulties of traditional testing mechanism for interposers. To improve production yield, a contactless testing mechanism for pre-bond interposers is proposed. Our testing mechanism attempts to detect a defective interposer from the thermal image after heating the interposer. We propose to extract special features from the thermal image and then use a clustering algorithm to determine whether the interposer is defective. Experimental results show that our testing mechanism can efficiently improve the yield from 70.5% to 96.84%. [ABSTRACT FROM AUTHOR]

Recently, many electronic products are employing a touchscreen as the user interface. This trend puts visually impaired persons into a difficult situation operating such products more and more. To tackle that situation, we have proposed a supporting system to operate a capacitive touchscreen. In the proposed system, the user attaches an image sensor to the middle finger and wears an earphone. The icons displayed on the touchscreen are replaced to the color barcodes. When the user traces the touchscreen by the middle finger with the image sensor, the color barcodes are detected and then the speech guidance is generated via the earphone. By hearing the speech guidance, the user pushes the icons by the forefinger to operate the electronic product. This paper develops a processing flow to detect the color barcode from the blurred image captured by the image sensor close to the surface of the touchscreen. In the experiments, we confirm how to construct the color barcode with several colored bars. Then, we evaluate the accuracy and performance of our processing flow. [ABSTRACT FROM AUTHOR]

In various pattern recognition applications, angle variation is always a main challenging factor for producing reliable recognition. To increase the endurance ability on angle variation, this paper adopts a Hierarchical Temporal Memory (HTM) algorithm which applies temporal information to organize time-sequence change of image features, and constructs invariant features so that the influence of angle variation can be effectively learnt and overcome. The proposed multi-angle HTM-based posture recognition method consists of two main modules of Hand Posture Image Pre-processing (HPIP) and Hand Posture Recognition (HPR). In HPIP, each input image is first processed individually by skin color detection, foreground segmentation and edge detection. Then, the three processed results are further combined linearly to locate a hand posture region. If a forearm exists in the located hand posture region, a forearm segmentation process will be executed to keep only the part of palm. In HPR, the normalized image is forwarded to a HTM model for learning and recognizing of different kinds of hand postures. Experiment results show that when using the same continuous unconstrained hand posture database, the proposed method can achieve an 89.1% high recognition rate for discriminating three kinds of hand postures, which are scissors, stone and paper, and outperforms both Adaboost (78.1%) and SVM (79.9%). [ABSTRACT FROM AUTHOR]

We present a new Spectral Difference Method using Dubiner polynomials on triangles which unifies high order accuracy and simplicity. A corresponding exponential filter is used to avoid oscillations near discontinuous solutions. Furthermore, we focus on the edge detection approach in two dimensions in order to reduce the number of filtered cells in our method. [ABSTRACT FROM AUTHOR]

In this paper, we propose a multiresolution edge fusion method based on Stationary Wavelet Transform (SWT) and Spatial Frequency Measurement (SFM). Our proposed method, Stationary Wavelet Transform (SWT) is firstly applied with the original image to get the edge image information both in level 1 and level 2. Next, both edge images are fused to get a complete edge image using Spatial Frequency Measurement, which is compared with a few simple fusion methods. In addition, we also apply our method to be used for edge detection in noisy original image. From the experiments, we found that our proposed method provides a complete edge image evaluated by the correlation value. [ABSTRACT FROM AUTHOR]